Development of a practical and scalable synthesis of a potent p38 mitogen-activated protein kinase inhibitor

Article abstract of DOI:10.1021/op300237b

Process research and development of a practical and scalable synthetic method toward a potent inhibitor of p38 mitogen-activated protein kinase 1 is described. The medicinal chemistry synthetic method had several issues in scale-up synthesis. In contrast, the synthetic method described here does not require purification by column chromatography for all steps, and the formation of impurities is suppressed well. Aminopyrazole ring formation was achieved by reaction between a new chiral amine building block 7 and bromoketone unit 4 as a key reaction. This highly efficient and scalable process was successfully demonstrated in the large-scale synthesis of 1·HBr.

Full text of DOI:10.1021/op300237b

Article
pubs.acs.org/OPRD
Development of a Practical and Scalable Synthesis of a Potent p38
Mitogen-Activated Protein Kinase Inhibitor
Shinya Yoshida,*,† Yasumasa Hayashi,† Kazuyoshi Obitsu,† Atsushi Nakamura,† Takashi Kikuchi,†
Tatsuya Sawada,† Takenori Kimura,‡ Takumi Takahashi,§ and Takashi Mukuta†
†Process Chemistry Laboratories, Astellas Pharma Inc., 160-2 Akahama, Takahagi-shi, Ibaraki 318-0001, Japan
‡Astellas Research Technologies Company, Ltd., 21, Miyukigaoka, Tsukuba-shi, Ibaraki 305-8585, Japan
§Supply Chain Management, Astellas Pharma Inc., 2-3-11, Nihonbashi-Honcho, Chuo-ku, Tokyo 103-8411, Japan
S
* Supporting Information
ABSTRACT: Process research and development of a practical and scalable synthetic method toward a potent inhibitor of p38
mitogen-activated protein kinase 1 is described. The medicinal chemistry synthetic method had several issues in scale-up
synthesis. In contrast, the synthetic method described here does not require purification by column chromatography for all steps,
and the formation of impurities is suppressed well. Aminopyrazole ring formation was achieved by reaction between a new chiral
amine building block 7 and bromoketone unit 4 as a key reaction. This highly efficient and scalable process was successfully
demonstrated in the large-scale synthesis of 1·HBr.
INTRODUCTION
RESULTS AND DISCUSSION
■
■
p38 mitogen-activated protein (MAP) kinase inhibitors have
been shown to effectively block the production of cytokines such
as interleukin-1 (IL-1), tumor necrosis factor-α (TNF-α), and IL-6.
Inhibitors of p38 MAP kinase therefore have significant therapeutic
potential for the treatment of autoimmune and inflammatory
diseases, such as rheumatoid arthritis (RA), multiple sclerosis,
chronic obstructive pulmonary disease (COPD), psoriasis, and
inflammatory bowel disease (IBD).1 (R)-6-[2-(4-Fluorophenyl)-
6-(hydroxymethyl)-4,5,6,7-tetrahydropyrazolo[1,5-a]pyrimidin-
3-yl]-2-(2-tolyl)-2,3-dihydropyridazin-3-one monohydrobromide
(1·HBr) (Figure 1) has been identified as a potent p38 MAP
The total synthesis of compound 1 on a small scale was recently
reported (Scheme 1).1a Many issues with this synthesis had to be
overcome to achieve successful scale-up, as described below.
The SiO2 column chromatography purification required in
many steps should be avoided.
Final chiral product 1 was separated using chiral column
chromatography, which was accomplished at a 260 g scale separation
level. However, separation on a larger scale would be difficult
because of the difficulty of chiral separation and cost performance.
The use of TBDMS-Cl should be avoided from the point of
view of economics.
The overall yield was low [3.0% from ethyl 2-(hydroxymethyl)-
acrylate (2) in nine steps].
New Synthetic Strategy for 1. Against this background, we
experienced strong demand for the development of a scalable
production process for 1 with a higher overall yield that does not
require column chromatography purification. We describe our
synthetic strategy for compound 1 in Scheme 2. The key reaction
is aminopyrazole ring formation using thiosemicarbazide 8,
which was derived from chiral amine 7 and bromoketone unit 4.
A useful synthesis method for chiral building block 7 has already
been described by our group.2 The synthesis method of
bromoketone unit 4 has already been reported,1a and it was
prepared from 3-chloro-6-methylpyridazine (9) and ethyl
4-fluorobenzoate (10). We describe the development of a practical
and scalable synthesis method for 1 below.
Figure 1. Structure of 1.
kinase inhibitor.1a We were recently required to execute the first
scale-up synthesis of 1. In the medicinal chemistry synthetic route,
the final product 1 was prepared by chiral column chromatography
of the racemate rac-1 (Scheme 1),1a meaning that scale-up synthesis
using this methodology would be difficult. In our new synthetic
route, compound 1 was constructed from the two key intermediates,
thiosemicarbazide unit 8 derived from chiral amine building block
72 and bromoketone unit 4 (Scheme 2). Here, we describe the
development of a practical and scalable synthetic method capable of
being operated on a 17 kg scale for the first GMP delivery of 1·HBr.
Optimization of Bromoketone Unit 4 Synthesis. The
synthesis of bromoketone 4 is shown in Scheme 3. In the
medicinal chemistry method, 4 was prepared from commercially
available 9 and 10 in 33% yield in six steps.1a This method
required a purification via SiO2 column chromatography. We
Received: August 29, 2012
Published: October 10, 2012
© 2012 American Chemical Society
1818
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
a
Scheme 1. Medicinal Chemistry Synthetic Route for 1
a
Reagents and conditions: (a) imidazole, TBDMSCl, CH2Cl2, rt (99% yield); (b) BnNH2, EtOH, 50 °C, SiO2 column chromatography (38% yield);
(c) Pd(OH)2, H2 (3 atm), EtOH (100% yield); (d) O-phenyl chlorothionoformate, saturated aqueous NaHCO3, toluene, rt, SiO2 column
chromatography (63% yield); (e) hydrazine monohydrate, EtOH, 40 °C, SiO2 column chromatography (75% yield); (f) compound 4, AcOH,
EtOH, 60 °C (54% yield); (g) MsCl, Et3N, CH3CN, 100 °C, SiO2 column chromatography (79% yield); (h) LiBH4, THF, SiO2 column
chromatography (82% yield); (i) chiral column chromatography separation (48% yield).
Scheme 2. New Synthetic Strategy for 1
therefore investigated a robust synthesis method that did not
require SiO2 column chromatography purification. First, the
reaction of 3-chloro-6-methylpyridazine 9 and ethyl 4-
fluorobenzoate 10 was tested in the presence of various kinds of
bases (Table 1). All of the bases tested gave complete conversion of
9, but the use of sodium hexamethyldisilazane (NaHMDS),
potassium hexamethyldisilazane (KHMDS), t-BuONa, t-BuOK, or
NaOEt resulted in high levels of unknown impurities.3 The order
of the rate of reaction was roughly as follows: Li > Na > K. The use
of LHMDS gave the best results for this reaction (12, 69.6%
isolated yield).
