Development of an efficient synthesis for a nipecotate-containing immunopotentiator

Article abstract of DOI:10.1021/op049941c

The preparation of Elanco Animal Health immunopotentiator (S)-ethyl-1-(2-thiopheneacetyl)-3-piperidinecarboxylate (1) is described. The synthesis includes a new resolution of racemic ethyl nipecotate with dibenzoyl-L-tartaric acid. The resolved salt is found to couple directly with commercially available 2-thiopheneacetyl chloride under environmentally friendly Schotten-Baumann conditions to afford the amide in high yield. The final product is an oil which is purified by wiped film evaporative distillation.

Full text of DOI:10.1021/op049941c

Organic Process Research & Development 2004, 8, 593−596
Development of an Efficient Synthesis for a Nipecotate-Containing
Immunopotentiator
Eric D. Moher,* Allie E. Tripp, Lawrence C. Creemer,† and Jeffrey T. Vicenzi
Eli Lilly and Company, Chemical Product Research and DeVelopment DiVision, Indianapolis, Indiana 46285, U.S.A., and
Elanco Animal Health Research and DeVelopment, Greenfield, Indiana 46140, U.S.A.
Scheme 1
Abstract:
The preparation of Elanco Animal Health immunopotentiator
(S)-ethyl-1-(2-thiopheneacetyl)-3-piperidinecarboxylate (1) is
described. The synthesis includes a new resolution of racemic
ethyl nipecotate with dibenzoyl-L-tartaric acid. The resolved
salt is found to couple directly with commercially available
2-thiopheneacetyl chloride under environmentally friendly
Schotten-Baumann conditions to afford the amide in high yield.
The final product is an oil which is purified by wiped film
evaporative distillation.
racemic ethyl nipecotate [(()-3], available for <$100/kg in
metric ton quantities.4 Initially, a literature procedure was
used which involved three recrystallizations of the diaster-
eomeric salts derived from L-tartaric acid and [(()-3] to
effect (R)-isomer removal.5 The combined mother liquors
were then treated with base to provide (S)-enriched 3 which
was reacted with D-tartaric acid to afford, after two crystal-
lizations, 3‚D-tartaric acid as a 1:1 complex in 99% de and
32% overall yield.6,7 Attempts to obtain the S-isomer more
directly from the racemate upon fractional crystallization of
the diastereomeric salts with D-tartaric acid were met with
moderate success. Best results were obtained using 95%
aqueous ethanol as solvent and 1 equiv of D-tartaric acid
which gave, after seeding and one recrystallization, 3‚D-
tartaric acid in 94% de and 32% yield.6,7 Concurrent with
these efforts, a screen of resolving agents was undertaken
in an attempt to find a more expedient means of accessing
high enantiopurity 3 (g98% ee). An examination of several
standard resolving agents at 100-mg scale revealed diben-
zoyl-tartaric acid and mandelic acid to be the most promising
acids with respect to crystal production in reasonable yield
(ca. 30-40%).7 Upon scale-up of these hits to 5 g of (()-3,
it was found that dibenzoyl-L-tartaric acid and (S)-mandelic
acid provided diastereomeric salts enriched in the desired
(S)-isomer of ethyl nipecotate (3). Using 10 volumes of
refluxing 95% aqueous ethanol and 0.5 equiv of dibenzoyl-
L-tartaric acid followed by slow cooling to room temperature
resulted in precipitation of the corresponding salt in 38%
yield.7 The salt was shown by 500 MHz 1H NMR spectros-
copy to be the 2:1 complex 4. Treatment of 4 with base
followed by Mosher amide formation and analysis (GC,
HPLC) indicated the diastereomeric excess to be g98%.
Similarly, (S)-mandelate salt 5 was produced in 32% yield
with a de of 94% using 8 volumes of ethyl acetate as solvent
Introduction
The Elanco Animal Health1 compound (S)-ethyl-1-(2-
thiopheneacetyl)-3-piperidinecarboxylate (1) is an investi-
gational immunopotentiating agent targeting direct stimula-
tion of immune cells, useful for the prevention of infectious
diseases in livestock.2 Promising in vivo studies using 1-day-
old broiler chicks as a model species warranted process
research involvement to meet increasing material demand
and enable further field and toxicological study. Strategically,
the preparation of 1 is straightforward where the key bond
disconnection involves the amide linkage connecting the
2-thiopheneacetic acid unit 2 with (S)-ethyl nipecotate (3)
(Scheme 1). Key requirements for successfully meeting large
material requests were the availability of the necessary raw
materials, a classical resolution of racemic 3, efficiency of
the amide bond formation, and since 1 is an oil (mp -0.5
°C by DSC), an economical and simple purification method
to give high-quality product.
