Development of a scalable synthesis of dipeptidyl peptidase-4 inhibitor ABT-279

Article abstract of DOI:10.1021/op900197r

A convergent, scalable synthesis of dipeptidyl peptidase-4 inhibitor, ABT-279, has been developed and demonstrated on multikilogram scale. The cis-2,5-disubstituted pyrrolidine is generated by cyclization of a Boc-amine onto an alkynyl ketone followed by stereospecific reduction of the resulting acyliminium intermediate. The amine coupling partner was prepared by a novel Hofmann rearrangement promoted by 1,3-dibromo-5,5-dimethylhydantoin. The final product was isolated as the L-malic acid salt. The scaleup campaign consisted of 15 steps and delivered 42 kg of ABT-279 in 14% overall yield. A second-generation synthesis that addresses some of the issues encountered during scale-up was developed and demonstrated on kilogram scale.

Full text of DOI:10.1021/op900197r

Organic Process Research & Development 2009, 13, 1145–1155
Development of a Scalable Synthesis of Dipeptidyl Peptidase-4 Inhibitor ABT-279
Todd S. McDermott,*,† Lakshmi Bhagavatula,‡ Thomas B. Borchardt,§ Kenneth M. Engstrom,‡ Jorge Gandarilla,‡
Brian J. Kotecki,‡ Albert W. Kruger,‡ Michael J. Rozema,‡ Ahmad Y. Sheikh,§ Seble H. Wagaw,‡ and Steven J. Wittenberger‡
Abbott Laboratories, Global Pharmaceuticals R&D, Formulation Sciences - Solids, 200 Abbott Park Road, D-R4P7, Building
AP31, Abbott Park, Illinois 60064, Abbott Laboratories, Global Pharmaceuticals R&D, Process R&D, 1401 Sheridan Road,
D-R450, Building R8, North Chicago, Illinois 60064, and Abbott Laboratories, Global Pharmaceuticals R&D, Solid State
Chemistry, 1401 Sheridan Road, D-R452, Building R13, North Chicago, Illinois 60064
Abstract:
A convergent, scalable synthesis of dipeptidyl peptidase-4 inhibitor,
ABT-279, has been developed and demonstrated on multikilogram
scale. The cis-2,5-disubstituted pyrrolidine is generated by cycliza-
tion of a Boc-amine onto an alkynyl ketone followed by stereospe-
cific reduction of the resulting acyliminium intermediate. The
amine coupling partner was prepared by a novel Hofmann
rearrangement promoted by 1,3-dibromo-5,5-dimethylhydantoin.
Figure 1. DPP-4 Inhibitor ABT-279 (1).
Scheme 1. Retrosynthesis of ABT-279 (1)
The final product was isolated as the L-malic acid salt. The scale-
up campaign consisted of 15 steps and delivered 42 kg of ABT-
279 in 14% overall yield. A second-generation synthesis that
addresses some of the issues encountered during scale-up was
developed and demonstrated on kilogram scale.
Introduction
Diabetes is a disease that affects 170 million people
worldwide. The prevalence of diabetes is expected to more than
double by 2030.1 Inhibition of dipeptidyl peptidase-4 (DPP-4)
has been shown to be effective at lowering levels in the blood
of glucose and hemoglobin A1c (HbA1c) and has the potential
to be disease altering by promoting regeneration of insulin-
producing cells in the pancreas.2 As part of an active program
in the area of DPP-4 inhibition, multikilogram quantities of
DPP-4 inhibitor ABT-279 (1), Figure 1, were necessary to
support regulatory toxicology studies and clinical evaluation.3
The retrosynthetic analysis of 1 leads to aminopyridine 2
and chloroacetamide 3 (Scheme 1). Aminopyridine 2 is
ultimately derived from commercially available ethyl isonipe-
cotate 4 and tert-butyl chloroisonicotinate 5, while chloroac-
etamide 3 comes from methyl N-Boc-pyroglutamate 6.
Scheme 2. Synthesis of amide 9a
a Reaction conditions: (a) Boc2O, tol, RT. (b) LDA; MeI, THF/tol, -10 °C.
(c) NaNH2, THF/tol, 50 °C. (d) p-TsOH, IPA, 80 °C; 69% overall yield.
Discussion
Synthesis of Amine Coupling Partner. Commercially
available ethyl isonipecotate, 4, was converted to salt 9·TsOH
by standard transformations (Scheme 2). Boc protection of 4
in toluene was followed by deprotonation with LDA in toluene/
THF at -10 °C and then addition of methyl iodide to give ester
7. After workup, the ester was converted to the corresponding
amide 8 by treatment with sodium amide in toluene/THF at 50
°C. Finally, the Boc group was removed, and the p-toluene-
sulfonic acid salt, 9·TsOH, was isolated as a crystalline solid.
The overall yield for the four-step sequence was 69%.
The pyridine-coupling partner necessary to access aminopy-
ridine 2 is tert-butyl-2-chloroisonicotinate, 5. Commercially
available 2-chloroisonicotinic acid, 10, was protected as its tert-
butyl ester by treatment with excess Boc2O and catalytic DMAP
in N-methyl-2-pyrrolidone (NMP) solvent (Scheme 3). Attempts
* Author to whom correspondence may be sent. E-mail: todd.mcdermott@
abbott.com.
† Formulation Sciences - Solids.
‡ Process R&D.
§ Solid State Chemistry.
(1) Wild, S.; Roglic, G.; Green, A.; Sicree, R.; King, H. Diabetes Care
2004, 27, 1047.
(2) Zerilli, T.; Pyon, E. Y. Clin. Ther. 2007, 29, 2614.
(3) Madar, D. J.; Kopecka, H.; Pireh, D.; Yong, H.; Pei, Z.; Li, X.;
Wiedeman, P. E.; Djuric, S. W.; Von Geldern, T. W.; Fickes, M. G.;
Bhagavatula, L.; McDermott, T.; Wittenberger, S.; Richards, S. J.;
Longenecker, K. L.; Stewart, K. D.; Lubben, T. H.; Ballaron, S. J.;
Stashko, M. A.; Long, M. A.; Wells, H.; Zinker, B. A.; Mika, A. K.;
Beno, D. W. A.; Kempf-Grote, A. J.; Polakowski, J.; Segreti, J.;
Reinhart, G. A.; Fryer, R. M.; Sham, H. L.; Trevillyan, J. M. J. Med.
Chem. 2006, 49, 6416.
10.1021/op900197r CCC: $40.75  2009 American Chemical Society
Published on Web 10/15/2009
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1145

Scheme 3. Synthesis of aminopyridine coupling partner 2
Figure 2. Major impurities formed during the synthesis of
to effect the tert-butyl ester formation under more standard
conditions4 resulted in low yields due to incomplete reactions.
After extraction with methyl tert-butyl ether (MTBE) and
concentration from toluene, compound 5 was obtained as an
oil containing toluene and di-tert-butyl carbonate, which do not
interfere with subsequent reactions.5
ABT-279.
2). In order to increase the rate of isocyanate hydrolysis, the
reaction was carried out in a biphasic system of THF/4 M
NaOH in the presence of a phase transfer catalyst. The use of
1 equiv of tetrabutylammonium bromide (TBAB) limited the
formation of urea while not greatly affecting the ester hydrolysis.
The amount of TBAB could be reduced to about 0.4 equiv
without an increase in urea formation, but the slower reaction
time resulted in additional ester hydrolysis. During the prepara-
tion for scale up, it was noted that the Hofmann reaction is
very sensitive to temperature. It was critical to keep the reaction
temperature below 5 °C during the addition of DBDMH, which
is very exothermic. On small scale, increasing the reaction
temperature from 5 to 10 °C resulted in a 4-fold increase in the
amount of urea 13 formed. During the pilot plant campaign,
the reaction temperature was controlled by limiting the rate of
addition of DBDMH and the reaction resulted in 87% yield of
the desired amine (Scheme 3).
Synthesis of Chloroacetamide Coupling Partner. The
synthesis of the other coupling partner, chloroacetamide 3, began
with chemistry reported by Martin et al. in 2004.9 They reported
opening of carbamate-protected pyroglutamates followed by
cyclization and reduction by treatment with BF3 ·OEt2 in the
presence of triphenylsilane to give 2,5-cis-pyrrolidines. In
the case of the isopropyl carbamate, the yield and selectivity
were excellent. In the case of Boc-protected pyroglutamates,
the protecting group was unstable to the Lewis acidic conditions,
and alternative cyclization/reduction conditions were employed.
The Martin chemistry provided a good starting point for the
synthesis of chloroacetamide 3. For our purposes, the Boc
protecting group was preferable to the isopropyl carbamate
because of our previous experience with Boc-protected inter-
mediates. The addition of the magnesium chloride salt of
trimethylsilylacetylene to pyroglutamate 6 gave clean conversion
to propargylic ketone 14 (Scheme 4). Although the ester was
added to an excess of Grignard reagent, very little bis-addition
product was detected, suggesting that the initially formed
tetrahedral intermediate is stabilized by the presence of the
adjacent Boc group. The major byproduct encountered was
ketone 15, the result of base-promoted self-condensation of 6
(Figure 2). This product could be avoided by slow addition of
a solution of the ester to a cold THF solution of the Grignard
The strategy for coupling chloropyridine 5 and various
substituted piperidines was initially focused on palladium-
catalyzed N-arylation reactions.6 Sensitivity of the tert-butyl ester
of 5 to the highly basic conditions necessary for these coupling
reactions led to low yields and complicated reaction mixtures.
The major impurities were related to hydrolysis of the tert-
butyl ester of starting material and product. An alternative
thermal coupling strategy was developed. Treatment of chlo-
ropyridine 5 with mild base and 1.5 equiv of 9·TsOH at 100
°C in dimethyl sulfoxide (DMSO) for 36 to 48 h led to amide
11 (Scheme 3). The major side reaction was cleavage of the
tert-butyl ester of the starting chloropyridine. Upon completion,
addition of water to the reaction mixture resulted in crystal-
lization of amide 11. Water washes of the solid completely
removed the chloroisonicotinic acid byproduct resulting in pure
material (>98 LC A%).
