Development of a scalable synthesis of a serotonin receptor antagonist

Article abstract of DOI:10.1021/op5003402

An efficient process was developed for the manufacture of MSA100, a serotonin receptor antagonist, via a five-step synthetic route furnishing a high quality of active pharmaceutical ingredient. Highlights of this synthesis include: (1) replacing carcinoge

Full text of DOI:10.1021/op5003402

Article
pubs.acs.org/OPRD
Development of a Scalable Synthesis of a Serotonin Receptor
Antagonist
Mahavir Prashad,* Yugang Liu, Wen-Chung Shieh,* Frank Schaefer, Bin Hu, and Michael Girgis
Chemical and Analytical Development, Novartis Pharmaceuticals Corporation, One Health Plaza, East Hanover, New Jersey 07936,
United States
ABSTRACT: An efficient process was developed for the manufacture of MSA100, a serotonin receptor antagonist, via a five-step
synthetic route furnishing a high quality of active pharmaceutical ingredient. Highlights of this synthesis include: (1) replacing
carcinogenic methyl iodide with methyl p-toluenesulfonate as the methylating reagent; (2) a hydrogenation protocol with
optimized temperature, pressure, and mass-transfer conditions that avoided one side product and reduced the other one
effectively; (3) chemical resolution employing D-camphoric acid in a mixed-solvent system; (4) amidation under anhydrous
conditions for controlling a Michael adduct impurity; and (5) plausible mechanisms for the formation of side products.
styrylpyridine 4 by reacting 2-nitrobenzaldehyde 2 with 2-
picoline 3 in acetic anhydride at 140 °C.1,5 Upon completion,
the reaction mixture was cooled to 15 °C and adjusted to pH
11 with 50% NaOH resulting in the formation of a granular
solid, which was collected by filtration. This process was scaled-
up on a 40-kg scale in the plant to afford 47.3 kg of 4 (79%
yield) with 99% purity.
INTRODUCTION
It was reported that the (S)-enantiomer of 2′-[2-(1-methyl-2-
piperidyl)ethyl]cinnamanilide (MSA100, 1) (Figure 1) is an
■
To prepare the methylpiperidine fragment of the target
molecule, the original synthesis treated 4 with methyl iodide
leading to methylpyridinium salt 5, which was then hydro-
genated to methylpiperidinium salt 6.1 We were concerned
about the carcinogenic property of methyl iodide and decided
to replace it with a less toxic methylating reagent. We found
that methyl p-toluenesulfonate is an excellent alternative.
Methylation of 4 with methyl p-toluenesulfonate in acetonitrile
at 82 °C for 24 h cleanly generated tosylate salt 10 (Scheme 2).
This salt precipitated out from the reaction mixture by adding
isopropyl acetate and was isolated by filtration. This simple and
efficient process was employed on a 47-kg scale in the plant to
furnish 80 kg of 10 (93% yield) with 99% purity.
Conversion of pyridinium salt 10 to the subsequent
piperidinium salt 11 by catalytic hydrogenation employing
platinum as the catalyst could generate side products 12 and 13
(Figure 2), about 25% each, if the hydrogen concentration and
its mass-transfer rate are not sufficient. Both side products were
isolated and characterized by NMR. The plausible mechanism
for their formations is shown in Scheme 3.
We postulated that the conversion of nitropyridinium
tosylate 10 to piperidinium salt 11 by hydrogenation went by
pathway A involving both nitroso 10a and aniline 10b (Scheme
3). Both intermediates can be detected and characterized by
LC/MS analyses. As suggested by HPLC analysis of the
reaction mixture, reductions of the double bond and the
pyridinium ring of 10b to piperidine 11 were much slower than
that of the nitro group of 10a to aniline 10b. If conditions were
not optimum for pathway A, such as low hydrogenation
Figure 1. MSA100, a serotonin receptor antagonist.
active 5-HT (5-hydroxytryptamine or serotonin) receptor
antagonist; however, its (R)-isomer is totally or substantially
devoid of the same activity.1 Because we had interests in
investigating its therapeutic potential in humans, we needed to
develop a scalable synthesis allowing the production of
kilogram quantities of this active pharmaceutical ingredient
(API). The original synthesis of 1 established the molecule’s
chirality by chemical resolution using dibenzoyl-L-tartaric acid
as the resolving agent (Scheme 1).1a Use of chemical resolution
for the synthesis of enantiomerically pure drug substances has
been extensively demonstrated in the pharmaceutical industry2
as well as in our lab.3 Since this approach is practical and
reliable, we decided to utilize this strategy and explore its
potential in making enantiomerically pure 1 on kilogram scale.