Results and Discussion
Classical Resolution of Ethyl Nipecotate. The first-
generation synthesis of 12 utilized commercially available
3. However, the only supplier at the time of this project was
the Italian company, Chemie S.p.A, who were able to supply
3 on an as-needed basis at a price of $7500 per 500 grams.3
The projected pricing for ton quantities was significantly
reduced at $1200/kg, yet remained prohibitive, particularly
for use in the preparation of an animal health product. We
thus turned our attention toward the classical resolution of
(4) At the time of this work, the French company Interor offered racemic ethyl
nipecotate through Interchem Corporation, U.S.A.
* To whom correspondence should be addressed. E-mail: emoher@lilly.com.
† Elanco Animal Health Research and Development.
(1) Elanco Animal Health is a division of Eli Lilly and Company.
(2) Creemer, L. C.; Herring, J. R.; McGruder, E. D. U.S. Patent 6,664,271,
2003.
(5) (a) Zheng, X.; Day, C.; Gollamudi, R. Chirality 1995, 7, 90. (b) Akkerman,
A. M.; De Jongh, D. K.; Veldstra, H. Recl. TraV. Chim. Pays-Bas. 1951,
70, 899.
(6) The enantiomeric purity of the free base was assessed by GC and/or HPLC
analysis of the derived Mosher amides (see Experimental Section).
(7) The resolution yields disclosed herein are calculated on the basis of 100%
theoretical yield.
(3) Aldrich currently offers both enantiomers of ethyl nipecotate for ap-
proximately $90/g.
10.1021/op049941c CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/05/2004
Vol. 8, No. 4, 2004 / Organic Process Research & Development
•
593

Scheme 3
and 1 equiv of (S)-mandelic acid. A single recrystallization
from 8 volumes of ethyl acetate provided 5 in 27% overall
yield and g99% de.6,7 Due to the relative efficiency,
simplicity, and high diastereomeric purity with which 4 was
produced, we chose the dibenzoyl-L-tartaric acid resolution
method for further scale-up and use in the preparation of 1.8
simple, we sought to use salt 4 directly in the coupling
reaction and at the same time avoid the use of methylene
chloride. Most gratifyingly, environmentally friendly Schot-
ten-Baumann conditions gave excellent results. Specifically
(Scheme 3), the dibenzoyl-L-tartrate salt 4 was reacted with
15% aqueous sodium carbonate followed by addition of an
ethyl acetate solution of 2-thiopheneacetyl chloride9 (2, X
) Cl) which provided 1 as a colorless to off-white oil in
near quantitative yield and high purity after silica gel
chromatography. The use of sodium hydroxide in place of
sodium carbonate as the base in the amidation reaction
resulted in yields depressed by as much as 10-12%,
primarily due to competitive acid chloride hydrolysis. The
carbonate method followed by chromatography routinely
produced material of greater than 99% enantiomeric and
chemical purity.10 Chromatography was introduced at this
early point primarily to remove color and trace impurities
which were evident by TLC.11 Obviously, it was desirable
to dispense with relying on a chromatographic purification
of 1 should the project progress further in development. An
attractive option was found in wiped film evaporation (WFE)
using a Pope Scientific 2-in. wiped film still. Crude material
from the amidation reaction was typically distilled at ∼160
°C and 0.06-0.09 Torr to provide 1 as a faint-yellow oil
with 90% recovery and 98% HPLC purity.
Enantiomer/Mother Liquor Recycle. With a suitable
method in hand for attaining 3 via a classical resolution we
examined the feasibility of recycling the (R)-enantiomer-
enriched mother liquor waste stream of the resolution
process. Accordingly, the mother liquor and filtrates were
concentrated in vacuo to generate (R)-enriched salt 6
(Scheme 2). Treatment with aqueous sodium carbonate
Scheme 2
followed by extractions with methyl tert-butyl ether and
concentration provided (R)-enriched ethyl nipecotate 7.
Dissolution of 7 in ethanol and treatment with catalytic
sodium ethoxide at reflux for 1 h effected complete racem-
ization. Neutralization with concentrated hydrochloric acid,
removal of sodium chloride by filtration, and treatment of
the filtrate with dibenzoyl-L-tartaric acid as conducted
previously provided 4 in 35% yield and g97% de.6,7 It was
observed that neutralization of the sodium ethoxide used for
the epimerization with hydrochloric acid was necessary since
neutralization with excess dibenzoyl-L-tartaric acid led to
inferior quality (tacky) 3 in low yields due to interference
from sodium dibenzoyl-L-tartrate. While not optimized for
yield and processing efficiency, the recycle of enantiomeric
3 is indeed possible, should it prove necessary, and cost-
effective relative to discarding the resolution filtrates and
purchasing fresh racemate.
