Chemistry development of a convergent route to trecetilide hemi-fumarate

Article abstract of DOI:10.1021/op049799f

A novel, efficient, stereoselective synthetic route for N-(4-{4-[ethyl(6-fluoro-6-methylheptyl)amino]-1-(S)-hydroxybutyl}-phenyl) methanesulfonamide hemi-fumaric acid salt (trecetilide hemi-fumarate, Figure 1) has been developed. The process features a convergent approach, which assembles two key intermediates in the last step to form the final molecule, which is then isolated by pH-controlled extraction. The new route offers significant yield and purity advantages over the previous route. However, the solvent volume and cycle time were not fully optimized due to the termination of the project.

Full text of DOI:10.1021/op049799f

Organic Process Research & Development 2005, 9, 174−178
Chemistry Development of a Convergent Route to Trecetilide Hemi-Fumarate
Sonja S. Mackey,‡ Haifeng Wu,*,† Michael E. Matison,† and Michael Goble§
Chemical Research & DeVelopment, Pfizer Corporation, 7000 Portage Road, Kalamazoo, Michigan 49001, U.S.A.
Abstract:
A novel, efficient, stereoselective synthetic route for N-(4-{4-
[ethyl(6-fluoro-6-methylheptyl)amino]-1-(S)-hydroxybutyl}-
phenyl)methanesulfonamide hemi-fumaric acid salt (trecetilide
hemi-fumarate, Figure 1) has been developed. The process
Figure 1. Trecetilide hemi-fumarate.
features a convergent approach, which assembles two key
intermediates in the last step to form the final molecule, which
of H2SO4 at elevated temperature. The product was crystal-
is then isolated by pH-controlled extraction. The new route
lized directly from MeOH at 0 °C. Residual acid was
offers significant yield and purity advantages over the previous
neutralized by adding Et3N to the cake wash.
route. However, the solvent volume and cycle time were not
Installation and retention of the benzylic chiral center in
fully optimized due to the termination of the project.
7 were major challenges in previous syntheses, and this route
was no different. Borane-mediated reduction with ox-
azaborolidine catalysts 1-3 did not yield a clean reaction. The
asymmetric hydrogenation with (1S,2S)-N(p-toluenesulfon-
yl)-1,2-diphenylethylenediamine as described by Noyori and
co-workers4 did not generate any product after 4 days at 80
°C. Blocking the sulfonamide with an N-acyl group did not
appear to help. Borohydride reduction with 1,2,5,6-di-O-
isopropylidene-a-D-glucofuranose as the chiral ancillary5
required a large excess of reagents to achieve higher than
99% conversion. The best reagent we found for ketone
reduction was (-)-B-chlorodiisopinocampheylborane [(-)-
DIP-Cl]. Brown and co-workers reported the use of (-)-
DIP-Cl to reduce prochiral ketones to their corresponding
secondary alcohols with high enantioselectivity.6 Using
Brown’s conditions, reduction of 6 in THF at 0-5 °C was
very slow, requiring 3 days for complete consumption of
starting material. The low reactivity is likely a consequence
of the poor solubility of 6 in THF. A solubility study revealed
that the ester had poor solubility in most solvents that were
compatible with the reduction conditions; therefore THF was
chosen based on the quality of product obtained. Although
increasing the reaction temperature helped to reduce the
reaction time, lower selectivity was observed (Table 1).
The reduction was quenched with acetone, and the HCl
byproduct was neutralized with NaHCO3. After a solvent
exchange to methanol, the R-pinene could be completely
Introduction
Trecetilide hemi-fumarate (1) was a compound developed
for the chronic treatment of atrial arrythmia. Previous
synthetic efforts followed a linear approach (Scheme 1). Each
involved early introduction of the unstable tertiary fluorine
group that led to quality issues with the final product due to
impurities arising from elimination and hydrolysis. Another
major issue was the introduction of the chiral aryl alcohol.