During the course of exploring the feasibility of the original
synthetic route, we found several drawbacks. First, methyl
iodide is not a desirable methylating agent for 4, because it is a
potential carcinogen.4 Second, resolution with dibenzoyl-L-
tartaric acid afforded diastereomeric salt 7 in only 28.6% yield
and resulted in an overall yield of 10.7% for the original
synthesis. Herein, we disclose an improved synthesis that was
employed successfully on plant-scale.
RESULTS AND DISCUSSION
Synthesis of Phenethylpiperidine 11. Our manufactur-
ing synthesis of MSA100 (1) began with the preparation of 2-
■
Received: October 29, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/op5003402 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
a
Scheme 1. Original Synthesis of 1
a
Reagents and conditions: (a) Ac2O, 140 °C, 100% as is (b) MeI, acetone, reflux, 80%; (c) H2, Pt/C, MeOH, 25 °C; 59%; (d) NaOH, then
dibenzoyl-L-tartaric acid, MeOH, 25 °C, 28.6%; (e) NaHCO3, EtOAc, 100% as is (f) K2CO3, EtOAc, 79%. Overall yield: 10.7%.
Scheme 2. Synthesis of Piperidinium Salt 11
product and side products during the hydrogenation of 10,
efforts were focused on optimizing these conditions. Employing
the improved conditions (30 °C, 75 psi, and 450 rpm), we were
able to generate 11 containing no detectable amount of 13 and
14% (by area relative to 11) of 12. Side product 12 was not a
problem because it was easily removed from 11 by
crystallization. This protocol was utilized on 20-kg scale three
times in the plant to afford a total of 40.6 kg (71% yield) of 11
Figure 2. Two hydrogenation side products.
with 99% purity.
Chemical Resolution. Dibenzoyl-L-tartaric acid was
utilized in the original synthesis as the resolving agent for
separating the free base of racemic 11 and resulted in the
isolation of the desired diastereomeric salt in 28.6% yield
(57.2% theory).1a To increase the throughput, alternative
resolving agents were investigated. We found that D-camphoric
acid (D-CA) consistently gave superior results under various
conditions (Scheme 4). The best diastereomeric ratio of
98.6:1.4 for salt 14 was obtained in 47% yield (94% theory)
when the resolution was conducted in a mixture of ethanol and
isopropyl alcohol. Further recrystallization of 14 in the same
solvent system can enhance the diastereomeric ratio to 99.9:0.1
in 94% recovery, an overall yield of 44% (88% theory) from 11.
This process was scaled up on 40-kg scale in the plant to
produce 17.4 kg (41% yield) of 14 with 99.9% diastereomeric
purity.
Amidation. To synthesize the final active pharmaceutical
ingredient 1, diastereomaric salt 14 was liberated to free-base 8
by treating it with an aqueous NaOH solution in isopropyl
acetate (Scheme 5). Amidation of 8 with cinnamoyl chloride 9
employing a biphasic system was problematic. For instance,
treating the isopropyl acetate solution of 8 with 9 employing
the Schotten−Baumann conditions7 with an aqueous solution
concentration or slow agitation rate (mass transfer), both
pathways B and C could compete with pathway A. Through
pathway B, an intramolecular cyclization reaction of aniline 10b
could give a spiro-intermediate 10c. After subsequent three
transformations: hydrogenation to 10d, rearrangement to 10e,
and hydrogenation of 10e, 10b could lead to side product 12.
Under another scenario, a thermal [2 + 2]-cycloaddition
reaction between nitroso intermediate 10a and starting material
10 via pathway C is conceivable.6 The four-membered ring
intermediate 10f thus formed could undergo a retro [2 + 2]-
cycloaddition reaction to afford an imine intermediate 10g.