The final transformation to produce amine 2 was a Hofmann
rearrangement of amide 11. This transformation could be carried
out under typical Hofmann rearrangement conditions; however,
low yields or hard-to-control reactions resulted.7 An alternative
Hofmann reaction was developed, utilizing 2,5-dibromo-3,3-
dimethylhydantoin (DBDMH) as the bromine source (Scheme
3).8
Initially, the Hofmann reaction was hampered by formation
of acid 12, by hydrolysis of the ester, and urea 13 due to buildup
of product in the presence of the intermediate isocyanate (Figure
(4) (a) Baldwin, J. E.; Bamford, S. J.; Fryer, A. M.; Rudolph, M. P. W.;
Wood, M. E. Tetrahedron 1997, 53, 5233. (b) Wright, S. W.;
Hageman, D. L.; Wright, A. S.; McClure, L. D. Tetrahedron Lett.
1997, 38, 7345. (c) Dhaon, M. K.; Olsen, R. K.; Ramasamy, K. J.
Org. Chem. 1982, 47, 1962.
(5) During the pilot-plant campaign, distillation of the MTBE/toluene
product solution resulted in a higher than expected concentration, and
compound 5 crystallized in the reactor. A crystallization procedure
from MeOH/water was subsequently developed and demonstrated on
1-kg scale.
(6) Wagaw, S.; Buchwald, S. L. J. Org. Chem. 1996, 61, 7240.
(7) The use of NaOCl as oxidant led to incomplete reaction, whereas the
use of 2,5-dichloro-3,3-dimethylhydantoin led to loss of the tert-butyl
ester group due to long reaction times. The used of bromine as oxidant
was possible, but addition rate and stoichiometry were difficult to
control.
(8) Engstrom, K.; Henry, R.; Hollis, S.; Kotecki, B.; Marsden, I.; Pu, Y.-
M.; Wagaw, S.; Wang, W. J. Org. Chem. 2006, 71, 5369–5372.
(9) Brenneman, J. B.; Machauer, R.; Martin, S. F. Tetrahedron 2004, 60,
7301.
1146
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development

Scheme 4. Initial cyclization/reduction sequence
Table 1. Cyclization/reduction of 18
entry
solvent
reductant
cis/trans
1
2
3
4
5
6
7
toluene
Et3SiH
1:8
DCM
Et3SiH
1:7
acetonitrile
DCM
toluene
acetonitrile
EtOAc
Et3SiH
1:4
1.5:1
2:1
5.1:1
11:1
NaBH(OAc)3
NaBH(OAc)3
NaBH(OAc)3
NaBH(OAc)3
Scheme 6. Synthesis of Boc-acid 19a
Scheme 5. Synthesis of enyne 18
a Reaction conditions: (a) NaBH(OAc)3; TFA, IPAc, -10 °C. (b) LiOH ·H2O,
THF/EtOH, 0 °C, crystallization from IPAc/heptane, (65%).
due to the lability of the TMS group. However, it was not
necessary to isolate the enyne intermediate. Ketone 14, when
exposed to TFA in the presence of NaBH(OAc)3, gave the
desired pyrrolidine (17) in good yield and selectivity (Scheme
6).11 In practice, a mixture of 14 and 1.5 equiv of NaBH(OAc)3
in IPAc was treated with 3 equiv of TFA at -10 °C. Upon
warming to room temperature the reaction was complete.12 The
crude product was solvent switched to EtOH and treated with
LiOH to hydrolyze the methyl ester and the TMS group. On
pilot-plant scale the cyclization, reduction, and hydrolysis
sequence gave a 12:1 cis/trans ratio. Crystallization of acid 19
from IPAc/heptane resulted in material that contained less than
0.5% of the trans isomer, but substantial losses to the mother
liquors occurred (about 19%). The overall yield of the process
was 65% on 50-kg scale (Scheme 6), and the material produced
was greater than 97% pure by HPLC analysis.
Carboxylic acid 19 was converted to the corresponding
amide and then dehydrated to nitrile 20 under Vilsmeier
conditions (Scheme 7). The Boc group was removed with
p-TsOH. One of the side products of this reaction was the
pyrrolidine tert-butyl amide 21 (Figure 2), resulting from attack
of the nitrile on the tert-butyl cation produced during the Boc
deprotection. The use of CH3CN as solvent provided an excess
of an alternative nitrile and resulted in formation of significantly
lower amounts of 21 (2-4 LC A%) and the production of 1
equiv of tert-butylacetamide as a byproduct, which does not
interfere with the next reaction. After removal of the Boc
protecting group, the chloroacetamide (3) was formed directly
reagent. This reaction gave 85% yield during the pilot plant
campaign and the product, which was about 95 LC A% pure
(excluding IPAc), was used without purification after extractive
workup.
Initial attempts to effect cyclization/reduction of 14 were
unsuccessful. Using the conditions reported by Martin, the
compound rapidly decomposed. Changing the solvent to THF
led to cyclization to the corresponding enyne (16), which slowly
reduced to pyrrolidine 17. However, during the long reaction
time degradation of enyne 16 and product was seen (Scheme
4).
In order to work with a more stable system, the triisopro-
pylsilyl (TIPS)-protected enyne (18) was produced by addition
of tri-isopropylalkynyl lithium to 6 and treatment with BF3 ·OEt2
followed by silica gel chromatography (Scheme 5). Enyne 18
is a stable material isolated in 45% yield and allowed for the
controlled investigation of the desired reduction.
The desired reduction could be achieved by treatment of 18
with trifluoroacetic acid (TFA) in the presence of reducing
agents. Indeed, 18 could be selectively reduced to either the
trans- or cis-pyrrolidine, depending on the conditions. Use of
Et3SiH as the reducing agent led preferentially to the trans
product (Table 1, entries 1-3). Reduction with a bulkier reagent,
sodium triacetoxyborohydride (NaBH(OAc)3), led preferentially
to the cis-pyrrolidine product (Table 1, entries 4-7). For both
reducing agents, the use of polar solvents (IPA or THF) led to
removal of the Boc group and decomposition. The highest cis-
selectivity was observed when EtOAc was used as the solvent.10
Although the above reduction produced the desired cis-
pyrrolidine product, it was preferable to use the TMS-protected
acetylene, rather than the harder to remove TIPS protecting
group. Unfortunately, isolation of enyne 16 was not possible
(11) For the use of TFA to convert a Boc-protected amino-ketone to a
Boc-enamine see: Clark, R. D.; Muchowski, J. M.; Fisher, L. E.;
Flippin, L. A.; Repke, D. B.; Souchet, M. Synthesis 1991, 871.
(12) Other protic acids (HCl, H2SO4, MeSO3H, H3PO4) and TMSCl failed
to effect the cyclization/reduction. The selectivity of the reaction is
independent of temperature (12:1 at 0 or 25 °C), although slightly
higher yields are obtained at 0 °C. IPAc and EtOAc are comparable
solvents for the reduction reaction. IPAc was used to avoid issues
with EtOAc hydrolysis in the subsequent step.
(10) An investigation of the scope and limitations of the cyclization/
reduction reaction is underway and will be reported separately.
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1147

Scheme 7. Synthesis of chloroacetamide 3a
Scheme 9. Alternative Hofmann and amine coupling
sequencea
a Reaction conditions: (a) i-BuOCOCl; NH4OH, IPAc, 0 °C. (b) SOCl2/DMF,
THF, rt, (87%). (c) TsOH, CH3CN RT, (d) ClCH2COCl, Hunig’s base, CH3CN,
0 °C; crystallization from IPA/water, (78%).
Scheme 8. Endgame strategya
a Reaction conditions: (a) DBDMH, aq KOH/CH3CN, 0 °C. (b) p-TsOH,
MeOH, 60 °C, (92%). (c) Boc2O, DMAP, NMP; Et3N, RT (88%). (d) K3PO4,
NMP 80 °C (97%). (e) H2, Pd/Al2O3, K3PO4, NMP, 40 °C (96%).
the DBDMH Hofmann reaction is very sensitive to reaction
temperature (see Scheme 3). The primary reason for this
sensitivity is that amide 11 has low solubility in most solvents,
making hydrolysis of the intermediate isocyanate slow. As an
alternative, the Hofmann reaction was carried out on amide 8,
a compound with more favorable solubility characteristics
(Scheme 9). Because of the improved solubility of 8, a phase
transfer catalyst was not necessary.13 The resulting Boc amine
was treated with p-TsOH in MeOH at 60 °C to effect cleavage
of the Boc group. The product was isolated as its bis-p-TsOH
salt directly from the reaction mixture. This sequence gave 92%
isolated yield of salt 23 for two steps, which is a substantial
improvement of the previous Hofmann protocol.
An additional improvement to the scale-up campaign was
the use of the more activated tert-butyl 2,6-dichloroisonicotinate
(25) as the pyridine coupling partner. Protection of the com-
mercially available 2,6-dichloroisonicotinic acid (24) was carried
out using conditions similar to those reported earlier for t-butyl
ester formation.14 In this case, the ester (25) was isolated as a
crystalline solid in 88% yield. The N-arylation reaction using
compound 25 is considerably more facile. Treatment of 25 with
a slight excess of 23 in NMP in the presence of base gave a
nearly quantitative yield of chloropyridine 26 in a few hours at
80 °C. Dechlorination of 26 was carried out by treatment with
Pd/Al2O3 in the presence of H2 and base in NMP to give 96%
yield of coupling partner 2 (Scheme 9). After workup, the
product was used in the final coupling reaction without further
purification. During the first pilot-plant campaign the sequence
from amide 8 to amine 2 proceeded in four steps and 66%
overall yield. The alternative sequence proceeded in five steps
and 76% overall yield and gave material that was indistinguish-
able from material produced during the pilot-plant campaign.