Early methods involved a nonselective reduction followed
by an enzymatic resolution. The enzymatic resolution was
highly selective, but chromatographic separation was neces-
sary to purge achiral impurities. Several intermediates also
had handling problems due to poor crystallinity. Additionally,
the final product prepared by the original method did not
meet the developmental purity goals (>96% ee and >95 wt
% achiral purity) making it necessary to develop this new
process.
Thus, improving both the chiral and chemical purities
without column chromatography became the main focus of
our efforts to develop a commercializable new route. The
success of the new route relied heavily on an asymmetric
reduction approach to install the aryl alcohol functionality
and efficient late stage installation of the labile tertiary
fluoroalkyl group. The convergent route developed for large-
scale preparation is outlined in Scheme 2.
(1) (a) Hirao, A.; Itsuno, S.; Nakahama, S.; Yamazaki, N. J. Chem. Soc., Chem.
Commun. 1981, 315-317. (b) Itsuno, S.; Nakano, M.; Miyazaki, K.;
Masuda, H.; Ito, K.; Hirao, A.; Nakahama, S. J. Chem. Soc., Perkin Trans.
1 1985, 2039-2044.
(2) Didier, E.; Loubinoux, B.; Ramos Tombo, G. M.; Rihs, G. Tetrahedron
1991, 47, 4941-4958.
(3) (a) Corey, E. J.; Bakshi, R. K.; Shibata, S. J. Am. Chem. Soc. 1987, 109,
5551-5553. (b) Review: Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed.
1998, 37, 1986-2012.
Results and Discussion
The new route also starts with 4-{4-[(methylsulfonyl)-
amino]phenyl}-4-oxobutanoic acid (2), which is commer-
cially available. The Fisher esterification was straightforward
to convert 2 to its methyl ester 6 in the presence of 5 mol %
(4) Hashiguchi, S.; Fujii, A.; Takehara, J.; Ikariya, T.; Noyori, R. J. Am. Chem.
Soc. 1995, 117, 7562-7563.
(5) Hirao, A.; Nakahama, S.; Mochizuki, D.; Itsuno, S.; Ohowa, M.; Yamazaki,
N. J. Chem. Soc., Chem. Commun. 1979, 807-808.
(6) (a) Chandrasekharan, J.; Ramachandran, P. V.; Brown, H. C. J. Org. Chem.
1985, 50, 5446-5448. (b) Brown, H. C.; Ramachandran, P. V. Acc. Chem.
Res. 1992, 25, 16-24.
* To whom correspondence should be addressed. Telephone: 269-833-3546.
E-mail: haifeng.wu@pfizer.com.
† Chemical Research & Development, Pfizer Corporation, Kalamazoo, MI.
‡ Present address: 3M Pharmaceuticals-Chemistry, St. Paul, MN.
§ Present address: Pfizer Global Research & Development, Pfizer Corporation,
La Jolla, CA.
174
•
Vol. 9, No. 2, 2005 / Organic Process Research & Development
10.1021/op049799f CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/01/2005

Scheme 1. Initial synthetic route
Scheme 2. Convergent route to trecetilide hemi-fumarate
Table 1. Temperature effect on the (-)-DIP-Cl reduction
protonation of the lactone would set up an equilibrium
between 13 and the ring opened p-benzoquinone monoimine
methide (12). Subsequent ring closure of that compound to
the lactone would be expected to be an achiral process. We
found that the racemization could be minimized by running
the reaction at 0 °C. The amount of p-TsOH also had
significant impact on the chiral purity of the product (Table
2). A catalytic amount of p-TsOH, 1.5 mol %, was selected
as it provided the highest %ee. Other acids such as
trichloroacetic acid (TCA) and trifluoroacetic acid (TFA) also
promoted the ring closure at 0 °C, but both required higher
loading, 1 equiv and 0.5 equiv, respectively, to achieve
complete conversion (Table 3).
area %
purity of
crude
temp
(°C)
time
(day)
% ee
of crude
23
4
-15
1
3
5
93.3
94.0
93.3
93
95
97
removed through multiple heptane washes. Another solvent
exchange to methylene chloride allowed the isolation of 7
as a crystalline solid in 75-85% yield with greater than 98%
ee.