Further hydrogenation of 10g could generate side product 13.
To prove that 13 could be formed under hydrogen-deficient
conditions, an experiment was conducted in which the
hydrogen supply was interrupted as soon as 10 was converted
to the nitrosopyridium tosylate 10a. The reaction mixture was
then heated to 65 °C in the absence of hydrogen gas and stirred
for an additional 1.5 h. At this point, hydrogen supply was
resumed and the hydrogenation reaction was allowed to go to
completion. As expected, we observed more than 25% side
product 13.
Because we noticed that both hydrogen concentration and
mass-transfer rate had significant impacts on the profiles of
B
dx.doi.org/10.1021/op5003402 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 3. Plausible Mechanism for the Formation of Side Products 12 and 13
15 was decreased to 1.3%. Further reduction of 15 to less than
0.1% was achieved by crystallization. To obtain the anhydrous
solution of 8 without using any drying agent, the isopropyl
acetate solution of 8 containing residual amounts of water was
concentrated by azeotropic distillation to ensure a complete
removal of water. As a result, amidation of 8 with cinnamoyl
chloride in the presence of solid K2CO3 was carried out on 16-
kg scale in the plant to produce 1 (10.9 kg, 83% yield) with
99.1% purity and 99.8% ee.
Scheme 4. Chemical Resolution
of either K2CO3 or NaHCO3 resulted in the formation of a side
product in 33−42% range. This side product was characterized
by NMR and LC/MS as a Michael adduct 15, arising from
Michael addition of aniline 8 to product 1 (Scheme 6).
Replacing the above biphasic solution with an anhydrous
system employing K2CO3 as the acid scavenger, similar
conditions used in the original synthesis,1a the side product
CONCLUSIONS
■
An efficient process was developed for the manufacture of
MSA100 (1), a serotonin receptor antagonist, via a five-step
synthetic route furnishing a high quality of active pharmaceut-
ical ingredient. The overall yield of the manufacturing synthesis
Scheme 5. Amidation
C
dx.doi.org/10.1021/op5003402 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
Scheme 6. Formation of Side Product 15
Scheme 7. Manufacturing Synthesis of 1
a
Reagents and conditions: (a) Ac2O, 140 °C; (b) methyl p-toluenesulfonate, CH3CN, 82 °C; (c) H2 (75 psi), 10% Pt/C, MeOH, 30 °C, 450 rpm;
(d) NaOH, then (1R, 3S)-(+)-camphoric acid, EtOH-iPrOH, 65−23 °C; (e) NaOH, iPrOAc; (f) K2CO3, iPrOAc, 85 °C. Overall yield: 17.8%.
of 1 (Scheme 7) was increased from the original 10.8 to 17.8%.
Highlights of this synthesis include the following: (1) replacing
carcinogenic methyl iodide with methyl p-toluenesulfonate as
the methylating reagent; (2) a hydrogenation protocol with
optimized temperature, pressure, and mass-transfer conditions
that avoided one side product and reduced the other one
effectively; (3) chemical resolution employing D-camphoric
acid in a mixed-solvent system; (4) amidation under anhydrous
conditions for controlling a Michael adduct impurity; and (5)
plausible mechanisms for the formation of side products.
8.64 min); 3 (tR = 1.63 min): Waters Symmetry-C18 150 × 4.6
mm, flow rate = 1 mL/min, 25 °C, gradient elution from 25:75
A−B (held for 5 min) to 90:10 A−B over 2 min, and held at
90:10 A−B (5 min); A = acetonitrile; B = 0.1% TFA in water;
UV λ = 254 nm.
1-Methyl-2-[(E)-2-(2-nitrophenyl)-ethenyl]-pyridinium 4-
methylbenzenesulfonate (10). A mixture of 4 (80 g, 0.354
mol), methyl p-toluenesulfonate (98.8 g, 0.530 mol), and
acetonitrile (400 mL) was heated to 82 °C and stirred for 24 h.
Isopropyl acetate (400 mL) was added at 82 °C. The
suspension was cooled to 25 °C over 1 h and stirred for 4 h.