Amide Formation and Purification. With an efficient
procedure for the resolution of ethyl nipecotate in hand our
attention was turned toward fashioning the amide linkage in
1. The first-generation procedure2 employed commercially
available 2-thiopheneacetyl chloride 2 (X ) Cl) along with
purchased 3 and i-Pr2NEt in methylene chloride followed
by extractive workup and Kugelrohr distillation to provide
1 as a light-yellow oil in 69% yield. While overall this direct
coupling between an acid chloride and an amine was quite
Conclusions
A simple and efficient two-step preparation of 1 has been
developed. The synthesis features a new method for the
resolution of ethyl nipecotate [(()-3], an enantiomer recycle,
(9) For the present study, 2-thiopheneacetyl chloride was prepared from
2-thiopheneacetic acid and thionyl chloride and purified by distillation
according to the following reference: Miller, W. H.; Dessert, A. M.;
Anderson, G. W. J. Am. Chem. Soc. 1948, 70, 500. Subsequently, Clariant
was identified as a potential bulk supplier of the acid chloride (ca. $100/
kg).
(10) (a) The enantiomeric purity of 1 was determined by chiral capillary
electrophoresis. (b) The chemical purity of 1 was determined by HPLC
and/or GC analysis (see Experimental Section). The largest impurity, apart
from enantiomer, in typical lots of 1 has been identified as 3-thiopheneacetyl
isomer i due to trace isomer contamination of purchased 2-thiopheneacetic
acid. The contamination level is consistently less than 0.5%.
(11) These impurities were identified as aqueous carbonate induced hydrolysis
product ii and ethyl 2-thiopheneacetate (iii). The origin of iii has not been
rigorously determined. It is suspected to arise from reaction of 2 (X ) Cl)
with residual ethanol present in 4.
(8) After the completion of this work, a report was published in this journal
describing the efficient resolution of ethyl nipecotate with D-tartaric acid
using aqueous 2-propanol as solvent. See: Cohen, J. H.; Bos, M. E.; Cesco-
Cancian, S.; Harris, B. D.; Hortenstine, J. T.; Justus, M.; Maryanoff, C. A.;
Mills, J.; Muller, S.; Roessler, A.; Scott, L.; Sorgi, K. L.; Villani, F. J., Jr.;
Webster, R. R. H.; Weh, C. Org. Process Res. DeV. 2003, 7, 866.
594
•
Vol. 8, No. 4, 2004 / Organic Process Research & Development

and a high-yielding Schotten-Baumann coupling. API
purification by WFE was shown to be a feasible and
attractive option with long-term potential for commercial
operations. A 34% yield was achieved for the overall process
from readily available starting materials, not including
enantiomer recycle.