This sequence was carried out on 1-kg scale in order to
demonstrate its scalability. In this case, each of the reactions
was worked up as usual (see Experimental Section). However,
the tert-butyl ester formation, the N-arylation, the dechlorination,
and the final coupling (Scheme 9) are each carried out in the
NMP/base system, allowing the possibility of telescoping these
a Reaction conditions: (a) K3PO4/KI, NMP, 35 °C. (b) D-tartaric acid, IPA,
(88%). (c) H3PO4/H2O, 60 °C. (d) L-malic acid, EtOH/H2O, 45 °C (93%).
by treatment with chloroacetyl chloride in the presence of
Hunig’s base. Chloroacetamide 3 was crystallized from IPA/
water to give 78% yield of material that was 99 LC A% pure
and >200:1 cis/trans.
Endgame Strategy. With the coupling partners in place,
all that remained was the final coupling and deprotection. The
coupling of hindered amine 2 required slightly forcing condi-
tions. An NMP solution of chloroacetamide 3 was added to a
mixture of 2, milled K3PO4, and KI in NMP solvent at 35 °C
(Scheme 8). The key concern with this reaction was epimer-
ization of the nitrile stereocenter. At temperatures above 40 °C,
the amount of epimerization was beyond the level that could
be removed during isolation. At 45 °C, the cis/trans ratio was
87:13, and salt formation and crystallization resulted in a final
ratio of 89:11. A reaction temperature of 35 °C gave a good
balance between reaction rate (completion in ∼4 h) and
epimerization (less than 1%). The product, 22, was isolated by
extraction with MTBE followed by crystallization as its
D-tartaric acid salt. The isolation of the penultimate product
allowed very clean material (>99 LC A%) to be taken into the
final deprotection step.
Early attempts at the tert-butyl ester deprotection suffered
from formation of Ritter reaction byproducts, as had been seen
with the Boc deprotection of 20. The strategy of using CH3CN
as a solvent, which had been successful previously, did not
completely alleviate the problem. Carrying out the final depro-
tection with phosphoric acid in water avoided formation of the
tert-butyl amide byproduct. After the deprotection step, ABT-
279 (1) was isolated as its crystalline L-malic acid salt from
EtOH/H2O. The final deprotection/isolation sequence proceeded
in 93% yield and gave 42 kg of product. The material was 99.8
LC A% pure and was 99.4% ee by HPLC analysis. The cis/
trans diastereomeric ratio of the final solid was >99.9:0.1.
Alternative Synthetic Sequence. The synthesis carried out
in the pilot plant produced material to support the project
through phase I and into phase II development. However, there
were several issues with the scale-up synthesis that needed to
be addressed prior to further scale-up work. As described earlier,
(13) Mixtures of 8, CH3CN, and 2 M KOH became homogeneous upon
addition of DBDMH. Running the reaction under dilute conditions,
where layer separation occurred, resulted in formation of the corre-
sponding urea byproduct.
(14) One equivalent of Et3N was added to the reaction to increase the
nucleophilicity of the acid. In the absence of base the reaction exhibits
a long induction period and a hard-to-control exotherm.
1148
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development

reactions. In a small-scale run, these reactions were carried out
without workup of any of the intermediates. Optimization of
this sequence should allow telescoping of the final four steps
of the process.
µL injection, λ 210 nm; Mobile phase A: 0.1% H3PO4; Mobile
phase B: ACN; Gradient: 0 min 45% B, 30 min 50% B, 35
min 80% B, 38 min 80% B, 40 min 45% B, 50 min 45% B;
Retention time (min): N-Boc ethyl isonipecotate (14.2), toluene
(15.5), 7 (21.0), ethyl benzene (21.5). The product was carried
forward as a toluene solution. An analytical sample was
prepared by short-path distillation (1.5 mmHg, collected
Conclusion
We have described a convergent and practical synthesis of
ABT-279, a DPP-4 inhibitor. The synthesis was carried out in
the pilot plant to produce 42 kg of ABT-279. The key
transformations include cyclization of a Boc-protected ami-
noketone, stereospecific reduction of the resulting acyliminium
ion with sodium triacetoxyborohydride, and a novel DBDMH-
promoted Hofmann rearrangement to give the amine coupling
partner. Substantial improvements were made to the initial
synthesis and were demonstrated on 1-kg scale.
1
128-134 °C). H NMR (400 MHz, CDCl3) δ 1.25 (td, J )
7.10, 0.75 Hz, 3 H) 1.45 (d, J ) 0.69 Hz, 9 H) 1.53-1.69 (m,
3 H) 1.81-1.91 (m, 2 H) 2.37-2.47 (m, J ) 11.01, 11.01,
3.95, 3.95 Hz, 1 H) 2.82 (t, J ) 11.87 Hz, 2 H) 3.93-4.07 (m,
1 H) 4.13 (qd, J ) 7.11, 0.75 Hz, 2 H). 13C NMR (101 MHz,
CDCl3) δ 14.54, 28.24, 28.69, 41.35, 43.0 (br), 60.57, 79.53,
154.26, 174.04. IR (KBr) 2977, 1730, 1690 cm-1. Anal. Calcd
for C13H23NO4: C 60.68, H 9.01, N 5.44; found: C 60.58, H
9.13, N 5.46.
Methylation. LDA solution (27.3 wt % active, 189 kg, 482
mol, 1.2 equiv) and THF (190 kg) were chilled to -10 °C.
The solution of starting material, Boc-4, (60.7 wt % active, 169
kg, 399 mol, 1 equiv) was added over 1 h with a maximum
internal temperature of 5 °C (target -10 °C). The addition vessel
was rinsed with THF (25 kg), which was added to the reaction
vessel. The reaction was stirred for about 30 min at an internal
temperature of approximately -10 °C. A mixture of methyl
iodide (71 kg, 500 mol, 1.25 equiv) and THF (35 kg) was added
to the reaction mixture over 6 h at not more than 5 °C. After
30 min, HPLC analysis showed no starting material remained
(see HPLC conditions above). The reaction was quenched into
-5 °C 1 M H3PO4 solution (prepared from 700 kg of water
and 86 of kg H3PO4), while maintaining the temperature <20
°C. While mixing, the temperature was adjusted to RT. The
layers were separated. HPLC analysis showed 105.4 kg (97.4%
yield) in the organic layer. The organic layer was concentrated
several times from toluene to remove water to <0.1 wt % as
determined by Karl Fisher analysis. After filtration of precipi-
tated salts, the filtrate was concentrated to ∼65-70 wt %
product in toluene and used directly in the next step. An
analytical sample was prepared by short-path distillation (0.5
mmHg, collected 124 °C) 1H NMR (400 MHz, CD3OD) δ 1.19
(s, 3 H) 1.25 (t, J ) 7.14 Hz, 3 H) 1.31-1.40 (m, 2 H) 1.44 (s,
9 H) 1.99-2.09 (m, 2 H) 2.88-3.07 (m, 2 H) 3.75 (dt, J )
13.86, 4.12 Hz, 2 H) 4.16 (q, J ) 7.14 Hz, 2 H) 4.82 (m, 4 H).
13C NMR (101 MHz, CD3OD) δ 14.69, 26.34, 28.79, 35.66,
43.0 (br), 42.79, 61.74, 80.84, 155.87, 177.07. IR (KBr) 2974,
1725, 1692 cm-1. Anal. Calcd for C14H25NO4: C 61.97, H 9.29,
N 5.16; found: C 61.81, H 9.45, N 5.14.
Experimental Section
All reactions were carried out under a nitrogen atmosphere
in either glass vessels or glass-lined stainless steel vessels with
overhead mechanical stirring unless otherwise noted. HPLC data
were collected using an Agilent 1100 series HPLC. Gas
chromatography data were collected using a Hewlett-Packard
6890 series gas chromatograph. Chromatographic conditions
are reported as part of the experimental descriptions below.
Retention times are uncorrected. Elemental analysis samples
were run by Quantitative Technologies Inc., Whitehouse, NJ
(U.S.A.). Reagents were purchased from commercial suppliers
and used as received.
Preparation of 1-tert-Butyl 4-ethyl 4-methylpiperidine-
1,4-dicarboxylate (7). Boc protection. A solution of Boc2O (97
kg, 444 mol, 1.03 equiv) and toluene (140 kg, KF 0.16 wt %
water) at 15 °C was treated with ethyl isonipecotate (4, 68 kg,
433 mol, 1 equiv) at <30 °C. The addition vessel and lines were
rinsed with toluene (2 kg), which was added to the reaction
vessel. The reaction was mixed for 15 min, at which time an
in-process GC sample showed the starting ethyl isonipecotate
(4) was consumed. GC conditions: Column: J&W Scientific,
DB-5, 30 m × 0.53 mm, 5 µm, injector: 150 °C, split ratio
1:1, 5.0 mL/min constant pressure, 2.0 µL injection; Oven
program: 110 °C then ramp 4 °C/min to 200 °C then ramp 40
°C/min to 250 °C and hold 2 min, run time: 25 min; Detector:
240 °C; Retention time (min): 4 (15.2). The excess Boc2O was
quenched by addition of N,N-dimethylethylenediamine (2 kg,
23 mol) and stirred until Boc2O was consumed as determined
by GC (typically about 30 min). GC conditions: Column:
Alltech AT-1, 30 m × 0.53 mm, 1.2 µm, injector: 150 °C, split
ratio 1:1, flow 5.0 mL/min constant pressure, 2.0 µL injection;
Oven program: 110 °C then ramp 4 °C/min to 200 °C then
ramp 30 °C/min to 270 °C and hold 5 min, run time: 30 min;
Detector: 240 °C; Retention time (min): Boc2O (6.7), 4 (7.4),
N-Boc ethyl isonipecotate (19.8), 7 (20.0). The reaction was
quenched with 1 M H3PO4 solution (prepared from 75 kg water
and 10 kg of H3PO4). The layers were separated, and the organic
phase was washed with a basic brine solution (prepared from
75 kg of water, 0.75 kg of NaOH pellets and 11 kg of NaCl).
The organic layer was assayed (HPLC) for product at 110 kg
(99% yield). HPLC conditions: Column: Phenomenex, Luna
phenyl hexyl, 250 mm × 4.6 mm, 5 µm, 25 °C, 1 mL/min, 20
Preparation of tert-Butyl 4-carbamoyl-4-methylpiperi-
dine-1-carboxylate (8). A mixture of sodium amide (38 kg,
974 mol, 2.5 equiv) and THF (400 kg) was warmed to 40 °C.