Conversion of the hydroxy methyl ester (7) to lactone 8
under acidic conditions resulted in some erosion of chiral
purity. An interesting observation was that, under identical
conditions, the N-acylated hydroxy ester lactonized without
racemization. This suggested participation of the sulfonamide
functionality in the racemization pathway. A possible mech-
anism is described in Scheme 3. Thus, after lactonization,
Lactone 8 in CH2Cl2 was subjected to DIBAL-H reduction
in toluene to yield lactol 9 in 64% yield and greater than
93% chemical purity. The crude product was then recrystal-
lized from EtOAc to achieve greater than 99% ee. Reductive
amination of the lactol 9 with amine 10 gave the free base
(11). Due to the equipment limitations that ensue from the
use of hydrogenation, we preferred the use of hydride for
Vol. 9, No. 2, 2005 / Organic Process Research & Development
•
175

Scheme 3. Proposed mechanism for lactone racemization
Table 2. Influence of p-TsOH quantity on chiral purity
extraction of 11. This method provided the free base (11)
with greater than 90% chemical purity and greater than 90%
yield. The chemical purity of the product was further
upgraded to greater than 98% during the hemi-fumarate salt
formation. The chiral purity remained unchanged in the last
two steps.
equiv of
p-TsOH
time
(h)
% conv
% ee
0.25
0.1
0.015
1
1
4
99
98
99
85
96
98
Table 3. Results for lactonization at 0 °C using TCA and
TFA
Conclusions
A convergent synthesis for trecetilide hemi-fumarate has
been developed. The new process was demonstrated on a
20 g scale in the lab to obtain the final product in 50.9%
overall yield in six steps with 99% ee and only 1-1.5 wt %
achiral impurities, thus meeting the developmental goals. The
intermediates were all easily filterable solids. Clean instal-
lation of the benzylic chiral center was accomplished by
reduction with (-)-DIP-Cl. The hydroxy methyl ester (7)
and the lactol (9) were capable of chiral upgrading through
crystallization. The final coupling of lactol 9 and fluoroamine
10 was achieved through a reductive amination using sodium
triacetoxyborohydride. Installation of the labile tertiary
fluoroalkyl functionality late in the sequence provided quality
advantages over the previous route. A unique use of ethyl
acetate as the solvent for the triacetoxyborohydride reductive
amination simplified the workup. Use of a pH-controlled
extractive procedure allowed the removal of many of the
achiral impurities from the free base (11) without chroma-
tography. Both the chemical and chiral purity profiles of the
hemi-fumarate salt (1) prepared by this new route are superior
to those from the previous route. However, the efforts to
further improve this process, such as increasing the solvent
and cycle-time efficiency, were curtailed due to termination
of the project.
equiv of
acid
time
(h)
% conv.
%ee
1 equiv of
TCA
0.5 equiv of
TFA
3
97
99
98
97
3.5
this application. Sodium cyanoborohydride is commonly used
in laboratory scale reductive amination reactions, but the
toxicity of the reagent and its byproducts make it unattractive
for use on large scale. An alternative, sodium triacetoxy-
borohydride, has been reported as a general reducing agent
for reductive aminations7 and was successful in this case.
We chose to run the reductive amination as a two-step
process as shown in Scheme 4. The selection of EtOAc as
the solvent simplified the workup. The reaction of lactol 9
and fluoroamine 10 at 20 °C afforded the iminium salt 14
as a solution, which was subsequently transferred to a NaBH-
(OAc)3 slurry at 0-4 °C to afford the desired product. The
crude product was purified via a pH-controlled extractive
workup. The free base was completely soluble in the aqueous
phase below pH 6.5, while most of the impurities were
eliminated in the organic phase. After separation, the pH of
the aqueous layer was raised to 8-8.5, followed by EtOAc
Scheme 4. Stepwise reductive amination
176
•
Vol. 9, No. 2, 2005 / Organic Process Research & Development

Experimental Section
under reduced pressure and swapped to MeOH (200 mL).