13 in 91% yield. Subsequent ketal protection of 13 with ethylene
glycol4 gave the desired protected compound 14 in the presence
of a catalytic amount of TsOH·H2O in toluene solvent. Chan−
Evans−Lam coupling with commercially available 2-methylphe-
nylboronic acid5 was accomplished using pyridine and Cu-
(OAc)2 in DMF at 25 °C under an atmosphere of air. The
medicinal chemistry synthetic method required SiO2 column
chromatography at this stage. In contrast, our procedure did not
require this chromatography but rather used optimization of the
post-treatment sequence. Details of the phase separation
procedure are provided in the Experimental Section. After the
deketalization of 15 in the presence of aqueous HCl in MeOH,6
highly pure 16 was obtained in 88% yield. Meanwhile, the direct
Subsequent reaction of the chloropyridazine 12 was
performed in aqueous acetic acid with NaOAc to give pyridazone
1819
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
Scheme 3. Synthesis of Bromoketone Unit 4
a
unit 4 from starting material 9 was 46.0%, which was an
improvement of 12.8% compared with that of the medicinal
chemistry method.
Preparation of Thiosemicarbazide Unit 8 (Scheme 4).
Attempts to improve the impurity profile of the conversion of 7
to 8 are summarized in Table 2.
Thiocarbamate intermediate 18 was synthesized from the
reaction between chiral amine unit 7 and O-phenyl chlorothiono-
formate (17) with Et3N as a base. When the reaction was performed
at 25 or −10 °C, unacceptable levels of cyclic impurity 19 and dimer
20 (Figure 2) were observed (see Table 2, entries 1 and 2). After
Table 1. Base Screening for the Preparation of 12
b
HPLC (%)
c
entry
base
12
impurities
1
2
3
4
5
6
LHMDS
NaHMDS
KHMDS
t-BuONa
t-BuOK
84
75
72
66
62
47
8
17
21
27
30
49
NaOEt
a
Base, 2.1 equiv; reaction temperature, 0 °C; reaction time, 1 h;
solvent, THF. Determined by HPLC method A (see the
Experimental Section). The different bases generated the same
kinds of many impurities, and the total quantity is given.
b
c
a
Table 2. Preparation of Thiosemicarbazide Unit 8
b
HPLC (%)
synthesis of 16 from 13 was attempted without protection.
However, the desired reaction did not occur, and many small
impurities were observed via HPLC. In the α-bromination of 16,
pyridinium tribromide was used as a bromination reagent in the
medicinal chemistry route but carried the risk of an unstable
supply of suitable quality. As an alternative, the reaction was
accomplished using bromine in good yield. Consequently, we
were able to prepare 32.8 kg of bromoketone unit 4 on a large
scale without column chromatography. The yield of bromoketone
temperature
entry
conditions (equiv)
(°C), time (h) 19
20
1
2
3
Et3N (2.5), 17 (1.3)
Et3N (2.5), 17 (1.3)
25, 1
7
2
2
−10 to −5, 1
−10 to −5, 2
5
TMS-Cl (2.0), Et3N (4.5),
then added 17 (1.0)
0.1
not detected
a
b
Conversion, >95%; solvent, CH2Cl2. Determined by HPLC method
C (see the Experimental Section). Impurities 19 and 20 were observed
before the addition of hydrazine hydrate.
Scheme 4. Synthesis of Thiosemicarbazide Unit 8
1820
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
1:1 ratio of thiadiazine 22 and 23 was observed on HPLC analysis.
The subsequent desulfurization (sulfur extrusion) of thiadiazine
22 and 23 was not accelerated under these conditions. From these
results, it was determined that the subsequent desulfurization
reaction was rate-limiting. To improve the sulfur extrusion
reaction, basic conditions were attempted, for example, reaction
in the presence of KOH, NaOEt, NaHCO3, and Et3N (entries 2−5,
respectively).7b−d,i As a result, the sulfur extrusion proceeded
faster that in the absence of base, but new, unidentified impurities
were formed, which resulted in a poor HPLC profile. Alternatively,
acidic conditions were investigated, for example, reaction with an
aqueous solution of HBr, HCl, or HBr with AcOH.1a,7e,f,h
However, these conditions did not provide a good reaction profile,
with a large amount of deprotected impurity 24 observed on HPLC
(entries 6−8). Because the enantiopurity was decreased by the
formation and residual amount of 24, the formation of 24 should be
suppressed during the reaction. To improve the reaction profile, the
reaction was attempted in the presence of a weak acid, including
KH2PO4 (entry 9) and PPTS (entry 10). Results showed that the
addition of aqueous KH2PO4 buffer (pH 4.1), which buffered the
HBr generated during this reaction, worked well and gave a good
reaction profile on HPLC. The suppression of the impurities by
buffering the reaction mixture with aqueous KH2PO4 demonstrates
that the pH is critical during sulfur extrusion. Consequently, the
yield of the reaction was improved from 39% to typically 83%, and
the formation of 24 was suppressed well (<1%). Impurity 24 was
completely purged into the filtrate during precipitation of 11. This
robust procedure was demonstrated on a large scale, with 30.6 kg of
desired compound 11 prepared in 71.0% yield in two steps (see the
Experimental Section).
Figure 2. Structures of impurities.
completion of the conversion of 7, the levels of 19 continued to
increase at 25 °C. The slow addition of 17 gave a higher level of
formation of dimer 20. In contrast, attempts to directly convert
impurity 19 with hydrazine hydrate to 8 failed. As an alternative
method, in situ double protection of amine and alcohol with
TMS-Cl was attempted to prevent formation of these impurities
(entry 3), which resulted in the complete suppression of their
formation. Further, after the removal of O-TMS with citric acid in
MeOH, desired compound 8 was obtained in good yield
(typically 80%). However, this method required inefficient
protection and deprotection that should be avoided. Accord-
ingly, we screened the reaction solvent to optimize this reaction
and thereby obviate the need for this problematic TMS
protection methodology (Table 3).
a
Table 3. Solvent Screening
b
HPLC (%)
entry
solvent
7
19
20
1
2
3
CH3CN
i-PrOAc
EtOH
1.2
3.7
1.2
3.4
5.9
0.2
27
Tetrahydropyrazolopyrimidine Ring Formation. For
formation of the tetrahydropyrazolopyrimidine ring, Et3N and
MsCl were used in the medicinal chemistry method. However,
this produced a complicated reaction because of the replacement
of mesylate with chloride (Scheme 6 and Table 5, entry 1).8
Whereas the conversion of mesylate 25 to 27 was fast, that of
chloride 26 to 27 was very slow and not completed despite
continuous reaction for 48 h at 80 °C. Moreover, this reaction
was not reproducible. Because this issue would likely interfere
with an efficient scale-up, a base screen was performed (Table 5).