Ester 7 was charged as a solution in toluene (105 kg active,
387 mol, 1 equiv) over 4 h at a temperature of not more than
65 °C (target 50 °C). The addition vessel was rinsed with THF
(25 kg), which was added to the reaction vessel. The reaction
was adjusted to 50 °C and held 2 h. In a separate vessel was
charged water (320 kg), concentrated HCl (60 kg), and toluene
(200 kg), which was then chilled to <5 °C. The heterogeneous
reaction mixture was quenched by addition to the 3 M HCl/
toluene mixture, while maintaining the temperature at less than
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1149

25 °C. The reactor was rinsed with toluene (75 kg) and water
(140 kg), which were added to the quench vessel. After warming
to RT, the layers were separated. The organic solution was
concentrated to about 300 L. Toluene (200 kg) was added, and
the mixture was distilled to a total volume of ∼300 L. The
slurry was chilled to ∼0 °C over 4 h, stirred for 3 h, and then
filtered. The cake was rinsed with toluene (400 kg) and dried
under vacuum at 50 °C for 48 h to provide 72.5 kg (77% yield)
Retention time (min): 10 (7.9), toluene (18.8), 5 (21.5). The
reaction was quenched by the addition of a 0 °C solution of
NaCl (15 kg) and KH2PO4 (15 kg) in water (160 kg). MTBE
(140 kg) was added, and the mixture was stirred and allowed
to settle. The layers were separated. The organic phase was
washed three times with water (120 kg each), then filtered and
diluted with toluene (120 kg) and distilled under vacuum. The
product unexpectedly crystallized after the distillation. THF (10
kg) was added to dissolve the solid. The solution was assayed
for 52.1 kg of 5 (96.3% yield). An analytical sample was
prepared by crystallization from toluene. 1H NMR (400 MHz,
CDCl3) δ 1.58 (s, 9 H), 7.69 (dd, J ) 5.08, 1.37 Hz, 1 H), 7.78
(dd, J ) 1.37, 0.69 Hz, 1 H), 8.47 (dd, J ) 5.08, 0.82 Hz, 1
H). 13C NMR (101 MHz, CDCl3) δ 28.16, 83.03, 121.28,
1
of amide 8. H NMR (400 MHz, CDCl3) δ 1.23 (s, 3 H)
1.37-1.51 (m, 2 H) 1.44 (s, 9H) 1.88-1.98 (m, 2 H) 3.19-3.32
(m, 2 H) 3.51-3.68 (m, 2 H) 5.56-5.76 (m, 2 H). 13C NMR
(101 MHz, CDCl3) δ 25.82, 28.70, 34.98, 41.35, 79.53, 154.37,
178.22. IR (KBr) 3388, 3172, 1665 cm-1. Anal. Calcd for
C12H22N2O3: C 59.48, H 9.15, N 11.56; found: C 59.57, H 9.37,
N 11.58.
123.61, 141.75, 149.76, 151.69, 162.14. IR (KBr) 1716 cm-1
.
Preparation of 4-Methylpiperidine-4-carboxamide 4-Me-
thylbenzenesulfonate (9 ·TsOH). A mixture of amide 8 (72
kg, 299 mol, 1 equiv), p-toluenesulfonic acid monohydrate
(TsOH·H2O, 73 kg, 384 mol, 1.3 equiv), and isopropanol (IPA,
285 kg) was warmed to 80 °C. After 2 h at 80 °C HPLC
analysis showed that no SM remained. HPLC Conditions:
Column: Phenomenex, Synergi Hydro-RP, 250 mm × 4.6 mm,
4 µm, 25 °C, 1 mL/min, 20 µL injection, λ 210 nm; Mobile
phase A: 10 mM phosphate buffer, pH 6.5; Mobile phase B:
ACN; Gradient: 0 min 2% B, 30 min 60% B, 35 min 100% B,
38 min 100% B, 40 min 2% B, 50 min 2% B; Retention time
(min): 9 (3.8), TsOH (11.5). The reaction solution was allowed
to slowly cool; at ∼78 °C the product began to precipitate from
the reaction mixture.15 When the mixture reached 70 °C, heptane
(62 kg) was added to the mixture, which was then stirred for
about 30 min. The mixture was cooled to 20 °C at a rate of 5
°C/h and was stirred for 3 h. The suspension was filtered, and
the reactor and cake were rinsed with a mixture of IPA (142
kg) and heptane (125 kg). The combined mother liquor and
rinse contained 3.1 kg of product salt (3.3% yield) by HPLC
analysis. The cake was dried under vacuum at 55 °C to give
87.7 kg (93.4% yield) of 9·TsOH that was 96.7 wt % pure by
HPLC analysis. 1H NMR (400 MHz, CD3OD) δ 1.28 (s, 3H),
1.67 (ddd, J ) 15.2, 11.5, 4.0, 2H), 2.33-2.16 (m, 2H), 2.39
(s, 3H), 3.07 (ddd, J ) 13.1, 11.5, 3.1, 2H), 3.31-3.24 (m,
2H), 7.33-7.16 (m, 2H), 7.80-7.63 (m, 2H). 13C NMR (101
MHz, MeOH-d4) δ 21.61, 26.94, 32.97, 41.30, 43.04, 126.65,
129.55, 141.40, 143.08, 179.57. Anal. Calcd for C14H22N2O4S:
C 53.48, H 7.05, N 8.91; found: C 53.36, H 6.96, N 8.70.
Preparation of tert-Butyl 2-Chloroisonicotinate (5). A
mixture of 2-chloroisonicotinic acid (10, 40 kg, 253 mol, 1
equiv), Boc2O (115 kg, 528 mol, 2.3 equiv), and NMP (80 kg)
was cooled to 10 °C. A solution of DMAP (5.8 kg, 47.5 mol,
0.2 equiv) dissolved in NMP (30 kg) was added over 10 min,
and the temperature was adjusted to 25 °C. After 12 h HPLC
analysis showed no starting material remained. HPLC Condi-
tions: Column: Phenomenex, Synergi Hydro-RP, 250 mm ×
4.6 mm, 4 µm, 25 °C, 1 mL/min, 20 µL injection, λ 210 nm;
Mobile phase A: 0.1% perchloric acid; Mobile phase B: ACN;
Gradient: 0 min 20% B, 8 min 25% B, 25 min 80% B, 30 min
95% B, 35 min 95% B, 36 min 20% B, 45 min 20% B;
Anal. Calcd for C10H12ClNO2: C 56.21, H 5.66, N 6.56; found:
C 56.27, H 5.83, N 6.54.
Preparation of tert-Butyl 2-(4-carbamoyl-4-methylpip-
eridin-1-yl)isonicotinate (11). The crude solution of tert-butyl-
2-chloroisonicotinate (5, 93 kg of a 56 wt % solution, 244 mol,
1 equiv) was concentrated to about 87 wt %, and DMSO (119
kg) was added. The resulting solution was added to a mixture
of 9·TsOH (88 kg, 280 mol, 1.5 equiv) and K2CO3 (325 mesh,
90 kg, 651 mol, 3.5 equiv). The mixture was stirred at 100 °C
until HPLC analysis showed less than 1 LC A% of 5 remained
(36 h total). HPLC Conditions: Column: Phenomenex, Synergi
Hydro-RP, 250 mm × 4.6 mm, 4 µm, 25 °C; 1 mL/min, 20
µL injection, λ 210 nm; Mobile phase A: 0.1% perchloric acid;
Mobile phase B: ACN; Gradient: 0 min 20% B, 8 min 25% B,
25 min 80% B, 30 min 95% B, 35 min 95% B, 36 min 20% B,
45 min 20% B; Retention time (min): 10 (7.8), 11 (9.1), 5 (21.5).
The mixture was cooled to room temperature, and water (544
kg) was added, slowly at first. The addition of water caused
the product to precipitate, and slow addition was necessary to
control the rate of precipitation. After stirring for about 1 h,
the mixture was filtered, and the solid was washed twice with
water (272 kg each) and dried under vacuum to give 65.1 kg
1
(83.6% yield) of white solid. H NMR (400 MHz, CDCl3) δ
1.28 (s, 3 H) 1.53-1.63 (m, 2 H) 1.58 (s, 9 H) 2.03-2.11 (m,
2 H) 3.42-3.51 (m, 2 H) 3.78-3.86 (m, 2 H) 5.41-5.69 (m,
2 H) 7.03 (dd, J ) 5.08, 1.24 Hz, 1 H) 7.18 (s, 1 H) 8.22 (dd,
J ) 5.15, 0.75 Hz, 1 H). 13C NMR (101 MHz, CDCl3) δ 25.88,
28.33, 34.62, 41.53, 42.61, 81.87, 106.59, 111.48, 140.35,
148.12, 159.25, 164.53, 178.43. IR (KBr) 3425, 1723 cm-1.
Anal. Calcd for C17H25N3O3: C 63.93, H 7.89, N 13.16; found:
C 63.56, H 7.80, N 12.90.
Preparation of tert-Butyl 2-(4-amino-4-methylpiperidin-
1-yl)isonicotinate (2). A mixture of amide 11 (49 kg, 153 mol,
1 equiv) and tetrabutylammonium bromide (TBAB, 50 kg, 153
mol, 1 equiv) and THF (132 kg) were cooled to 0 °C. A portion
of 4 M NaOH (393 kg, of a solution prepared by dissolving 96
kg of NaOH in 578 kg water) was added, and the temperature
of the slurry was adjusted to -5 °C. 1,3-Dibromo-5,5-dimeth-
ylhydantoin (DBDMH, 24 kg, 84 mol, 0.55 equiv) was added
in portions over 20 min, resulting in an exotherm to 1.6 °C. It
is important to keep the reaction temperature below 5 °C in
order to avoid formation of impurities. The reaction was stirred
at 0 °C for 90 min, at which time HPLC analysis showed less
(15) If product does not precipitate, the reaction may be seeded with
approximately 0.2 wt % seeds.