This MeOH solution was washed with heptanes (4 × 200
mL), followed by distillation to a minimum volume under
reduced pressure. CH2Cl2 (3 × 100 mL) was added followed
by distillation to a volume of 20 mL. After adding CH2Cl2
(200 mL), the mixture was cooled to -15 °C and held for 4
h, the product was collected by filtration in 84.9% yield (17.1
General. All reagents were purchased from commercial
suppliers and used as received unless otherwise noted. NMR
spectra were measured on a Bruker AM-400 operated at 400
and 100 MHz, for 1H and 13C, respectively. Elemental
analysis was obtained from Pharmacia and Upjohn Physical
and Analytical Chemistry. HPLC analyses were carried out
on a Dionex DX 500 Chromatography System. Analysis of
the chemical purity was conducted by HPLC using the
following conditions: column, 4.6 × 150 mm2 Luna 3m C18;
mobile phase, 45:55 acetonitrile/water; flow rate, 1.0 mL/
min; detector, 254 nm. Analysis of the chiral purity of 8
was conducted by HPLC using the following conditions:
column, 4.6 × 250 mm2 Chiralpak AS; mobile phase, EtOH
(0.05% TFA); flow rate, 0.4 mL/min; detector, 230 nm.
Analysis of the chiral purity of 9 was conducted by HPLC
using the following conditions: column, 4.6 × 250 mm2
Chiralcel OF; mobile phase, 50:50 IPA/heptanes (0.05%
TFA); flow rate, 0.5 mL/min; detector, 230 nm.
Methyl 4-{4-[(Methylsulfonyl)amino]phenyl}-4-oxobu-
tanoate (6). A 1000 mL round-bottom flask equipped with
a reflux condenser, heating mantle, overhead stirrer, and
nitrogen inlet was charged with 4-{4-[(methylsulfonyl)amino]-
phenyl}-4-oxobutanoic acid (2) (50 g, 0.18 mol), methanol
(500 mL), and concentrated sulfuric acid (0.92 g, 0.0094
mol). The slurry became a solution after being heated at 60
°C for 4 h. The reaction was monitored by HPLC. After the
reaction was complete, the solution was cooled to 0 °C and
stirred for 4 h to allow the product to crystallize. The solid
product was isolated by filtration and washed with a mixture
of cold methanol and triethylamine, followed by cold
methanol. Drying afforded ethyl {[(methylsulfonyl)amino]-
phenyl}-4-oxobutanoate (6) (51.2 g, 97.3% yield) of 99 area
1
g, 92.1% chem, 98.4% ee). Melting point: 70-72 °C. H
NMR (400 MHz, CDCl3) δ 7.09 (d, J ) 8.4 Hz, 2 H), 6.96
(dd, J ) 4.9, 8.4 Hz, 2 H), 4.51 (dd, J ) 6.3, 6.4 Hz, 1 H),
3.45 (s, 3 H), 2.77 (s, 3 H), 2.21 (t, J ) 7.2 Hz, 2 H), 1.81
(dt, J ) 6.8, 6.8 Hz, 2 H); 13C NMR (100 MHz, CDCl3) δ
C 174.4, 141.2, 136.1; CH 127.1, 120.9, 72.8; CH2 33.7,
30.3; CH3 51.8, 39.2; MS m/z calcd for C13H19NO5S 286.075
(M - H)+, found 285.94 (M - H)+.
N-{4-[(2S)-5-Oxotetrahydrofuran-2-yl]phenyl}-
methanesulfonamide (8). A 1000 mL round-bottom flask
equipped with an overhead stirrer and nitrogen inlet was
charged with hydroxy ester 7 (17.9 g, 62.3 mmol) and CH2-
Cl2 (350 mL). The resulting slurry was stirred at -5 to 0 °C
followed by the addition of p-TsOH (0.12 g, 0.94 mmol).
After the reaction was determined to be complete by HPLC,
the mixture was washed with cold H2O (2 × 250 mL) and
brine (250 mL) while maintaining the temperature at 0 °C.