The use of K2CO3 or NaHCO3 did not give full conversion
(entries 3 and 4). In contrast, the use of pyridine worked well,
with the formation of intermediate 26 suppressed completely.
The isolated yield of 27 was typically 85%. In a pilot plant-scale
synthesis, 25.0 kg of compound 27 was obtained (85.1% yield).
Preparation of Final Product 1·HBr (Scheme 7). In the
final stage, debenzylation of primary alcohol 27 with hydro-
chloric acid in EtOH at 60 °C was performed, and the desired
free base 1 (22.0 kg) was prepared in 98.6% yield in the first scale-
up synthesis. The subsequent salt formation was performed in
aqueous HBr in n-propanol in good yield. Details of the
procedure are described in the Experimental Section. Using this
newly developed synthetic method, 16.7 kg of final product
1·HBr was synthesized with a purity of 98.9% in a highly practical
and scalable way. The overall yield from 3-chloro-6-methylpyr-
idazine (9) was 19.6%. The enantioisomer was detected at 0.27%
at the release testing of drug substance 1·HBr, with this being the
same level of enantiopurity that was seen for the key intermediate
chiral amine 7.
not detected
not detected
a
21, 1.05 equiv; Et3N, 2.1 equiv; reaction temperature, −5 to 0 °C;
b
reaction time, 0.5−1 h. Determined by HPLC method C (see the
Experimental Section).
As a result of this, EtOH was selected as the optimal solvent,
and the formation of impurities was completely suppressed
because of the low solubility of intermediate 18 in EtOH (entry 3).
The use of CH3CN or i-PrOAc produced unfavorable stirring
conditions because of the Et3N·HCl formed by the reaction
(entries 1 and 2) and gave higher levels of impurities. After
treatment with hydrazine hydrate, thiosemicarbazide compound 8
could be crystallized as a salt of hydrochloric acid. However,
crystallization was unnecessary from the point of view of the
quality of this compound. Thiosemicarbazide compound 8 was
used in the next step without isolation and purification. The yield
of this step was typically 86%.
Aminopyrazole Ring Formation (Scheme 5). In the
medicinal chemistry route, aminopyrazole ring formation was
performed in an AcOH/EtOH solvent at 60 °C (Table 4, entry 1).
However, the reaction profile on HPLC analysis in this case was
characterized by its slowness, with the reaction incomplete at
3 days, at which time the levels of intermediates 22 and 23 remained
at 12% (entry 1). Further, the isolated yield of desired compound
11 was low (typically 39%). To increase this yield, we screened
other conditions. Liquid chromatography−mass spectrum studies
during this reaction indicated a tendency for the very rapid
formation of aminal intermediate 21 with liberation of 1 equiv of
HBr and cyclization, which was completed in EtOH at 25 °C.
Subsequent dehydration was accelerated by the addition of AcOH
or warming of the reaction mixture to around 40 °C, and an almost
CONCLUSION
■
In summary, we developed a practical and scalable synthetic
route for the potent p38 MAP kinase inhibitor 1. Aminopyrazole
1821
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
Scheme 5. Aminopyrazole Ring Formation1a,7
a
Table 4. Aminopyrazole Ring Formation
c
HPLC (%)
intermediate 22 and 23
entry
additive, temperature (°C), time (h)
4 and 8
11
impurity 24
unknown impurities
1
2
AcOH (6 v/w), 60, 72
<1
<1
<1
<1
<1
<1
<1
<1
<1
<1
53
30
29
74
71
41
37
39
90
81
12
1
11
23
68
66
19
21
13
8
b
KOH, 40, 24
not detected
b
3
NaOEt, 40, 24
4
not detected
b
4
NaHCO3, 40, 24
6
not detected
b
5
Et3N, 40, 24
7
not detected
b
6
aqueous HBr, 40, 8
4
41
49
55
<1
3
b
7
HBr/AcOH, 40, 8
5
b
8
aqueous HCl, 40, 8
3
2
b
9
aqueous KH2PO4, 40, 24
2
6
b
10
PPTS, 40, 24
7
8
a
b
c
8, 1 equiv; EtOH, 6 v/w. One equivalent of additive was added after the formation of 22 and 23. Determined by HPLC method C (see the
Experimental Section).
significant increase in comparison to that of the medicinal
chemistry method, which required chiral separation of racemate
using chiral column chromatography in the final stage (3.0%
overall yield).
ring formation was achieved by reaction of a key chiral amine unit
7 that was synthesized from a triol compound using an enantio-
selective lipase reaction2 and bromoketone unit 4. The formation
of impurities 19 and 20 was controlled well at low levels during
preparation of the thiosemicarbazide unit. Finally, this process
was performed on a large scale, and we prepared 16.7 kg of final
product 1·HBr for the first GMP delivery. The overall yield was
19.6% from 3-chloro-6-methylpyridazine (9) and 16.1% from
2-(hydroxymethyl)-1,3-propanediol.2 This yield represented a
EXPERIMENTAL SECTION
General. Starting materials, reagents, and solvents were
obtained from commercial suppliers and used without further
■
1
purification. H and 13C NMR spectra were recorded in the
1822
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
Scheme 6. Tetrahydropyrazolopyrimidine Ring Formation
a
270 nm: 3-chloro-6-methylpyridazine (9), 3.2 min; 12, 5.9 min; 13,
3.2 min; 14, 3.6 min; toluene, 9.4 min; 2-methylphenylboronic acid,
3.8 min; o-cresol, 5.2 min; 15, 8.1 min; 16, 6.5 min.
HPLC Method B. YMC-Pack ODS-A, 5 μm, 4.6 mm × 150 mm
column, elution with 0.01 M aqueous KH2PO4/CH3CN (4/6),
over 20 min, 1.0 mL/min, at 40 °C, with UV detection at 270 nm:
16, 3.7 min; 4, 6.0 min.
HPLC Method C. YMC-Pack ODS-A, 5 μm, 4.6 mm × 150 mm
column, elution with 0.01 M aqueous KH2PO4/CH3CN (1/1),
over 30 min, 1.0 mL/min, at 40 °C, with UV detection at 220 nm:
7, 1.5 min; intermediate 18 (thiocarbamate compound), 7.7 min;
O-phenyl chlorothionoformate (17), 15 min; 8, 2.2 min; phenol,
2.7 min; 11, 5.7 min; 4, 11 min; intermediate 21, 3.5 min; inter-
mediate 22, 7.4 min; intermediate 23, 7.6 min; 27, 17 min;
pyridine, 2.2 min; 1, 2.7 min.