1150
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development

than 1 LC A% 11 remaining and less than 1 LC A% isocyanate
intermediate. HPLC conditions: Column: Phenomenex, Synergi
Hydro, 250 mm × 4.6 mm, 4 µm, 25 °C, 1 mL/min, 20 µL
injection, λ 210 nm; Mobile phase A: 0.1% perchloric acid;
Mobile phase B: ACN; Gradient: 0 min 10% B, 30 min 90%
B, 35 min 90% B, 37 min 10% B, 45 min 10% B; Retention
time (min): 2 (12.4), 11 (14.4), urea 13 (20.5), isocyanate (21.2).
The reaction was transferred into the 10 °C quench solution of
6 M HCl (294 g), tert-butyl methyl ether (MTBE, 91 kg), and
Na2SO3 (20 kg, 156 mol), which resulted in an exotherm to
about 15 °C and the evolution of carbon dioxide gas. The rate
of addition of the reaction mixture to the quench solution
allowed the rate of CO2 release to be controlled. The reaction
vessel was rinsed with MTBE (91 kg), which was added to the
quench vessel. The quenched reaction solution had a pH of
approximately 3 as determined by pH paper. After stirring for
5 min, 4 M NaOH (196 kg) was charged to the quenched
reaction solution such that the temperature remained below 20
°C. The quenched reaction solution had a pH of approximately
12 as determined by pH paper. The layers were separated, and
the aqueous layer was back extracted with MTBE (91 kg). The
combined organic layers were washed with a brine-sulfite
solution (prepared from 12.3 kg Na2SO3, 24.5 kg NaCl, and
208 kg water). The combined organic layers were distilled to
about half the original volume, NMP (60 kg) was added, and
the distillation was continued until less than 1% MTBE
remained as determined by NMR analysis. The NMP solution
of product was assayed at 38.7 kg (86.8% yield) and 96.2 LC
A% pure. Compound 2 was characterized as the succinate salt,
formed by treatment of the amine in MeOH with 1 equiv of
succinic acid and isolated by crystallization from 10% MeOH/
IPAc. (1:1 salt ratio). 1H NMR (400 MHz, DMSO-d6) δ 1.30
(s, 3 H), 1.53 (s, 9 H), 1.64 (t, J ) 5.63 Hz, 4 H), 2.25 (s, 4 H),
3.30-3.39 (m, 2 H), 3.91 (ddd, J ) 13.72, 4.87, 4.73 Hz, 2
H), 6.95 (dd, J ) 5.1, 1.2, 1 H), 7.16 (s, 1 H), 8.22 (d, J ) 5.1,
1 H). 13C NMR (101 MHz, DMSO-d6) δ 23.72, 27.69, 32.07,
34.89, 40.75, 51.07, 81.45, 105.59, 110.64, 139.72, 148.11,
158.31, 163.67, 174.34. IR (KBr) 1753, 1709 cm-1. Anal. Calcd
for C20H31N3O6: C 58.66 H 7.63 N 10.26; found: C 58.60 H
7.70 N 10.21.
Preparation of (S)-1-tert-Butyl 2-methyl 5-((triisopropyl-
silyl)ethynyl)-2,3-dihydro-1H-pyrrole-1,2-dicarboxylate (18).
To a THF solution of triisopropylsilylacetylene (12.0 mL, 53.5
mmol, 1.3 equiv) cooled in an ice bath was added dropwise
butyllithium (2.5 M in hexane, 20 mL, 50 mmol, 1.2 equiv),
keeping the temperature less than 5 °C. The resulting mixture
was stirred for 1 h at 0 °C. The solution of alkynyl lithium was
cooled to -20 °C, and a solution of 6 (10.0 g, 41.1 mmol) in
THF (50 mL) was added dropwise over 2 h. After an additional
1.5 h, BF3 ·OEt2 (50 mL, 395 mmol, 9.6 equiv) was added,
and the resulting mixture was stirred for 3 h at -20 °C. The
reaction mixture was quenched into saturated aq NaHCO3 (50
mL) and stirred for 15 min, and the layers were separated. The
organic layer was washed with saturated aq NaHCO3 (50 mL),
and the combined aqueous layers were extracted with EtOAc
(100 mL). The combined organic layers were washed with
brine, dried over Na2SO4, filtered, and concentrated. Purification
by silica gel chromatography eluting with 5% EtOAc in hexane
afforded 7.6 g (45%) of clear colorless oil. 1H NMR (400 MHz,
CDCl3) δ 5.40 (t, J ) 3.1, 1 H), 4.70 (dd, J ) 11.9, 5.4, 1 H),
3.75 (s, 3 H), 2.98 (ddd, J ) 18.1, 11.9, 2.9, 1 H), 2.59 (ddd,
J ) 18.1, 5.4, 3.3, 1 H), 1.47 (s, 9 H), 1.18-0.99 (m, 21 H).
Preparation of (S)-Methyl 2-(tert-butoxycarbonylamino)-
5-oxo-7-(trimethylsilyl)hept-6-ynoate (14). A solution of oc-
tylmagnesium chloride (2.1 M, 190 kg, 429 mol, 1.1 equiv)
and THF (75 kg) was cooled to 0 °C. Trimethylsilylacetylene
(44 kg, 448 mol, 1.15 equiv) was added by subsurface addition
over about 25 min such that the temperature remained below
15 °C. The solution was stirred at 0 °C for 1 h and was then
cooled to -10 °C. A solution of ester 6 (95 kg, 391 mol, 1
equiv) in THF (180 kg) was added by subsurface addition over
a 2-h period. Slow, subsurface addition was necessary to
minimize local concentration of 6. The vessel and lines were
rinsed with about 5 kg of THF, which was added to the reaction
vessel. HPLC analysis after 2.5 h showed less than 2 LC A%
6 remained. HPLC conditions: Column: Phemomenex, Luna
C8(2), 4.6 mm × 250 mm, 5 µm, 25 °C, 1.0 mL/min, 20 µL
injection, λ 210 nm; Mobile phase A: 0.1% H3PO4; Mobile
phase B: 0.1% H3PO4 in ACN; Gradient: 0 min, 30% B; 12
min, 40% B; 16 min, 72% B; 26 min, 74% B; 30 min, 95% B;
35 min, 95% B; 36 min, 30% B; 45 min, 30% B; Retention
time (min): 6 (10.9), 14 (23.0). The reaction solution was added
to a 0 °C mixture of IPAc (290 kg) and 20% NH4Cl solution
(378 kg, prepared by dissolving 126 kg of NH4Cl in 630 kg
water) over about 30 min such that the temperature remained
below 10 °C. The mixture was allowed to warm to above 10
°C, and the layers were separated. The aqueous layer was
extracted with IPAc (250 kg), and the organic layers were
combined. HPLC analysis of the aqueous layer showed no
product. The organic layer was washed with the remainder of
the 20% NH4Cl solution (378 kg) and with 20% NaCl solution
(377 kg, prepared by dissolving 62 kg of NaCl in 315 kg water).
HPLC analysis of the organic layer showed 111.9 kg (84.8%
yield) of ketone 14. An analytical sample was prepared by silica
gel chromatography using 10% EtOAc/hexanes as an eluent
followed by crystallization from heptane. 1H NMR (400 MHz,
CDCl3) δ 0.18 (s, 9 H), 1.39 (s, 9 H), 1.82-1.97 (m, 1 H),
2.14 (tdd, J ) 5.3, 6.7, 11.9, 1 H), 2.62-2.73 (m, 2 H), 3.70
(s, 3 H), 4.25 (bs, 1 H), 5.08 (bs, 1 H). 13C NMR (101 MHz,)
δ -0.2, 26.9, 28.7, 41.5, 52.7, 53.0, 80.2, 98.5, 101.6, 155.1,
172.2, 185.4. MS (DCI - NH3) (M + 18) 359.1 m/e IR (KBr)
3378, 2972, 2157, 1737, 1719, 1673 cm-1. [R]D ) +14.89° (c
) 0.991, CHCl3). Anal. Calcd for C16H27NO5Si: C 56.28, H
7.97, N 4.10; found: C 56.18, H 7.92, N 4.06.
Preparation of (2S,5R)-1-(tert-Butoxycarbonyl)-5-ethyn-
ylpyrrolidine-2-carboxylic Acid (19). Cyclization/Reduction.
A mixture of sodium triacetoxyborohydride (94 kg, 444 mol,
1.3 equiv), IPAc (110 kg), and a 26.6 wt % IPAc solution of
TMS ketone 14 (421 kg, 328 mol, 1 equiv) was cooled to -10
°C. Trifluoroacetic acid (160 kg, 1400 mol, 4.3 equiv) was
added over a 2-h period at an internal temperature of less than
10 °C. The temperature was adjusted to 10 °C, and the mixture
was stirred until HPLC analysis indicated complete consumption
of starting material (about 8 h). HPLC conditions: Column:
Phenomenex, Hydro-RP, 4.6 mm × 250 mm, 4 µm, 25 °C, 1
mL/min, 20 µL injection, λ 210 nm; Mobile phase A: 0.1%
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1151

H3PO4; Mobile phase B: 0.1% H3PO4 in ACN; Gradient: 0 min,
45% B; 10 min, 70% B; 25 min, 80% B; 30 min, 95% B; 35
min, 95% B; 36 min, 45% B; 45 min, 45% B; Retention time
(min): 14 (7.5), 17 (19.2). The reaction was added to 750 kg of
25% K2HPO4 (prepared by dissolving 250 kg of K2HPO4 in
750 kg of water). The pH of the mixture was adjusted to 6.6
using 20% KOH (prepared by dissolving 100 kg of KOH in
400 kg water). The layers were separated. The organic layer
was washed with the remaining 25% K2HPO4 solution (250
kg) and with water (250 kg). The organic layer was distilled to
remove IPAc, which was chased with EtOH to a final volume
of about 250 L. An analytical sample was prepared by silica
gel chromatography using EtOAc/hexane as an eluent. 1H NMR
(400 MHz, DMSO-d6, 333 K) δ -0.01 (s, 9 H), 1.25 (s, 9 H),
1.66-1.90 (m, 2 H), 1.92-2.16 (m, 2 H), 3.51 (s, 3 H),
4.00-4.07 (m, 1 H), 4.27-4.38 (m, 1 H), 12.15 (br s, 1 H).