The organic layer was dried with MgSO4, filtered, and
concentrated under reduced pressure. The resulting solids
were recrystallized from EtOAc (25 mL) to provide 12.1 g
(76% yield) of white solids (95 area %, 97% ee). Mp 101-
1
102 °C. H NMR (400 MHz, CDCl3) δ 7.24 (dt, J ) 8.0,
8.6 Hz, 4 H), 5.43 (m, 1 H), 2.97 (s, 3 H), 2.63 (m, 3 H),
2.14 (m, 1 H); 13C NMR (100 MHz, CDCl3) δ C 177.0,
137.2, 136.0; CH 126.8, 120.7, 80.9; CH2 30.7, 29.0; CH3
39.4; MS m/z calcd for C11H13NO4S 255.056 (M+), found
254.95 (M+).
1
% purity. Mp 181-183 °C. H NMR (400 MHz, DMSO-
d6) δ 10.32 (s, 1 H), 7.95 (d, J ) 8.4 Hz, 2 H), 7.30 (d, J )
8.8 Hz, 2 H), 3.58 (s, 3 H), 3.32 (s, 1 H), 3.24 (t, J ) 6.4
Hz, 2 H), 3.1 (s, 3 H), 2.63 (t, J ) 6.4 Hz, 2 H); 13C NMR
(100 MHz, DMSO-d6) δ C 196.8, 172.8, 143.0, 131.0; CH
129.6, 117.5; CH2 32.7, 27.6; CH3 51.3, 39.8; MS m/z calcd
for C12H14NO5S 284.31 (M - H)+, found 284.00 (M - H)+.
Methyl (4S)-4-Hydroxy-4-{4-[(methylsulfonyl)amino]-
phenyl}butanoate (7). A 1000 mL round-bottom flask
equipped with an overhead stirrer and nitrogen inlet was
charged with methyl ester 6 (20.0 g, 0.070 mol) and
tetrahydrofuran (300 mL) at 0 °C. A solution of (-)-DIP-
Cl (48.3 g, 0.15 mol) in 100 mL of tetrahydrofuran was
added dropwise to the above slurry while maintaining the
pot temperature below 0 °C. The reaction mixture was stirred
for 72-96 h at 0 °C to give a colorless solution. When the
reaction was determined to be complete by HPLC, cold
acetone (75 mL) was added slowly to quench the reaction
while maintaining the pot temperature between 0 and 5 °C.
The reaction mixture was stirred at 0 °C for 1 h followed
by the addition of EtOAc (100 mL) and warming to 23 °C.
The mixture was washed with a 9% aqueous solution of
NaHCO3 (2 × 250 mL). The organic layer was concentrated
N-{4-[(2S)-5-Hydroxytetrahydrofuran-2-yl]phenyl}-
methanesulfonamide (9). A 500 mL round-bottom flask
equipped with an overhead stirrer and nitrogen inlet was
charged with lactone 8 (9.8 g, 38.5 mmol), CH2Cl2 (125 mL),
and toluene (75 mL). After cooling the mixture to -30 °C,
DIBAL-H (59 mL, 1.5 M in toluene) was added at a rate of
1 mL/min, keeping the temperature below -25 °C. After
the addition was complete, the solution was stirred at -30
°C until HPLC analysis indicated the reaction was complete.