Table 5. Base Screening
b
HPLC (%)
entry
base
Et3N
11
27
intermediate 26
c
1
not detected
10
93
75
40
81
2
3
4
pyridine
K2CO3
<0.1
20
not detected
<0.1
NaHCO3
45
not detected
a
MsCl, 2.2 equiv; base, 5 equiv; CH3CN, 10 v/w; reaction
temperature, 60 °C; reaction time, 5−15 h. Determined by HPLC
method C (see the Experimental Section). The reaction was not
b
c
reproducible.
Scheme 7. Preparation of Final Product 1·HBr
HPLC Method D. Waters Xterra RP18, 5 μm, 4.6 mm × 150
mm column, elution A; 0.01 M aqueous KH2PO4 (adjusted to
pH 3.0 with aqueous H3PO4), elution B; CH3CN, over 70 min,
1.0 mL/min, at 25 °C, with UV detection at 220 nm.
The gradient program was as follows (1, 12 min):
time (min)
elution A
elution B
0 to 20
70
30
20 to 60
60 to 70
70 to 20
20
30 to 80
80
specified deuterated solvent. Chemical shifts of 1H NMR spectra
are reported in parts per million on the δ scale from an internal
standard of residual solvent (DMSO-d6, 2.50 ppm) or TMS. Data
are reported as follows: chemical shift, multiplicity (s, singlet; d,
doublet; dd, doublet doublet; t, triplet; m, multiplet; br, broad),
coupling constant (hertz), and integration. Chemical shifts of
proton-decoupled 13C NMR spectra are reported in parts per
million from the central peak of DMSO-d6 (39.5 ppm) on the
δ scale. Mass spectra were measured using UPLC/SQD-LC/
MS, JEOL JMS-GC mate II, Waters ZQ 2000, and JEOL JMS-
LX2000 instruments. IR spectra were measured using the KBr
disk method (JP16). KF were measured using the JP method.
HPLC was performed using a Hitachi D-2500 or D-7500 system.
HPLC methods are described below.
HPLC Method E. DAICEL CHIRALCEL OJ-H, 5 μm, 4.6 mm ×
250 mm column, elution with n-hexane/MeOH/i-PrOH (5/4/3),
over 60 min, 0.5 mL/min, at 40 °C, with UV detection at 228 nm: 1,
14 min; enantiomer of 1, 18 min.
2-(6-Chloropyridazin-3-yl)-1-(4-fluorophenyl)ethan-1-
one (12). A solution of 3-chloro-6-methylpyridazine 9 (23.5 kg,
183 mol) and ethyl 4-fluorobenzoate 10 (32.3 kg, 192 mol) in
THF (47 L) was cooled to −8 °C. To the batch was added LHMDS
(28% in THF, 229 kg, 384 mol) at −8 to 6 °C and the mixture aged
for 1 h; then HPLC analysis indicated <2% 9 remained (HPLC
method A). The batch was then poured into an aqueous solution of
HCl [35 wt % HCl (83.9 kg) and water (353 L)] at 0−30 °C and
washed with THF (23.5 L). The resulting slurry was aged for 16 h at
20 °C, filtered, and washed with i-PrOH (47 L) and then water
(70.5 L). Loss to the filtrate was 13%. The wet cake was dried in
HPLC Method A. Waters XTERRA RP18, 5 μm, 4.6 mm × 250
mm column, elution with 0.05 M aqueous K2HPO4/CH3CN
(1/1), over 20 min, 1.0 mL/min, at 40 °C, with UV detection at
1823
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
vacuo at 50 °C to afford the desired 12 at 98.5% purity via HPLC
method A (31.9 kg, 69.6% yield): MS (ESI, positive mode) m/z
251.0; MS (ESI, negative mode) m/z 249.0; 1H NMR (400 MHz,
DMSO-d6) δ 8.15−8.20 (2H, m), 7.93 (1H, d, J = 8.8 Hz), 7.79
(1H, d, J = 8.8 Hz), 7.38−7.45 (2H, m), 4.85 (2H, s); 13C NMR
(100 MHz, DMSO-d6) δ 194.8, 166.5, 164.0, 158.3, 155.2, 132.8
(JCF = 3.0 Hz), 131.6 (JCF = 44.8 Hz), 131.1 (JCF = 40.2 Hz), 129.3
(JCF = 161 Hz), 128.2 (JCF = 8.9 Hz), 115.8 (JCF = 21.8 Hz), 44.6.
Anal. Calcd for C12H8ClFN2O: C (57.50%), H (3.22%), N
(11.18%), F (7.58%). Found: C (57.33%), H (3.11%), N (11.32%),
F (7.77%). Water: 0.3% (KF).
6-[2-(4-Fluorophenyl)-2-oxoethyl]-2,3-dihydropyridazin-
3-one (13). A solution of 12 (31.7 kg, 127 mol), AcOH
(133 kg), NaOAc (10.9 kg, 133 mol), and water (31.7 kg) was
heated to 100−108 °C, and the batch was aged for 5 h; then
HPLC analysis indicated <1% 12 remained (HPLC method A).
Water (256.3 L) was then added to the batch at 38−80 °C; the
resulting slurry was heated to 82 °C, and the resulting solution
was cooled to 0−5°C and aged for 19 h. The batch was then filtered,
and the residue was washed with water (127 L). Loss to the filtrate
was 6%. The wet cake was dried in vacuo at 50 °C to afford the
desired 13 at 97.0% purity via HPLC method A (26.6 kg, 90.6%
yield): MS (ESI, positive mode) m/z 233.1; MS (ESI, negative
mode) m/z 231.0; 1H NMR (400 MHz, DMSO-d6) δ 12.9 (1H, s),
8.09−8.15 (2H, m), 7.36−7.44 (3H, m), 6.87 (1H, d, J = 9.7 Hz),
4.43 (2H, s); 13C NMR (100 MHz, DMSO-d6) δ 195.2, 166.5,
163.9, 160.3, 142.7, 135.3, 132.8 (JCF = 3.0 Hz), 131.2 (JCF = 9.7
Hz), 129.1, 115.9, 115.7, 43.7. Anal. Calcd for C12H9FN2O2: C
(62.07%), H (3.91%), N (12.06%), F (8.18%). Found: C (62.21%),
H (3.99%), N (11.92%), F (8.33%). Water: 0.1% (KF).