13C NMR (101 MHz, DMSO-d6) δ 0.03, 28.2, 32.4, 49.5, 51.8,
59.5, 79.8, 107.0, 172.7. MS (DCI - NH3) (M + 1) 326.1 m/e.
IR (KBr) 2959, 2176, 1760, 1703 cm-1. [R]D ) +61.67° (c )
1.028, CHCl3). Anal. Calcd for C16H27NO4Si: C 59.04, H 8.36,
N 4.30; found: C 58.76, H 8.29, N 4.27.
Hydrolysis. The EtOH solution of 17 was cooled to 0 °C,
and a 3.5 M LiOH solution (287 kg, prepared by dissolving 37
kg of LiOH ·H2O in 250 kg of water) was added such that the
temperature remained below 30 °C. The reaction progress was
followed by HPLC until no ester remained (2 h). HPLC
conditions: Column: Zorbax, Eclipse, XDB-C8, 250 mm × 4.6
mm, 5 µm, 25 °C, 1 mL/min, 20 µL injection, λ 210 nm; Mobile
phase A: 0.1% H3PO4; Mobile phase B: 0.1% H3PO4 in ACN;
Gradient: 0 min, 25% B; 25 min, 72% B; 30 min, 95% B; 35
min, 25% B; 45 min, 25% B; Retention time (min): 19-cis
(10.2), 19-trans (11.0), 17 (28.5). The ethanol was replaced
with MTBE (185 kg) by distillation under vacuum, and the
layers were separated. The MTBE layer was extracted with
water (100 kg) and discarded. The aqueous layer was cooled
to 0 °C and neutralized to pH 7 with concentrated HCl. IPAc
(220 kg) was added, and the pH was adjusted to 3 with
concentrated HCl. The layers were separated, and the aqueous
layer was extracted with IPAc (220 kg). The combined organic
layers were washed with water (100 kg). The organic solution
was distilled under vacuum to a total volume of about 180 L.
IPAc (200 kg) was added, and the organic solution was distilled
under vacuum to a total volume of about 180 L and then cooled
to 0 °C. Crystals formed over a 30-min period. Heptanes (50
kg) was added over a 2-h period, and the resulting mixture was
stirred for 2 h at 0 °C. The mixture was filtered, and the solid
was washed with cold 1:1 IPAc/heptanes (50 L) and dried under
vacuum at 40 °C. The procedure gave 52 kg of acid 19 (65%
yield). The final cis/trans ratio was >99.5 cis/<0.5% trans. 1H
NMR (600 MHz, DMSO-d6, 343 K) δ 1.39 (s, 9 H), 1.91 (dt,
J ) 7.7, 15.6, 1 H), 1.98 (dt, J ) 7.2, 15.8, 1 H), 2.08-2.16
(m, 2 H), 2.22 (td, J ) 7.1, 12.1, 1 H), 4.05-4.12 (m, 1 H),
4.40-4.49 (m, 1 H). 13C NMR (151 MHz, DMSO-d6, 343 K)
δ 27.7, 28.4, 31.8, 48.2, 59.0, 71.7, 78.9, 83.8, 152.4, 172.6.
MS (DCI - NH3) (M + 18) 257.1 m/e IR (KBr) 1745, 1648
cm-1. [R]D ) +23.45° (c ) 0.99, CHCl3). Anal. Calcd for
C12H17NO4: C 60.24, H 7.16, N 5.85; found: C 60.26, H 7.30,
N 5.83.
Preparation of (2S,5R)-1-(2-Chloroacetyl)-5-ethynylpyr-
rolidine-2-carbonitrile (3). Amide Formation. A mixture of
acid 19 (50 kg, 209 mol, 1 equiv), IPAc (214 kg) and
N-methylmorpholine (NMM, 27 kg, 267 mol, 1.25 equiv) was
cooled to 0 °C. Isobutyl chloroformate (34 kg, 249 mol, 1.2
equiv) was added over about 30 min at a temperature of less
than 30 °C, and the charge vessel and lines were rinsed with
IPAc (5 kg), which was added to the reaction vessel. The
resulting solution was stirred at 0 °C for 1 h. A portion of the
reaction was quenched into benzyl amine and analyzed by
HPLC to evaluate the progress of the mixed anhydride forma-
tion, which is typically complete at 1 h reaction time. HPLC
conditions: Column: Phenomenex, Synergi 4 µm Fusion-RP
80, 250 mm × 4.6 mm column, 1.0 mL/min, 35 °C, λ 205 nm,
10 µL injection; Mobile phase A: 0.1% H3PO4; Mobile phase
B: 0.1% H3PO4 in ACN; Gradient starts at 85% A and decrease
to 65% A at 12 min and decrease to 5% A at 25 min, hold at
5% A to 35 min; Retention time (min): amide intermediate
(13.9), 3 (14.6) 19 (17.4), 20 (21.0), benzyl amide intermediate
(25.0). The reaction mixture was added to cold (0 °C) 28%
NH4OH solution (135 kg) such that the internal temperature of
the quench remained below 30 °C. The reaction vessel was
rinsed with IPAc (20 kg), which was added to the quench vessel.
The mixture was diluted with 26% NaCl solution (150 kg,
prepared by dissolving 39 kg of NaCl in 111 kg of water), and
the layers were separated. The aqueous layer was extracted with
IPAc (100 kg). The combined organic layers were washed with
20% KH2PO4 (2 × 250 kg, prepared by dissolving 100 kg of
KH2PO4 in 400 kg water). The organic solution was concen-
trated under vacuum to a total volume of about 85 L. IPAc
(219 kg) was added, and the solution was concentrated under
vacuum to a total volume of about 85 L. IPAc (219 kg) was
added, and the solution was concentrated under vacuum to a
total volume of about 85 L. The solution contained 0.4% water
by Karl Fischer titration.
Nitrile Formation. A solution of THF (233 kg) and DMF
(37 kg, 506 mol, 2.4 equiv) was cooled to 0 °C. Thionyl chloride
(SOCl2, 57 kg, 479 mol, 2.3 equiv) was added slowly such that
the internal temperature remained below 30 °C. The Vilsmeier
solution was stirred at 0 °C for 1.5 h. The IPAc solution of the
amide from above was added such that the internal tempera-
ture remained below 30 °C. IPAc (10 kg) was used to rinse the
vessel containing the solution of amide and was added to the
reaction vessel. The resulting mixture was stirred for 1 h at 0
°C and was analyzed for reaction completion by HPLC (see
HPLC conditions above). The reaction solution was added to
cold (0 °C) 5 M NaOH solution (300 kg, prepared by dissolving
50 kg of NaOH in 250 kg of water) such that the internal
temperature remained below 25 °C. The layers were separated,
and the organic layer was washed with 20% KH2PO4 solution
(250 kg, prepared by dissolving 50 kg of KH2PO4 in 200 kg of
water) and 5% NaCl solution (250 kg, prepared by dissolving
12 kg of NaCl in 238 kg of water). The organic layer was
distilled to a total volume of about 80 L and was diluted with
CH3CN (197 kg) a total of three times. The organic solution
was analyzed by GC to verify that the level of IPAc was below
10% and by HPLC for yield. The organic layer contained 40.1
kg of nitrile 20 (87% yield) and was used directly in the next
1152
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development

step. The HPLC of the crude solution typically contains DMF
(9 LC A%), IPAc (9 LC A%), and N-formyl-isobutylcarbamate
(18 LC A%). These materials do not interfere in subsequent
reactions.
at 40 °C in a vacuum oven for 48 h to obtain 25.4 kg (78.6%
yield, 99.2 LC A%) of tan solid. The minor diastereomer was
not detectable; HPLC Conditions: Column: Daicel Chiralcel
OD-H, 250 mm × 4.6 mm, 0.8 mL/min, 35 °C, λ 205 nm, 5
µL injection; Mobile phase: 80:20 heptane/ethanol; Retention
Boc Deprotection. The CH3CN solution of nitrile 20 (40 kg
active, 183 mol, 1 equiv) was further diluted with CH3CN (39
kg) to bring the total to 10 L of CH3CN/kg 20. p-Toluene-
sulfonic acid monohydrate (TsOH ·H2O, 71 kg, 365 mol, 2
equiv) was added in portions, keeping the internal temperature
below 30 °C. CO2 gas evolution was vigorous; thus, venting
for CO2 was essential. The reaction was stirred for 8 h at 25
°C during which time the reaction became heterogeneous.
Disappearance of the starting material was followed by HPLC
(see HPLC conditions above). The product solution was used
directly in the next reaction. An analytically pure sample was
prepared by crystallization from cold (0 °C) IPAc (10 g/g
1
time (min): trans-3 (11.2), cis-3 (14.0). H NMR (400 MHz,
CDCl3) δ 2.33-2.49 (m, 4 H), 2.63 (d, J ) 2.06 Hz, 1 H),
4.26 (d, J ) 13.50 Hz, 1 H), 4.46 (d, J ) 13.50 Hz, 1 H),
4.69-4.76 (m, 2 H). 13C NMR (101 MHz, CDCl3) δ 28.65,
33.29, 41.74, 46.82, 48.43, 74.33, 80.18, 117.35, 164.50. IR
(KBr) 3252, 1669 cm-1. Anal. Calcd for C9H9ClN2O: C 54.97,
H 4.61, N 14.25; found: C 54.89, H 4.39, N 14.31.
Preparation of tert-Butyl 2-(4-(2-((2S,5R)-2-cyano-5-ethyn-
ylpyrrolidin-1-yl)-2-oxoethylamino)-4-methylpiperidin-1-yl)-
isonicotinate (2S,3S)-2,3-Dihydroxysuccinate (22). A mixture
of amine 2 in NMP (123 kg of a 31.4 wt % solution, 131 mol,
1 equiv), milled K3PO4 (42 kg, 212 mol, 1.5 equiv), and KI
(2.2 kg, 13.3 mol, 0.1 equiv) was degassed by nitrogen sparging
for 15 min, then the solution was warmed to 35 °C. Chloro-
acetamide 3 (27.5 kg, 140 mol, 1.05 equiv) was dissolved in
NMP (45 kg) and added over 30 min, maintaining the
temperature between 35 and 40 °C. The transfer lines were
rinsed with NMP (5 kg), which was added to the reaction vessel.