The precipitate was filtered, rinsed with isopropyl acetate (2 ×
160 mL), and dried under reduced pressure (15−40 mbar) at
60 °C for 16 h to afford 10 (135 g, 93% yield) as a solid: mp
172−173 °C; HPLC for 10 (tR = 2.60 and 3.47 min) 99.7%
purity; 4 (tR = 4.60 min); methyl p-toluenesulfonate (tR = 9.33
min): Waters Symmetry-C18 150 × 4.6 mm, flow rate = 1 mL/
min, 25 °C, gradient elution from 25:75 A−B (held for 5 min)
to 90:10 A−B over 2 min, and held at 90:10 A−B (5 min); A =
acetonitrile; B = 0.1% TFA in water; UV λ = 254 nm.
2-[2-(1-Methyl-2-piperidinyl)ethyl]-benzenamine 4-meth-
ylbenzenesulfonate (1:1) (11). To an inert hydrogenation
vessel was charged 10 (43.9 g, 0.106 mol), 10% Pt/C (1.87 g,
62.4% wet), and methanol (396 g) under N2 atmosphere. The
headspace was purged with N2, and the vessel was charged with
H2 by pressurizing with H2 to 65 psi, followed by
depressurizing to 14.5 psi. The H2 pressurization/depressuriza-
tion cycle was repeated 4 times. After the final depressurization,
the reactor’s temperature, hydrogen pressure, and agitation
speed were set at 30 °C, 75 psi, and 450 rpm, respectively.
Hydrogenation was maintained under these conditions for an
EXPERIMENTAL SECTION
■
General. Reverse phase HPLC analyses were performed on
an Agilent HPLC system with a DAD detector (area
normalization).
2-[(E)-2-(2-Nitrophenyl)ethenyl]-pyridine (4). A mixture of
2-nitrobenzaldehyde 2 (150 g, 0.99 mol), 2-picoline 3 (130 g,
1.39 mol), and acetic anhydride (282 mL) was heated to 140
°C and stirred for 28 h. The resulting dark mixture was cooled
to 10 °C. Water (750 mL) was added while maintaining the
batch temperature below 35 °C. The mixture was cooled to 15
°C. A 50% (w/w) aqueous NaOH solution (263 g) was added
while maintaining the batch temperature below 35 °C. Seeds
(4, 120 mg) were added. More 50% (w/w) aqueous NaOH
solution (263 g) was added while maintaining the batch
temperature below 35 °C. The resulting suspension was stirred
at 35 °C for 1 h. The precipitate was filtered, rinsed with water
(4 × 375 mL), and dried under reduced pressure (15−40
mbar) at 70 °C for 16 h to afford 4 (195 g, 87% yield) as a
solid: mp 95−96 °C (lit. ref 5, mp 98−99 °C); HPLC for 4 (tR
= 4.60 min, identical to authentic sample) 99.5% purity; 2 (tR =
D
dx.doi.org/10.1021/op5003402 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
additional 6 h. The batch was cooled to 25 °C. The vessel was
depressurized, purged with N2 by pressurizing to 65 psi, and
depressurized. The pressurizing−depressurizing cycle was
repeated five times. The mixture was filtered over a pad of
Celite (8 g) and rinsed with methanol (2 × 45 g). The
combined filtrate was concentrated at 35−45 °C under reduced
pressure (80−160 mbar) until a final volume of ∼150 mL was
reached. 2-Propanol (353 g) was added and the mixture was
concentrated at 35−45 °C under reduced pressure (80−160
mbar) until a final volume of ∼150 mL was reached. 2-
Propanol (353 g) was added one more time and the mixture
was concentrated at 35−45 °C under reduced pressure (80−
160 mbar) until a final volume of ∼150 mL was reached. The
residue was heated to 60 °C and isopropyl acetate (44 g) was
added while maintaining the temperature at 55−65 °C. The
mixture was cooled to 40 °C over a period of 20 min and
seeded with 11 (160 mg). The mixture was cooled to 20 °C
over a period of 1 h and stirred for 4 h. The precipitate was
filtered, rinsed with a solution of 2-propanol/isopropyl acetate
(1:2 v/v) (2 × 42 g), and dried under reduced pressure (15−40
mbar) at 60 °C for 16 h to afford 11 (26.3 g, 64% yield) as a
solid: mp 133−135 °C; HPLC for 11 (tR = 5.20 min) 98.8%
purity: Waters YMC ODS-AQ S-3 120 A, 150 × 3.0 mm, flow
rate = 0.8 mL/min, 25 °C, gradient elution from 90:10 A−B to
40:60 A−B over 22 min; A = 10 mM NH4OAc in water ; B =
acetonitrile; UV λ = 240 nm.