(Radiolite, 14.1 kg). The suspended substance was removed during
filtration. After filtration of the batch, the resulting cake was washed
with isopropyl acetate (56 L). After phase separation, the organic
layer was washed twice with an aqueous solution of NaOH [NaOH
(11.2 kg) and water (281 L)]. o-Cresol derived from 2-methyl-
phenylboronic acid was purged into aqueous layers. The resulting
organic layer was washed with an aqueous solution of HCl [35 wt %
HCl (28.1 kg) and water (259 L)] and concentrated in vacuo using
a vacuum pump to afford the desired 15 at 92.4% purity via HPLC
method A, which was used in the next step without isolation or puri-
fication: MS (ESI, positive mode) m/z 367.2; 1H NMR (400 MHz,
DMSO-d6) δ 7.46 (1H, d, J = 9.7 Hz), 7.31−7.40 (4H, m), 7.22−
7.30 (1H, m), 7.12−7.20 (2H, m), 7.04 (1H, d, J = 7.5 Hz), 6.98
(1H, d, J = 9.7 Hz), 3.95−4.02 (2H, m), 3.73−3.77 (2H, m), 3.17
(2H, s), 1.84 (3H, s); 13C NMR (100 MHz, DMSO-d6) δ 163.1,
160.7, 158.5, 143.2, 140.7, 137.6, 137.5, 135.0, 134.3, 130.5, 129.3,
128.7, 127.7, 127.6, 127.1, 126.5, 114.8 (JCF = 21.6 Hz), 108.4, 64.5,
44.4, 16.6.
6-[2-(4-Fluorophenyl)-2-oxoethyl]-2-(2-tolyl)-2,3-dihy-
dropyridazin-3-one (16). To the residue of 15 was added
MeOH (142 L). To the resulting solution was added 35 wt % HCl
(21.2 kg, 204 mol), and the mixture was aged for 8 h at 25 °C; then
HPLC analysis indicated <2% 15 remained (HPLC method A).
The batch was then seeded (2.8 g) at 25 °C, and water (141.5 L)
was added at 25 °C. The resulting slurry was cooled to 2 °C and
aged for 13 h. The batch was then filtered and washed with a
MeOH (28.1 L)/water (28.1 L) solution. Loss to the filtrate was
1%. The wet cake was dried in vacuo at 50 °C to afford the desired
16 at 98.5% purity via HPLC method A (28.9 kg, 88.1% yield from
14 in two steps): MS (ESI, positive mode) m/z 323.1; MS (ESI,
1
6-{[2-(4-Fluorophenyl)-1,3-dioxolan-2-yl]methyl}-2,3-
dihydropyridazin-3-one (14). A batch of 13 (26.4 kg, 114 mol),
p-toluenesulfonic acid monohydrate (2.16 kg, 11.4 mol), and
ethylene glycol (35.3 kg, 569 mol) in toluene (264 L) was heated to
105−111 °C and aged for 12 h; then HPLC analysis indicated <7%
13 remained (HPLC method A). During the reaction, the resulting
water was removed at atmospheric pressure by azeotropic distillat-
ion. The batch was then cooled to 20 °C and concentrated in vacuo.
To the residue was added acetonitrile (264 L), and the mixture was
concentrated in vacuo. To the residue were then added acetonitrile
(132 L) and an aqueous solution of NaOH [NaOH (0.64 kg) and
water (15.1 L)] and water (264 L) at <20 °C. The resulting slurry
was cooled to 0−30 °C and aged for 16 h. The batch was then
filtered and washed with water (52.8 L). Loss to the filtrate was 7%.
The wet cake was dried in vacuo at 50 °C to afford the desired 14 at
98.5% purity via HPLC method A (28.3 kg, 90.1% yield): MS (GC,
positive mode) m/z 277.1; MS (ESI, negative mode) m/z 275.2; 1H
NMR (400 MHz, DMSO-d6) δ 12.74 (1H, s), 7.35−7.42 (2H, m),
7.28 (1H, d, J = 9.7 Hz), 7.12−7.20 (2H, m), 6.77 (1H, d, J = 9.7
Hz), 3.88−3.95 (2H, m), 3.68−3.75 (2H, m), 3.11 (2H, s); 13C
NMR (100 MHz, DMSO-d6) δ 163.0, 160.6, 160.2, 142.7, 137.8
(JCF = 3.0 Hz), 135.1, 128.6, 127.7, 127.6, 114.9, 114.7, 108.3, 64.4,
44.3. Anal. Calcd for C14H13FN2O3: C (60.87%), H (4.74%), N
(10.14%), F (6.88%). Found: C (60.61%), H (4.99%), N (10.01%),
F (6.77%). Water: 0.2% (KF).
negative mode) m/z 321.1; H NMR (400 MHz, DMSO-d6) δ
8.08−8.13 (2H, m), 7.52 (1H, d, J = 9.7 Hz), 7.35−7.42 (4H, m),
7.28−7.34 (1H, m), 7.24 (1H, d, J = 7.4 Hz), 7.08 (1H, d, J = 9.7
Hz), 4.51 (2H, s), 2.02 (3H, s); 13C NMR (100 MHz, DMSO-d6)
δ 195.0, 166.5, 163.9, 158.6, 143.3, 140.8, 135.1, 134.3, 132.7 (JCF
=
3.0 Hz), 131.2 (JCF = 9.7 Hz), 130.5, 129.8, 128.7, 127.2, 126.6,
115.9, 115.7, 43.7, 16.8. Anal. Calcd for C19H15FN2O2: C (70.80%),
H (4.69%), N (8.69%), F (5.89%). Found: C (70.51%), H (4.79%),
N (8.61%), F (5.69%). Water: 0.1% (KF).
(RS)-6-[1-Bromo-2-(4-fluorophenyl)-2-oxoethyl]-2-(2-
tolyl)-2,3-dihydropyridazin-3-one (4). To a solution of 16
(28.8 kg, 89.4 mol) in AcOH (181 kg) was added Br2 (15.04 kg,
94.1 mol) at 25 °C, and the mixture was aged for 4 h at the same
temperature; then HPLC analysis indicated <1% 16 remained
(HPLC method B). To the batch were added toluene (346 L)
and water (173 L). The resulting organic layer was concentrated in
vacuo. To the residue was added isopropyl acetate (173 L) while the
mixture was heated to 87 °C. The resulting solution was cooled to
30 °C, and diisopropyl ether9(346 L) was added at 30 °C. The slurry
was cooled to 2 °C and aged for 12 h. The batch was then filtered
and washed with diisopropyl ether (28.8 L, precooled to 2 °C). Loss
to the filtrate was 4%. The wet cake was dried in vacuo at 50 °C to
afford the desired 4 at 99.2% purity via HPLC method B (32.8 kg,
91.5% yield): MS (ESI, positive mode) m/z 401.0 (d, 403.0); MS
(ESI, negative mode) m/z 399.0 (d, 401.1); 1H NMR (400 MHz,
DMSO-d6) δ 8.12−8.17 (2H, m), 7.80 (1H, d, J = 9.7 Hz), 7.34−
7.42 (4H, m), 7.29−7.33 (1H, m), 7.19−7.23 (2H, m), 7.08 (1H, s),
1.87 (3H, s); 13C NMR (100 MHz, DMSO-d6) δ 196.0, 166.3,
163.8, 158.6, 146.6, 140.5, 134.3, 132.0, 131.9, 131.3, 131.0, 130.6,
128.9, 127.2, 126.6, 115.8, 115.6, 74.1, 16.6. Anal. Calcd for
C19H14BrFN2O2: C (56.88%), H (3.52%), N (6.98%), F (4.74%),
Br (19.91%). Found: C (56.71%), H (3.71%), N (6.69%), F (4.69%),
Br (19.82%). Water: 0.1% (KF).