Excess DIBAL-H was prequenched with EtOAc (15 mL) at
-30 °C, and the reaction was treated with 1 M aqueous
disodium citrate (200 mL). The mixture was allowed to warm
to 23 °C. Ethyl acetate (200 mL) was added, and the phases
were separated. The aqueous layer was extracted with EtOAc
(2 × 100 mL), and the combined organic layers were washed
with brine (2 × 150 mL). After the organic volume was
reduced to 60 mL under reduced pressure, the solution was
cooled to 0 °C and held at this temperature for 3 h. The
product was collected on a filter as a white, powdery solid
(6.4 g, 64.9% yield, 93 area %, >99% ee). Mp 129-130
1
°C. H NMR (400 MHz, CD3OD) δ 7.42 (d, J ) 8.2 Hz, 1
H), 7.29 (d, J ) 8 Hz, 1 H), 7.21 (d, J ) 8.1 Hz, 2 H), 5.65
(m, 0.5 H), 5.53 (m, 0.5 H), 5.14 (t, J ) 6.9 Hz, 0.5 H),
4.93 (t, J ) 7.6 Hz, 0.5 H), 2.92 (s, 3 H), 2.49 (m, 1 H),
(7) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.; Shah,
R. D. J. Org. Chem. 1996, 61, 3849-3862
Vol. 9, No. 2, 2005 / Organic Process Research & Development
•
177

2.25 (m, 1 H), 1.98 (m, 1 H), 1.92 (m, 1 H); 13C NMR (100
MHz, CD3OD) δ C 142.9, 141.2, 140.6, 138.6, 138.5; CH
128.4, 127.9, 121.7, 121.6, 99.9, 99.5, 83.2, 80.3, 74.4, 74.3;
CH2 35.5, 35.1, 34.4, 34.2, 33.9; CH3 39.1, 39.0. MS (EI)
m/z calcd for C11H15NO4S 257.07 (M+); found 256.93 (M+).
N-{4-[(2S)-5-Hydroxytetrahydrofuran-2-yl]phenyl}-
methanesulfonamide (11). A mixture of lactol 9 (4.02 g,
15.62 mmol) and fluoroamine 10 (2.77 g, 15.62 mmol) was
stirred in EtOAc (30 mL) in a 100 mL flask at 23 °C until
a clear solution was obtained. The solution was cooled to 0
°C before being added to a cooled (0 °C) vigorously stirred
slurry of sodium triacetoxyborohydride (4.63 g, 21.87 mmol)
in EtOAc (10 mL). After HPLC indicated the reaction was
complete, cold water (40 mL) was added to quench the
excess reagent at 0 to 4 °C. While cooling was maintained,
the pH was raised to 6-6.5 using a 10% aqueous sodium
hydroxide solution. The layers were separated, and the
aqueous phase was extracted with EtOAc (2 × 40 mL). The
pH of the aqueous layer was further adjusted to 8-8.5 using
a 10% aqueous sodium hydroxide solution, followed by
extraction with EtOAc (3 × 30 mL). The combined organic
phase was concentrated under reduced pressure to provide
11 as a colorless oil (5.62 g) in 91.1% yield and 96.5% purity.
N-(4-{4-[Ethyl(6-fluoro-6-methylheptyl)amino]-1-
hydroxybutyl}phenyl)methanesulfonamide Hemi-fuma-
rate Salt (1). The free base (11) (7.78 g, 18.7 mmol) was
stirred in 10 mL of THF in a 100 mL flask equipped with
an overhead stirrer and a nitrogen inlet. A 125 mL flask was
charged with fumaric acid (1.08 g, 9.34 mmol) and THF
(80 mL) and heated at 40-45 °C to obtain solution. The
THF solution of 11 was then added to the fumaric acid
solution, and the mixture was stirred at 45 °C for 30 min.
After reducing the total volume to 45 mL under vacuum, 14
mL of branched octane was added while maintaining the pot
temperature at 45 to 50 °C. The solids were isolated by
filtration, and the cake was washed in two parts using a
mixture of 20 mL of THF and 5 mL of branched octane.
After a branched octane cake wash (2 × 25 mL) the product
was dried in vacuo at 40-45 °C to give 8.1 g of white solid
(LC assay 98.9 wt % purity, 92% yield).
Acknowledgment
We are grateful to Timothy W. Evans, Paul M. Herrinton,
Dilip K. Joshi, Vikram D. Kalthod, Moses W. McMillan,
Mark T. Maloney, and Thomas A Runge for their collabora-
tion on this project. We would especially like to thank Larry
D. Kissinger, Simpson D. Putnam, Michael T. Radvan, Fred
J. Antosz, and Jerome A. Morris for analytical support.
Received for review November 4, 2004.
OP049799F
178
•
Vol. 9, No. 2, 2005 / Organic Process Research & Development