(S)-2-[2-(1-Methyl-2-piperidinyl)ethyl]-benzenamine
(1R,3S)-(+)-camphoric acid salt (1:1) (14). To a mixture of 11
(30 g, 76.8 mmol) and isopropyl acetate (200 mL) was added a
solution of NaOH (4 g, 100 mmol) in water (50 mL). The
suspension was stirred until a solution was obtained. The
organic layer was separated and saved. The aqueous layer was
extracted with isopropyl acetate (67 mL). The combined
organic layer was washed with water (50 mL) and concentrated
at 20−40 °C under reduced pressure (20−100 mbar) until a
final volume of ∼25 mL was reached. 2-Propanol (50 mL) was
added and concentrated at 20−40 °C under reduced pressure
(20−100 mbar) until a final volume of ∼25 mL was reached.
More 2-propanol (50 mL) was added and concentrated at 20−
40 °C under reduced pressure (20−100 mbar) until a final
volume of ∼25 mL was reached. The residue was dissolved in
2-propanol (100 mL) and added to a solution of (1R,3S)-
(+)-camphoric acid (15.4 g, 76.8 mmol) in anhydrous ethanol
(100 mL) at 65 °C. More 2-proponol (100 mL) was added and
seeded with 14 (10 mg). The mixture was cooled to 23 °C over
a period of 2 h and stirred for an additional 2 h. The precipitate
was filtered, rinsed with 2-proponol (2 × 50 mL), dried at 45−
50 °C under reduced pressure (13−40 mbar) for 16 h to obtain
crude 14 (15.1 g) as a white solid (enantiomeric purity of free
base 8: 98.6:1.4 er).
(2E)-N-[2-[2-[(2S)-1-Methyl-2-piperidinyl]ethyl]phenyl]-3-
phenyl-2-propenamide (1). To a mixture of 14 (6.28 g, 15
mmol) and isopropyl acetate (60 mL) was added a solution of
NaOH (1.6 g, 40 mmol) in water (20 mL). The suspension was
stirred until a solution was obtained. The organic layer was
separated and saved. The aqueous layer was extracted with
isopropyl acetate (20 mL). The combined organic layer was
washed with water (20 mL) and concentrated at 20−40 °C
under reduced pressure (20−100 mbar) until a final volume of
∼65 mL was reached to obtain a solution of free base 8.
Potassium carbonate (6.22 g) and cinnamoyl chloride 9 (3.75 g,
22.5 mmol) was added. The suspension was stirred at 85 °C for
2 h and cooled to 25 °C. Water (50 mL) was added and stirred
for 30 min to obtain a biphasic solution. The organic layer was
separated and stirred with 0.5 N aqueous HCl solution (80 mL,
40 mmol). The aqueous solution was separated, diluted with
isopropyl acetate (60 mL), and treated with a solution of
NaOH (2 g, 50 mmol) in water (25 mL). The organic layer was
separated and saved. The aqueous layer was extracted with
isopropyl acetate (60 mL). The combined organic layer was
washed with water (40 mL) and concentrated at 20−40 °C
under reduced pressure (20−100 mbar) until a final volume of
∼22 mL was reached. The residue was heated to 85 °C, and
heptane (96 mL) was added while maintaining temperature at
85 °C. The mixture was cooled to 25 °C over a period of 1 h,
and stirred for 2 h. The precipitate was filtered, rinsed with a
solution of heptane/isopropyl acetate (6/1 v/v) (2 × 14 mL),
dried at 45−50 °C under reduced pressure (13−40 mbar) for
16 h to obtain 1 (4.06 g, 78% yield) as an off-white solid: mp
125−127 °C (lit. ref 1a, mp 128 °C); Chiral HPLC for (S)-1
(tR = 19.3 min), >99.9% ee; (R)-1 (tR = 18.5 min): Chiralcel
AD-H, 250 × 4.6 mm, flow rate = 1.0 mL/min, 25 °C,
900:100:1 A:B:C isocratic; A = hexanes; B = ethanol; C =
diethylamine; UV λ = 230 nm. HPLC for 1 (tR = 11.2 min)
99.8% purity; 8 (tR = 5.4 min); 9 (tR = 12.3 min): Waters
Symmetry-C18 150 × 4.6 mm, flow rate = 1 mL/min, 25 °C,
gradient elution from 93:7 A−B to 85:15 A−B over 5 min, to
10:90 A−B over 10 min and held for 2 min, to 93:7 A−B over 1
min; A = 0.1% TFA in water; B = acetonitrile; UV λ = 230 nm.