6-{[2-(4-Fluorophenyl)-1,3-dioxolan-2-yl]methyl}-2-(2-
tolyl)-2,3-dihydropyridazin-3-one (15). To the mixture of
2-methylphenylboronic acid (30.4 kg, 224 mol), 14 (28.1 kg, 102 mol),
and pyridine (32.2 kg, 407 mol) in DMF (169 L) was added
anhydrous Cu(OAc)2 (2.77 kg, 15.3 mol), and the batch was stirred
for 37 h at 25 °C under an atmosphere of air; then HPLC analysis
indicated <3% 14 remained (HPLC method A). To the batch were
then added isopropyl acetate (225 L), water (281 L), and filter aid
1824
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
(R)-4-[3-(Benzyloxy)-2-(hydroxymethyl)propyl]-
thiosemicarbazide (8). To a solution of 7 (18.5 kg, 79.8 mol)
in EtOH (146 kg) was added triethylamine (17.0 kg, 168 mol) at
25 °C. The solution was cooled to −5 °C, and O-phenyl chloro-
thionoformate (14.5 kg, 84.0 mol) was added at −5 to 1.5 °C, and
then the mixture was aged for 1 h; then HPLC analysis indicated
<1% 7 remained (HPLC method C). To the batch was then added
hydrazine monohydrate (4.40 kg, 87.9 mol) at −3 °C, and the
mixture was heated to 40 °C and aged for 16 h; then HPLC
analysis indicated <1% intermediate 18 remained (thiocarbamate
compound, HPLC method C). The reaction mixture was used in
the next step without isolation.
(R)-6-{6-[(Benzyloxy)methyl]-2-(4-fluorophenyl)-
4,5,6,7-tetrahydropyrazolo[1,5-a]pyrimidin-3-yl}-2-(2-
tolyl)-2,3-dihydropyridazin-3-one (27). To a solution of 11
(30.4 kg, 56.3 mol) in acetonitrile (238.3 kg) were added
pyridine (22.3 kg, 282 mol) and methanesulfonyl chloride (14.2
kg, 124 mol) at 22−25 °C, and then the mixture was heated to
60 °C. The batch was aged for 5 h at 60 °C; then HPLC analysis
indicated <1% 11 remained (HPLC method C). Water (152.8
kg) was then added at 60 °C, and the solution was cooled to
40 °C and aged for 1 h. Water (153.7 kg) was then added at 40 °C,
and the resulting slurry was cooled to 20 °C and aged for 12 h. The
slurry was filtered and washed twice with water (30.4 kg). The wet
cake was dried in vacuo at 50 °C to afford the desired 27 at 98.2%
purity via HPLC method C (25.0 kg, 85.1% yield): MS (ESI,
An analytical sample of 8 was purified by crystallization as the
HCl salt from i-PrOH (10 v/w) and diisopropyl ether (20 v/w) in
80% yield: MS (FAB, positive mode) m/z 270.1; MS (FAB,
1
positive mode) m/z 522.3; H NMR (400 MHz, DMSO-d6) δ
1
7.46−7.52 (2H, m), 7.31−7.38 (8H, m), 7.20−7.30 (3H, m), 7.08
(1H, d, J = 9.8 Hz), 6.92 (1H, d, J = 9.8 Hz), 6.00 (1H, br), 4.50
(2H, s), 4.16 (1H, dd, J = 12.3, 5.1 Hz), 3.86 (1H, dd, J = 12.3, 8.0
Hz), 3.50 (2H, d, J = 6.7 Hz), 3.34−3.39 (1H, m), 3.08 (1H, t, J =
9.1 Hz), 2.40−2.45 (1H, m), 2.09 (3H, s); 13C NMR (100 MHz,
DMSO-d6) δ 163.0, 160.6, 157.8, 146.5, 144.6, 141.4, 141.1, 138.3,
134.5, 133.1, 130.4, 130.3, 130.2, 130.1, 130.0, 129.8, 128.6, 128.2,
127.4, 127.3, 127.2, 126.6, 115.4, 115.2, 94.0, 72.1, 69.1, 46.9, 40.9,
32.1, 17.1. Anal. Calcd for C31H28FN5O2: C (71.38%), H (5.41%),
N (13.43%), F (3.64%). Found: C (71.13%), H (5.68%), N
(13.20%), F (3.46%). Water: 0.1% (KF).
negative mode) m/z 268.1; H NMR (400 MHz, DMSO-d6) δ
10.01 (2H, br), 8.59 (1H, br), 7.20−7.37 (5H, m), 4.47 (2H, s), 4.02
(1H, br), 3.41−3.55 (7H, m), 2.01−2.04 (1H, m). Anal. Calcd for
C12H19N3O2S·HCl: C (47.13%), H (6.59%), N (13.74%), S
(10.48%), Cl (11.59%). Found: C (47.13%), H (6.58%), N
(13.72%), S (10.41%), Cl (11.46%).