1269 (s), 1196 (s), 1121 (s), 715 (s) cm-1; 500 MHz 1H NMR
(D2O) δ 7.96 (d, 4H, J ) 8.0 Hz), 7.55 (t, 2H, J ) 7.5 Hz),
7.42-7.37 (m, 4H), 5.55 (s, 2H), 4.09-3.97 (m, 4H), 3.27
(dd, 2H, J ) 13, 3.5 Hz), 3.11-3.00 (m, 4H), 2.88-2.80
(m, 2H), 2.77-2.63 (m, 2H), 1.94-1.90 (m, 2H), 1.72-
1.52 (m, 6H), 1.08 (t, 6H, J ) 7.3 Hz); 125 MHz 13C NMR
(D2O) δ 174.63, 173.58, 168.44, 134.66, 130.43, 129.52,
129.36, 75.87, 63.01, 44.71, 44.42, 38.77, 25.03, 21.23,
13.82. Anal. calcd. For C34H44N2O12: C, 60.69; H, 6.59; N,
4.18. Found: C, 60.58; H, 6.66; N, 4.27.
Experimental Section
1
General. H and 13C NMR spectra were recorded on a
Bruker ARX-500 spectrometer or a Varian MercuryVx 400
MHz spectrometer as noted. Diastereomeric purities (% de)
of the diastereomeric salts of ethyl nipecotate were deter-
mined by Mosher amide analysis using capillary gas chro-
matography on a Hewlett-Packard model 5890 instrument
equipped with a 25 m × 0.25 mm 30M DB-1 column (initial
temperature 60 °C ramped to 300 °C over 13 min, injector
temperature 250 °C) with FID detection at 250 °C and by
HPLC analysis on a Shimadzu SCL-10A instrument equipped
with a 4.6 mm × 250 mm Zorbax SB-Phenyl column with
gradient elution (1 mL/min, acetonitrile-water both with
0.5% TFA, 60% acetonitrile ramped to 90% over 25 min)
and detection at 220 nm. Absolute configuration of resolved
ethyl nipecotate was determined through Mosher amide
formation and comparison of retention times (GC and HPLC)
with Mosher amides derived from authentic 3 from Chemie
S.p.A. and (R)-3 obtained according to ref 5a. Analysis of
the chemical purity of 1 was conducted via the above GC
method as well as by HPLC using a 4.6 mm × 150 mm
Zorbax SB-C18 column with gradient elution (90% water
with 0.1% TFA/10% acetonitrile ramped to 80% acetonitrile/
20% water with 0.1% TFA over 70 min) at a 1 mL/min flow
rate and detection at 250 nm. Optical rotations were measured
with a Perkin-Elmer model 241 polarimeter. Melting points
were recorded on a Gallenkamp melting point apparatus and
are uncorrected. Elemental analyses, Fourier transform
infrared (FTIR) spectra, and mass spectra were performed
at the Structural and Organic Chemistry Research Laboratory,
Eli Lilly and Company, Indianapolis, Indiana, U.S.A.
Dibenzoyl-L-tartrate Salt 4. To a three-neck, 5-L flask
equipped with a heating mantle, mechanical stirrer, a
temperature probe, and a reflux condenser topped with a
calcium carbonate drying tube was charged 250.8 g (1.60
mol) of racemic ethyl nipecotate [(()-3] followed by 2100
mL of 95% aqueous 2B-ethanol (denatured with toluene).
To the solution was added 286.7 g (0.80 mol) of dibenzoyl-
L-tartaric acid as a solid resulting in a temperature rise to 37
°C. Residual resolving agent was rinsed into the reaction
mixture with 400 mL of 95% aqueous ethanol. The mixture
was heated to 78 °C, effecting complete dissolution. The
heat was turned off, and the clear yellow solution was
allowed to gradually cool to 68 °C at which time seed was
added followed by cooling slowly to room temperature and
stirring for a total of 15 h after seeding (crystal formation
was observed at 66 °C). The white precipitate was collected
and washed with ethanol (1 × 400 mL) followed by vacuum-
drying at 45-50 °C to provide 208 g (39%) of the title
compound as a white solid: mp 173-175 °C; [R]D -61 (c
1.22, MeOH), 98% de; FTIR (KBr) 3428 (m), 2996 (m),
2854 (m), 2317 (w), 1721 (s), 1623 (s), 1454 (s), 1383 (s),
Epimerization/Resolution/Recycle. To ca. 100 g (wet)
of concentrated resolution filtrates and mother liquors was
added tert-butyl methyl ether (500 mL) followed by 500 mL
of 15% sodium carbonate. After stirring mechanically for 1
h to effect dissolution, the layers were separated, and the
aqueous layer was extracted with tert-butyl methyl ether (1
× 500 mL, 2 × 250 mL). The combined organics were dried
(Na2SO4), filtered, and concentrated in vacuo to give 27.0 g
of (R)-enriched ethyl nipecotate as a yellow oil. To a solution
of the above oil (10 g, 63.5 mmol) in 58 mL of 2B-ethanol
(denatured with toluene) was added 5.9 mL (15.9 mmol) of
a 21 wt % solution of sodium ethoxide in ethanol. After
heating at reflux for 1 h, the mixture was cooled to room
temperature and treated with concentrated HCl (1.31 mL,
15.9 mmol) and filtered through Celite washing with 36 mL
of 2B-ethanol (denatured with toluene). The combined filtrate
and washings were treated with 11.29 g (31.7 mmol) of
dibenzoyl-L-tartaric acid, and the mixture was heated to
reflux. Addition of 2 mL of water effected complete
dissolution, after which time the mixture was allowed to cool
slowly. After seeding at 64 °C the mixture was allowed to
cool further to room temperature and to stir a total of 16 h.
The crystals were collected and dried in vacuo at 45-50 °C
to yield 7.30 g (35%) of 4 as a white solid (g97% de).