After 3 h, HPLC analysis of the reaction mixture indicated
>95% conversion. HPLC conditions: Column: Phenomenex,
Synergi Hydro, 250 mm × 4.6 mm, 4 µm, 25 °C, 1 mL/min,
20 µL injection, λ 210 nm; Mobile phase A: 0.1% perchloric
acid; Mobile phase B: ACN; Gradient: 0 min 10% B, 30 min
90% B, 35 min 90% B, 37 min 10% B, 45 min 10% B;
Retention time (min): 2 (12.4), 3 (13.3), 22 (16.0). After cooling
to 25 °C, MTBE (425 kg) and water (450 kg) were charged,
and the phases were separated. The MTBE layer was sequen-
tially washed with 5% KH2PO4 (453 kg, prepared by dissolving
23 kg of KH2PO4 in 430 kg of water) and then water (450 kg).
The MTBE layer was filtered through a Cuno R53SP carbon
filter, rinsing with MTBE (20 kg). The carbon filtration is
necessary to remove color that can be carried into the final
product. The source of the color was never identified. The
MTBE solution was distilled to approximately 340 L. As
the distillation continued, IPA (680 kg) was added to maintain
the solvent level around 340 L. The resulting IPA solution was
warmed to 60 °C, and D-tartaric acid (20 kg, 132 mol, 1.0 equiv)
was charged, and the solution was stirred for 30 min at 60 °C.
After nucleation occurred, the slurry was cooled to ambient
temperature over 30 min, filtered, and washed with IPA (2 ×
150 kg). The cake was dried under vacuum at 50 °C with a
nitrogen bleed to give 71 kg of a yellow solid. The color of the
product was the result of ineffective washing of the wetcake
due to an equipment malfunction. A reslurry procedure was
carried out to further purify the material. The isolated solid (71
kg) was suspended in IPA (330 kg). The mixture was stirred at
25 °C for 1 h and filtered. The wetcake was rinsed with two
portions of IPA (80 kg each) and dried under vacuum to give
1
product). H NMR (400 MHz, CD3OD) δ 2.17-2.29 (m, 1
H), 2.36 (s, 3 H), 2.39-2.51 (m, 2 H), 2.53-2.65 (m, 1 H),
3.41 (d, J ) 2.33 Hz, 1 H), 4.59 (td, J ) 7.03, 2.40 Hz, 1 H),
4.81 (dd, J ) 8.71, 6.79 Hz, 1 H), 4.86-5.03 (m, 2 H), 7.23
(d, J ) 7.82 Hz, 2 H), 7.70 (d, J ) 8.23 Hz, 2 H). 13C NMR
(101 MHz, CD3OD) δ 21.5, 30.5, 32.5, 47.3, 51.7, 77.0, 79.7,
115.9, 126.5, 129.4, 141.4, 142.7. IR (KBr) 3269, 2956, 2560,
2257, 2135 cm-1. Anal. Calcd for C14H16N2O3S: C 57.52 H
5.52 N 9.58; found: C 57.21, H 5.48, N 9.49.
Chloroacetamide Formation. Upon complete deprotection,
the reaction was cooled to 0 °C and N,N-diisopropylethylamine
(52 kg, 401 mol, 2.2 equiv) was added such that the internal
temperature remained below 10 °C. The transfer line was rinsed
with CH3CN (2 kg), which was added to the reaction vessel.
After the addition the reaction mixture was cooled to 0 °C, and
chloroacetyl chloride (25 kg, 219 mol, 1.2 equiv) was added
such that the internal temperature remained below 10 °C. The
reaction was mixed at 0 °C, followed by HPLC for the reaction
completion (see HPLC conditions above). The reaction was
slowly added to a cold (0 °C) solution of IPAc (360 kg) and 1
N K2HPO4 (490 kg, prepared by dissolving 109 kg of K2HPO4
in 625 kg of water). The transfer line was rinsed with CH3CN
(20 kg), which was added to the reaction vessel. The mixture
was concentrated to about 500 L and chased with IPAc (720
kg) and water (335 kg) to a total volume of about 450 L. Upon
removal of the CH3CN (less than 1% as determined by NMR)
the layers were separated. The aqueous layer was extracted with
IPAc (2 × 180 kg). The combined organic layers were washed
with the remaining 1 N K2HPO4 solution (244 kg).
Crystallization. The organic solution was azeotropically dried
by distillation from IPAc (400 kg) and filtered to remove
inorganic salts. Karl Fischer analysis of the solution showed
0.5 wt % water. The IPAc solution was solvent exchanged with
IPA by a constant volume distillation with the addition of 380
kg of IPA. The IPA solution was concentrated to about 80 L
and stirred at room temperature overnight. Solids typically
started to crash out during the solvent exchange to IPA. Water
(450 kg) was added over 2 h, and the mixture was stirred for
1 h at room temperature. The mixture was cooled to 0 °C to
bring the liquor concentration to about 4.5 mg/mL, as deter-
mined by HPLC. Product was filtered, and the solid was washed
with cold (2 °C) 1:7 IPA/water (72 kg). The product was dried
1
68.7 kg of white solid (88% yield). H NMR (400 MHz,
DMSO-d6) δ 1.20 (s, 3 H) 1.53 (s, 9 H) 1.55-1.74 (m, 4 H)
2.07-2.16 (m, 1 H) 2.17-2.43 (m, 3 H) 3.33-3.45 (m, 2 H)
3.61 (d, J ) 2.20 Hz, 1 H) 3.73-3.87 (m, 3 H) 4.12 (s, 2 H)
4.77 (t, J ) 7.00 Hz, 1 H) 4.91-4.98 (m, 1 H) 6.94 (dd, J )
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1153

5.08, 1.10 Hz, 1 H) 7.14 (s, 1 H) 8.21 (d, J ) 5.08 Hz, 1 H).
13C NMR (101 MHz, DMSO-d6) δ 22.06, 27.69, 28.44, 32.58,
34.33, 34.39, 40.88, 42.59, 45.97, 47.53, 53.17, 71.65, 75.67,
81.36, 81.65, 105.51, 110.42, 118.51, 139.59, 148.05, 158.49,
163.70, 167.44, 172.87. IR (MIC) 3417, 3007, 1719, 1666 cm-1.
Anal. Calcd for C29H39N5O9: C 57.89 H 6.53 N 11.64; found:
C 57.81, H 6.56, N 11.42.
injection, λ 245 nm; Mobile phase A: 0.2% acetic acid in
MeOH; Mobile phase B: 0.2% Et3N in MeOH; Mobile phase
composition: 60% A/40% B; Retention time (min): ent-ABT-
279 (6.9), ABT-279 (8.9). 1H NMR (400 MHz, DMSO-d6) δ
1.26 (s, 3 H) 1.56-1.79 (m, 4 H) 1.96-2.17 (m, 1 H)
2.17-2.46 (m, 4 H) 2.56 (dd, J ) 15.64, 7.14 Hz, 1 H)
3.23-3.38 (m, 2 H) 3.51-3.75 (m, 2 H) 3.82-3.99 (m, 3 H)
4.08 (dd, J ) 7.14, 6.17 Hz, 1 H) 4.62-4.83 (m, 1 H)
4.91-5.26 (m, 1 H) 6.99 (dd, J ) 5.01, 1.03 Hz, 1 H) 7.20 (s,
Preparation of (S)-2-Hydroxysuccinic Acid Compound
with 2-(4-(2-((2S,5R)-2-Cyano-5-ethynylpyrrolidin-1-yl)-2-
oxoethylamino)-4-methylpiperidin-1-yl)isonicotinic Acid
(ABT-279, 1). Ester Deprotection. A mixture of 22 (69 kg,
115 mol, 1 equiv), water (648 kg), and 85% H3PO4 (37 kg,
382 mol, 3 equiv) was heated to 60 °C and held at this
temperature for 16 h to drive the reaction to completion as
determined by HPLC analysis. HPLC conditions: Column:
Thermo Aquasil C18, 4.6 mm × 150 mm, 3 µm, 25 °C, 1 mL/
min, 20 µL injection, λ 210 nm; Mobile phase A: 0.1%
perchloric acid; Mobile phase B: ACN; Gradient: 0 min 5% B,
2 min 5% B, 30 min 90% B, 31 min, 5% B, 40 min 5% B;
Retention time (min): ABT-279 (10.4), 22 (18.0). Upon
completion the solution was cooled to room temperature and
filtered to remove solids. The filter was rinsed with water (60
kg). The product was crystallized at room temperature as a
zwitterion by neutralizing the filtrate. A solution of 8.5% NaOH
(prepared by dissolving 30 kg of NaOH in 325 kg water) was
added over 30 min to adjust the pH to 6.5. After stirring at 20
°C for 3 h, the slurry was cooled to 10 °C, stirred for 2 h,
filtered, and washed with two 162-kg portions of water. The
wetcake was 99 LC A% pure, and the zwitterion was carried
1 H) 8.20 (d, J ) 5.08 Hz, 1 H) 8.62-10.82 (br m, 4 H). 13
C
NMR (101 MHz, DMSO-d6) δ 20.82, 28.48, 32.56, 33.72,
40.36, 40.94, 42.17, 46.03, 47.64, 54.59, 66.48, 73.82, 75.88,
81.51, 106.03, 111.17, 118.44, 140.70, 147.78, 158.43, 166.45,
171.33, 174.78. IR (MIC) 3295, 3092, 1731, 1672 cm-1. Anal.
Calcd for C25H31N5O8: C 56.70, H 5.90, N 13.23: found: C
56.68, H 5.85, N 13.15.