(R)-6-(5-{[3-(Benzyloxy)-2-(hydroxymethyl)propyl]-
amino}-3-(4-fluorophenyl)-1H-pyrazol-4-yl]-2-(2-tolyl)-
2,3-dihydropyridazin-3-one (11). The reaction mixture of
the previous step was cooled to 28 °C, and then 4 (32.0 kg, 79.8
mol) and EtOH (15.3 kg) were added. The batch was heated to
40 °C, and to the resulting solution was added an aqueous solution
of KH2PO4 [KH2PO4 (10.9 kg, 80.1 mol) and water (55.5 kg)] at
36−42 °C. The reaction mixture was aged for 22 h; then HPLC
analysis indicated <1% intermediate 21 and <2% intermediate 22
and 23 remained (HPLC method C). During the reaction, the pH
was around 4. The reaction mixture was then cooled to 30 °C and
filtered, and the residue was washed with EtOH (28.6 kg). The
filtrate was concentrated in vacuo to 110 L. To the residue were
added MEK (150 kg) and water (187.0 kg), and the resulting
aqueous layer was re-extracted with MEK (74.5 kg). The collected
organic layer was washed with an aqueous solution of NaOH
[NaOH (7.4 kg) and water (185 kg)] and then an aqueous
solution of NaCl [NaCl (9.25 kg) and water (185 kg)]. To the
organic layer was added toluene (81.7 kg), and the mixture was
concentrated in vacuo to 70 L. To the residue was added toluene
(81.9 kg), and the mixture was concentrated in vacuo to 70 L. To
the resulting residue were added isopropyl acetate (33.4 kg) and
toluene (162 kg), and the mixture was heated to 100 °C. The
resulting solution was cooled to 65 °C and seeded (1.85 g) at
65 °C. The batch was cooled to −4 °C and aged for 41 h. The
slurry was filtered and washed with isopropyl acetate (16.1 kg,
precooled to 10 °C). The wet cake was dried in vacuo at 50 °C to a
constant weight to afford the desired 11 at 96.8% purity via HPLC
method C (30.6 kg, 71.0% yield from 7): MS (ESI, positive mode)
m/z 540.3; MS (ESI, negative mode) m/z 538.2; 1H NMR (400
MHz, DMSO-d6) δ 12.3 (1H, br), 7.50−7.55 (2H, m), 7.28−7.40
(8H, m), 7.22−7.27 (3H, m), 7.05 (1H, d, J = 9.8 Hz), 6.92 (1H, d,
J = 9.8 Hz), 5.50 (1H, br), 4.51 (1H, br), 4.26 (2H, s), 3.27−3.41
(4H, m), 3.18−3.26 (2H, m), 2.09 (3H, s), 1.87−1.93 (1H, m);
13C NMR (100 MHz, DMSO-d6) δ 157.7, 141.2, 138.5, 134.3,
132.5, 132.4, 130.9, 130.7, 130.6, 130.5, 130.1, 130.0, 128.7, 128.2,
128.1, 127.5, 127.3, 127.2, 127.1, 127.0, 126.7, 115.9 (two
carbons), 115.8, 115.7, 72.1, 59.8, 42.2, 42.1, 41.4, 17.0. Anal. Calcd
for C31H30FN5O3: C (69.00%), H (5.60%), N (12.98%), F
(3.52%). Found: C (68.83%), H (5.78%), N (13.00%), F (3.46%).
Water: 0.1% (KF).
(R)-6-[2-(4-Fluorophenyl)-6-(hydroxymethyl)-4,5,6,7-
tetrahydropyrazolo[1,5-a]pyrimidin-3-yl]-2-(2-tolyl)-2,3-
dihydropyridazin-3-one (1). A mixture of 27 (24.8 kg, 47.5
mol) in EtOH (118.8 kg) was heated to 45 °C, and 35 wt % HCl
(175.3 kg, 1682 mol) was added while the mixture was heated to
60 °C. The batch was aged for 1 day; then HPLC analysis
indicated <1% 27 remained (HPLC method C). The batch was
then cooled to 30 °C, and toluene (129.3 kg) was added. To the
resulting aqueous layer was added an aqueous solution of 5 M
NaOH (265.5 kg), and the pH was adjusted to 7.48. The
resulting slurry was aged at 25 °C for 13 h. The slurry was filtered
and washed with an EtOH (14.7 kg)/water (55.8 kg) solution
and twice with water (148.8 kg). The wet cake was dried in vacuo
at 50 °C to afford the desired 1a at 95.6% purity via HPLC
method C (20.22 kg, 98.6% yield): MS (ESI, positive mode) m/z
432.2; MS (FAB, positive mode) m/z 432.1; MS (FAB, negative
mode) m/z 430.1; 1H NMR (400 MHz, DMSO-d6) δ 7.47−7.51
(2H, m), 7.31−7.38 (4H, m), 7.20−7.25 (2H, m), 7.09 (1H, d,
J = 9.8 Hz), 6.92 (1H, d, J = 9.8 Hz), 5.97 (1H, br), 4.83 (1H, t,
J = 5.3 Hz), 4.12 (1H, dd, J = 12.3, 5.2 Hz), 3.82 (1H, dd, J = 12.3,
8.2 Hz), 3.46 (2H, t, J = 6.0 Hz), 3.30−3.38 (1H, m), 3.01−3.07
(1H, m), 2.15−2.30 (1H, m), 2.09 (3H, s); 13C NMR (100 MHz,
DMSO-d6) δ 163.0, 160.6, 157.8, 146.5, 144.7, 141.5, 141.1,
134.5, 133.1, 130.4, 130.2, 130.1, 129.8, 128.6, 127.4, 126.6,
115.4, 115.2, 93.9, 60.6, 46.9, 40.8, 34.5, 17.1. Anal. Calcd for
C24H22FN5O2: C (66.81%), H (5.14%), N (16.23%), F (4.40%).
Found: C (66.74%), H (5.17%), N (16.27%), F (4.45%). Water:
0.2% (KF).
(R)-6-[2-(4-Fluorophenyl)-6-(hydroxymethyl)-4,5,6,7-
tetrahydropyrazolo[1,5-a]pyrimidin-3-yl]-2-(2-tolyl)-2,3-
dihydropyridazin-3-one Monohydrobromide (1·HBr). To
the mixture of 1 (20.0 kg, 46.4 mol) in n-propanol (161.8 kg) was
added 47 wt % aqueous HBr (12.0 kg, 69.5 mol), and the mixture
was heated to 82 °C. The solution was then cooled to 50 °C and
concentrated in vacuo to 40 L. To the residue was added
n-propanol (79.0 kg), and the mixture was heated to 92 °C and
then cooled to 30 °C. The batch was then transferred to another
1825
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826

Organic Process Research & Development
Article
vessel, and the vessel was washed with n-propanol (82.3 kg) and
heated to 93 °C. The solution was filtered via a 1 μm filter into
another vessel, and the vessel was washed with n-propanol (16.2
kg). The filtrate was heated while being stirred to 91 °C, cooled
to 64 °C, and seeded (2.0 g). The batch was cooled to 20 °C,
aged for 6.5 h, heated to 60 °C, and aged for 1 h. The slurry was
then cooled to 0 °C and aged for 24 h. The slurry was then
filtered and washed with n-propanol (32 kg, precooled to 5 °C).
The wet cake was dried in vacuo at 50 °C to afford the crude
product 1 (18.5 kg, 77.9% yield). A batch of crude 1 (18.1 kg,
35.3 mol) in n-propanol (204 kg) was heated to 91 °C, and the
solution was then filtered via a 1 μm filter into another vessel and
washed with n-propanol (14.6 kg). The batch was heated to
90 °C, and the solution was cooled to 66 °C. To the batch were
then added 47 wt % aqueous HBr (2.98 kg, 17.7 mol, prefiltered
via a 1 μm filter) at 66 °C and n-propanol (7.28 kg, prefiltered via
a 1 μm filter), and the mixture was seeded (18.1 g) at 65 °C. The
batch was aged for 24 h at 65 °C, cooled to 25 °C (rate of 10 °C/h),
and aged for 1 h. To the slurry were then added 47 wt % aqueous
HBr (2.98 kg, 17.7 mol, prefiltered via a 1 μm filter) and n-propanol
(7.28 kg, prefiltered via a 1 μm filter) at 25 °C, and the mixture was
aged for 1 h. The slurry was then cooled to 3 °C, stirred for 13 h,
filtered, and washed with n-propanol (29.1 kg precooled to 5 °C and
prefiltered via a 1 μm filter). The wet cake was dried in vacuo at
50 °C to afford the desired 1·HBr at 99.1% purity via HPLC
method D (16.7 kg, 71.7% yield from 1). The optical isomer was
observed at 0.27% by chiral HPLC analysis (HPLC method E).