(S)-Ethyl-1-(2-thiopheneacetyl)-3-piperidinecarboxy-
late (1). To a three-neck, 5-L flask equipped with a
mechanical stirrer and a dropping funnel was charged 182 g
(271 mmol) of tartrate salt 4 followed by 590 mL of ethyl
acetate. To the rapidly stirring slurry was added 715 mL of
water followed by 482 mL (ca. 2.5 equiv) of 15% sodium
carbonate over 15 min via dropping funnel. After stirring
briefly, acid chloride 2 (X ) Cl)9 (91.3 g, 568 mmol) was
added as a solution in 111 mL of ethyl acetate over 10 min
via a dropping funnel (mild gas evolution observed). Upon
completion of the addition, the mixture was allowed to stir
for 30 min at which time the layers were separated, and the
aqueous layer was washed with ethyl acetate (2 × 350 mL).
The combined organics were dried (Na2SO4), filtered, and
concentrated in vacuo to a yellow oil. Chromatography (1200
g of flash SiO2, 1:1 then 1:3 hexanes/ethyl acetate) provided
150 g (99%) of the title compound as a faintly off-white
oil: [R]22D +63 (c 10.8, MeOH); 400 MHz 1H NMR (CD3-
OD; mixture of two rotamers) δ 1.24 (dt, J ) 5.37, 7.33
Hz, 3H), 1.28-1.40 (m, 0.5H), 1.40-1.51 (m, 0.5H), 1.61-
1.73 (m, 1.5H), 1.73-1.84 (m, 0.5H), 1.93-2.03 (m, 1H),
2.31-2.39 (m, 0.5H), 2.39-2.47 (m, 0.5H), 3.12 (dd, J )
13.19, 9.77 Hz, 0.5H), 3.15-3.26 (m, 1.5H), 3.52 (dd, J )
13.68, 8.3 Hz, 0.5H), 3.80-4.02 (m, 3H), 4.09-4.18 (m,
Vol. 8, No. 4, 2004 / Organic Process Research & Development
•
595

2H), 4.34 (dm, J ) 12.7 Hz, 0.5H), 6.91-6.97 (m, 2H), 7.28
(dt, J ) 5.37, 1.46 Hz, 1H); 100 MHz 13C NMR (CDCl3;
mixture of two rotamers) δ 172.98, 168.70, 136.82, 127.03,
126.30, 124.97, 61.03, 60.91, 48.19, 46.91, 44.11, 42.60,
41.68, 41.35, 35.51, 35.36, 27.35, 25.01, 24.01, 14.42; FTIR
(CHCl3) cm-1 3019 (br, m), 1725 (s), 1636 (s), 1455 (m),
1179 (m); ES-MS (Positive) 585 (dimer + Na+), 304 (M +
Na+), 282 (MH+); UV (EtOH) λmax nm 234; Anal. calcd.
For C14H19NO3S: C, 59.76; H, 6.81; N, 4.98; S, 11.39.
Found: C, 59.95; H 6.82; N, 5.10; S, 11.26.
Mosher Amide Formation for Enantiopurity Analysis.
General Procedure. To a slurry of 55 mg (0.082 mmol) of
salt 4 in 1 mL of tert-butyl methyl ether was added 200 µL
of 15% sodium carbonate. After stirring to dissolve, the
layers were separated, and the aqueous layer was washed
with tert-butyl methyl ether (2 × 1 mL). The combined
organics were dried (Na2SO4), filtered, and concentrated to
give 22 mg (0.14 mmol, 86%) of 3 as a colorless oil. The
free base was dissolved in 220 µL of CH2Cl2 and treated
sequentially with 66 mg (0.28 mmol) of (R)-(+)-R-methoxy-
R-(trifluoromethyl)phenylacetic acid [(R)-Mosher acid], 2 mg
(0.02 mmol) of 4-dimethylamino pyridine, and 58 mg (0.28
mmol) of 1,3-dicyclohexylcarbodiimide as a solution in 60
µL of CH2Cl2. After stirring for 30 min the cloudy, white
reaction was filtered through Celite, and the filtrate was
analyzed by GC and HPLC as indicated above in the General
Experimental section.
Acknowledgment
We kindly thank Dr. Robin Readnour, Ms. Carolyn M.
Stobba-Wiley, Ms. Christine Cleveland, and Mr. Douglas
Kiehl for generous analytical support. Mr. Robert Waggoner
is gratefully acknowledged for contributions in raw material
sourcing.
Received for review March 18, 2004.
OP049941C
596
•
Vol. 8, No. 4, 2004 / Organic Process Research & Development