AUTHOR INFORMATION
Corresponding Authors
*E-mail: mahavir.prashad@novartis.com.
*E-mail: wen-chung.shieh@novartis.com.
■
Notes
The authors declare no competing financial interest.
REFERENCES
■
(1) (a) Amer, M. S., U.S. Patent 5,780, 487, 1998. (b) Prashad, M.;
Liu, Y.; Hu, B.; Girgis, M. J.; Schaefer, F., WO2007/111705, 2007.
(c) Prashad, M.; Liu, Y.; Hu, B.; Girgis, M. J.; Schaefer, F., WO2007/
111706, 2007.
Recrystallization. A mixture of crude 14 (15.1 g),
anhydrous ethanol (40 mL), and 2-propanol (50 mL) was
heated to 78 °C and stirred for 1 h. The mixture was cooled to
25 °C over a period of 2 h. The resulting slurry was cooled to 5
°C over a period of 30 min and diluted with 2-propanol (40
mL) at 5 °C. The precipitate was filtered, rinsed with 2-
propanol (2 × 30 mL), dried at 45−50 °C under reduced
pressure (13−40 mbar) for 16 h to obtain 14 (14.2 g, 44%
yield) as a white solid: mp 139−142 °C; Chiral HPLC for 8S
free base (tR = 7.80 min), 99.9:0.1 er; 8R free base (tR = 10.3
min): Chiralcel AD-H, 250 × 4.6 mm, flow rate = 1.0 mL/min,
25 °C, 900:100:1 A−B−C isocratic; A = hexanes; B = ethanol;
C = diethylamine; UV λ = 230 nm.
(2) (a) Brands, K. M. J.; Davies, A. Chem. Rev. 2006, 106, 2711−
2733. (b) Novel Optical Resolution Technologies; Sakai, K., Hirayama,
N., Tamura, R., Eds.; Springer-Verlag: Berlin, Heidelberg, and New
York, 2007; Topics in Current Chemistry, Vol. 269.
(3) (a) Hu, B.; Prashad, M.; Har, D.; Prasad, K.; Repic, O.; Blacklock,
T. Org. Process Res. Dev. 2007, 11, 90−93. (b) Prashad, M.; Hu, B.;
Repic, O.; Blacklock, T.; Giannousis, P. Org. Process Res. Dev. 2000, 4,
55−59. (c) Prashad, M.; Har, D.; Repic, O.; Blacklock, T.; Giannousis,
P. Tetrahedron: Asymmetry 1999, 10, 3111−3116.
(4) Shieh, W.-C.; Dell, S.; Repic, O. Org. Lett. 2001, 3, 4279−4281.
(5) Dykstra, S. J.; Minielli, J. L.; Lawson, J. E. J. Med. Chem. 1973, 16,
1015−1020.
E
dx.doi.org/10.1021/op5003402 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX

Organic Process Research & Development
Article
(6) Tsuge, O.; Torii, A. Bull. Chem. Soc. Jpn. 1972, 45, 3187−3191.
(7) For a review of Schotten−Baumann reaction, see: Sonntag, N. O.
V. Chem. Rev. 1953, 52, 237−416.
F
dx.doi.org/10.1021/op5003402 | Org. Process Res. Dev. XXXX, XXX, XXX−XXX