The overall yield from 3-chloro-6-methylpyridazine 9 was
19.6%, and 16.1% from 2-(hydroxymethyl)-1,3-propanediol:2
REFERENCES
■
(1) (a) Asano, T.; Yamazaki, H.; Kasahara, C.; Kubota, H.; Kontani, T.;
Harayama, Y.; Ohno, K.; Mizuhara, H.; Yokomoto, M.; Misumi, K.;
Kinoshita, T.; Ohta, M.; Takeuchi, M. J. Med. Chem. 2012, 55, 7772.
(b) Boutros, T.; Chevet, E.; Metrakos, P. Pharmacol. Rev. 2008, 60, 261.
(c) Milburn, R. R.; Thiel, O. R.; Achmatowicz, M.; Wang, X.; Zigterman,
J.; Bernard, C.; Colyer, J. T.; DiVirgilio, E.; Crockett, R.; Correll, T. L.;
Nagapudi, K.; Ranganathan, K.; Hedley, S. J.; Allgeier, A.; Larsen., R. D.
Org. Process Res. Dev. 2011, 15, 31.
(2) Yoshida, S.; Obitsu, K.; Hayashi, Y.; Shibazaki, M.; Kimura, T.;
Takahashi, T.; Asano, T.; Kubota, H.; Mukuta., T. Org. Process Res. Dev.
2012, 16, 1527.
(3) To minimize the formation of these impurities, we performed the
reaction below −20 °C. However, it was difficult to suppress the
formation of these impurities, and it was difficult to identify impurities
because there were too many small impurities in this reaction.
(4) Cloonan, S. M.; Keating, J. J.; Butler, S. G.; Knox, A. J. S.; Jorgensen,
A. M.; Peters, G. H.; Rai, D.; Corrigan, D.; Lloyd, D. G.; Williams, D. C.;
Meegan, M. J. Eur. J. Med. Chem. 2009, 44, 4862.
(5) Ley, S. V.; Thomas, A. W. Angew. Chem., Int. Ed. 2003, 42, 5400.
(6) Albright, J. D.; Lieberman, D. F. J. Heterocycl. Chem. 1994, 31, 537.
(7) (a) Jira, T.; Pfeiffer, W. D.; Lachmann, K.; Epperlein, U. Pharmazie
1994, 49, 401. (b) Lee, L. F.; Howe, R. K. J. Org. Chem. 1984, 49, 4780.
(c) Schmidt, R. R.; Huth, H. Tetrahedron Lett. 1975, 33. (d) Ogura, H.;
Takahashi, H.; Kudo, E. J. Carbohydr., Nucleosides, Nucleotides 1978, 5,
329. (e) Beyer, H.; Wolter, G. Chem. Ber. 1956, 89, 1652. (f) Beyer, H.;
Wolter, G. Chem. Ber. 1956, 89, 2550. (g) Naka, T.; Furukawa, Y.
Heterocycles 1978, 9, 101. (h) Brandt, T. A.; Caron, S.; Damon, D. B.;
DiBrino, J.; Ghosh, A.; Griffith, D. A.; Kedia, S.; Ragan, J. A.; Rose, P. R.;
Vanderplas, B. C.; Wei, L. Tetrahedron 2009, 65, 3292. (i) Kaji, K.;
Nagashima, H.; Ohta, Y.; Nagao, S.; Hirose, Y.; Oda, H. Heterocycles
1984, 22, 479.
1
MS (ESI, positive mode) m/z 432.2; H NMR (400 MHz,
(8) Evans, M. E.; Long, L., Jr.; Parrish, F. W. J. Org. Chem. 1968, 33,
DMSO-d6) δ 7.50−7.55 (2H, m), 7.31−7.38 (4H, m), 7.26−7.30
(2H, m), 7.14 (1H, d, J = 9.8 Hz), 6.97 (1H, d, J = 9.8 Hz), 5.50
(1H, br), 4.17 (1H, dd, J = 12.2, 5.2 Hz), 3.88 (1H, dd, J = 12.2,
8.2 Hz), 3.45 (2H, d, J = 6.6 Hz), 3.37 (1H, dd, J = 12.2, 3.4 Hz),
3.07 (1H, dd, J = 12.2, 8.6 Hz), 2.20−2.32 (1H, m), 2.08 (3H, s);
13C NMR (100 MHz, DMSO-d6) δ 163.5, 161.1, 157.9, 145.7,
145.2, 141.0, 140.4, 134.5, 133.2, 130.6, 130.5, 130.4, 130.1, 128.7,
127.4, 126.7, 115.7, 115.5, 95.0, 60.4, 46.6, 40.6, 34.0, 17.2; IR (KBr,
cm−1) 3301, 2873, 1654, 1624, 1490, 1243, 1171, 846; mp 220.6 °C
(by DSC); heavy metals <20 ppm. Anal. Calcd for
C24H22FN5O2·HBr: C (56.26%), H (4.52%), N (13.67%), Br
(15.59%), F (3.71%). Found: C (56.27%), H (4.57%), N (13.67%),
Br (15.65%), F (3.72%). Water: 0.09% (KF).
1074.
(9) We also attempted to crystallize compound 4 not only with
diisopropyl ether but also with the other ethers, CPME, 2-Me THF, and
MTBE. However, the use of CPME, 2-Me THF, and MTBE confers
higher solubility than diisopropyl ether, making it difficult to prevent
large loss to the filtrate or to crystallization, in spite of combination with
n-heptane.
ASSOCIATED CONTENT
* Supporting Information
■
S
XRD chart of 1·HBr and DSC and TG curves of 1·HBr. This
material is available free of charge via the Internet at http://pubs.
acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: shinya.yoshida@astellas.com. Telephone: +81-293-23-5478.
Fax: +81-293-24-2708.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We are grateful to Mr. H. Yamazaki, Dr. C. Kasahara, and Dr.
Toru Asano for their helpful discussions about the medicinal
synthetic routes to 1. We also thank Ms. K. Fujita and Mr. K. Itoh
for analytical study.
1826
dx.doi.org/10.1021/op300237b | Org. Process Res. Dev. 2012, 16, 1818−1826