Alternative Hofmann/Amine Coupling Strategy. Prepa-
ration of 4-Methylpiperidin-4-amine bis(4-methylbenzene-
sulfonate) (23). Hofmann Reaction. A mixture of amide 8 (110
g, 0.454 mol, 1 equiv), CH3CN (260 g), and water (990 g) was
cooled to 10 °C, and KOH (131 g, 2.04 mol, 4.5 equiv) was
added, resulting in an exotherm from 10 to 24 °C. The slurry
was cooled to 1 °C, and 1,3-dibromo-5,5-dimethylhydantoin
(DBDMH, 71.4 g, 0.250 mol, 0.55 equiv) was added in one
portion, resulting in an exotherm from 1 to 3 °C. After 30 min
the reaction was warmed to 23 °C. After stirring for 1 h, no
starting material remained as determined by HPLC analysis.
HPLC conditions: Column: Phenomonex, Synergi Hydro RP,
4.6 mm × 250 mm, 4 µm, 35 °C, 1 mL/min, 10 µL injection,
λ 205 nm; Mobile phase A: 0.1% H3PO4; Mobile phase B:
ACN; Gradient: 0 min 10% B, 2 min 10% B, 17 min 90% B,
25 min, 90% B; Retention time (min): Boc amine (8.9), 8 (12.2).
Sodium sulfite (Na2SO3, 5.5 g, 0.044 mol, 0.1 equiv) was added,
and the reaction was stirred for 15 min. EtOAc (496 g) was
added and the reaction cooled to 11 °C. K3PO4 (110 g, 0.482
mol, 1.06 equiv) was added, resulting in an exotherm from 11
to 13 °C. The reaction was warmed to 23 °C, and the layers
were separated. The organic layer was washed once with a 25%
aqueous NaCl solution (138 g). The product lost to the
combined aqueous layers was 1.9% as determined by HPLC
analysis. The organic layer was distilled to an oil, which was
dissolved in MeOH (500 mL) and distilled to an oil. The oil
was dissolved in MeOH (740 mL) and held for use in the next
reaction.
1
on to the next step as a wetcake. H NMR (400 MHz,
CD3CO2D) δ 1.65 (s, 2 H) 2.09-2.22 (m, 2 H) 2.22-2.37 (m,
3 H) 2.37-2.54 (m, 3 H) 3.01 (d, J ) 2.06 Hz, 1 H) 3.46 (t, J
) 11.60 Hz, 2 H) 4.13-4.36 (m, 3 H) 4.38-4.53 (m, 1 H)
4.78-4.86 (m, 1 H) 4.86-4.94 (m, 1 H) 7.30 (dd, J ) 6.17,
0.96 Hz, 1 H) 7.64 (s, 1 H) 8.11 (d, J ) 6.18 Hz, 1 H). 13C
NMR (101 MHz, CD3CO2D) δ 19.36, 29.41, 33.47, 33.49,
33.78, 43.00, 43.13, 47.66, 49.30, 59.60, 75.79, 80.84, 112.06,
112.71, 117.94, 140.95, 146.40, 154.69, 165.28, 168.18, 177.08.
IR (MIC) 3190, 2113, 1674 cm-1.
Salt Formation/Crystallization. A mixture of the zwitterion
wetcake (57 kg, 76 wt %, 109 mol, 1 equiv), water (264 kg),
ethanol (215 kg), and L-malic acid (15.6 kg, 116 mol, 1.06
equiv) was heated to 70 °C to achieve dissolution. The clear
solution was filtered and 1:1 ethanol/water (59 kg) was used to
rinse the vessel and filter. The clear solution was cooled to 20
°C over 2 h and allowed to crystallize. The thin slurry was
distilled under reduced pressure to approximately 250 L and
further chased with ethanol (568 kg) to a final volume of 250
L. The desired solvent makeup of 15 wt % water in ethanol
was achieved by addition of ethanol (305 kg). In order to
minimize losses from product buildup on the reactor wall during
the distillation, the slurry was heated to 40 °C, held for 1 h,
cooled to 0 °C over 4 h, and stirred for 2 h. The product slurry
was filtered and washed with cold ethanol (2 × 83 kg). Product
was dried at 50 °C under reduced pressure to yield 41.6 kg of
ABT-279·L-malate salt (99.5 wt %, 99.6 LC A%, 99.4% ee,
93.2% yield). Chiral HPLC conditions: Column: Astec Chiro-
biotic T, 4.6 mm × 250 mm, 5 µm, 40 °C, 1 mL/min, 20 µL
Boc Deprotection. A mixture of p-toluenesulfonic acid
monohydrate (TsOH ·H2O, 197 g, 1.04 mol, 2.3 equiv) and IPA
(290 g) was heated to 60 °C. The MeOH solution from the
previous reaction was added over 30 min, during which time
the product crystallized. The resulting slurry was stirred for 19 h,
cooled to 0 °C, stirred for 1 h, and filtered. The wet cake was
washed twice with IPA (145 g) and dried under vacuum at 50
°C to give 192 g of 23 (92% yield for the two steps). 1H NMR
(400 MHz, DMSO-d6) δ 1.33 (s, 3 H) 1.77-1.91 (m, 4 H)
2.29 (s, 6 H) 3.03-3.14 (m, 2 H) 3.18-3.27 (m, 2 H) 3.33 (br
s, 2 H) 7.08-7.15 (m, 4 H) 7.44-7.52 (m, 4 H) 8.04-8.43 (br
s, 3 H). 13C NMR (DMSO-d6) δ 20.93, 22.35, 31.55, 39.17,
50.17, 125.01, 127.81, 137.71, 144.22. IR (KBr) 3336, 1724
1154
•
Vol. 13, No. 6, 2009 / Organic Process Research & Development

cm-1. Anal. Calcd for C20H30N2O6S2: C 52.38, H 6.59, N 6.11;
found: C 52.43, H 6.49, N 6.08.
cooling the reaction mixture to 25 °C, MTBE (4.8 L) and a
solution of K3PO4 (1.28 kg) in water (12 L) was added. The
addition resulted in three layers. The bottom two layers were
separated and extracted with MTBE (4.8 L). The combined
MTBE layers were extracted with a solution of K3PO4 (0.43
kg) dissolved in water (6.5 L). Analysis of the MTBE layer
showed 1.27 kg of chloroamine 26 (97% yield). The solution
was concentrated to an oil, which was dissolved in NMP (5
kg) prior to the final coupling reaction. Compound 26 was
characterized as the hemi-L-tartrate salt (2:1 base to salt ratio),
which was formed in and crystallized from MeOH (5 mL/g).
1H NMR (400 MHz, DMSO-d6) δ 1.24 (s, 3 H), 1.53 (s, 9 H),
1.59 (t, J ) 5.00 Hz, 4 H), 3.41-3.52 (m, 2 H), 3.74-3.83
(m, 3 H), 6.88 (d, J ) 1.00 Hz, 1 H), 7.11 (d, J ) 1.00 Hz, 1
H). 13C NMR (101 MHz, DMSO-d6) δ 25.47, 27.49, 35.82,
40.87, 49.85, 71.15, 82.45, 105.04, 109.50, 143.44, 149.25,
158.88, 163.44, 174.80. IR (KBr) 1724 cm-1. Anal. Calcd for
C36H54Cl2N6O10: C 53.93, H 6.79, N 10.48; found: C 53.84, H
6.85, N 10.45.
Preparation of tert-Butyl 2-(4-amino-4-methylpiperidin-
1-yl)isonicotinate (2). To a Parr hydrogenation vessel under
nitrogen was charged 5% Pd/Al2O3 (132 g, 10 wt % load) and
K3PO4 (908 g, 4.3 mol, 1.05 equiv) followed by an NMP
solution of 26 (4.17 kg solution, 31.8 wt %, 1.33 kg active, 4.1
mol, 1.0 equiv). NMP (650 mL) was used to rinse the product
flask and was added to the reaction vessel. The vessel was
sealed, sparged with hydrogen, and warmed to 40 °C with
shaking. An HPLC sample after 3 h indicated <0.1% of 26
remained (see HPLC conditions above). The vessel was sparged
with N2, cooled to ambient, and filtered. The solution assayed
for 1.14 kg of 2 (96% yield).
Preparation of tert-Butyl 2,6-dichloroisonicotinate (25).
A mixture of 2,6-dichloroisonicotinic acid 24 (875 g, 4.56 mol,
1.0 equiv), DMAP (110 g, 0.90 mol, 0.2 equiv), and NMP (2.4
kg) was stirred for 15 min to dissolve the solids. Boc2O (2.0
kg, 9.3 mol, 2.1 equiv) was charged as a melt. Triethylamine
(0.45 kg, 4.5 mol, 1.0 equiv) was charged over 15 min while
maintaining the temperature at 25 ( 10 °C. After 19 h, HPLC
analysis indicated 95% conversion. HPLC conditions: Column:
Zorbax, Eclipse XDB-C8, 4.6 mm × 250 mm, 5 µm, 35 °C, 1
mL/min, 10 µL injection, λ 210 nm; Mobile phase A: 0.1%
H3PO4; Mobile phase B: ACN; Gradient: 0 min 5% B, 14 min
55% B, 20 min 90% B, 30 min 90% B; Retention time (min):
2 (10.4), 24 (13.8), 26 (17.2), 25 (21.2). The reaction mixture
was diluted with MTBE (6 L) and water (8 L) containing
KH2PO4 (0.84 kg, 6.2 mol, 1.4 equiv). After cooling to 30 °C
the aqueous phase was separated, and the MTBE layer was
filtered through a Hyflo bed to remove insoluble material. The
MTBE layer was then extracted with water (2 × 6 L). To the
MTBE layer was charged Darco-G60 (85 g) and the mixture
was stirred for 1 h and filtered through a Hyflo pad. The Darco-
G60 treatment was necessary to remove color and extraneous
material present in some of the vendor lots of 24. The MTBE
was distilled under vacuum to a volume of approximately 5 L.
The distillation was continued, maintaining the volume by the
addition of MeOH (8 L). The slurry in MeOH was warmed to
50 °C to dissolve the solids and then cooled to RT to crystallize.
Water (1.5 L) was added over 30 min. The slurry was filtered
and washed with a mixture of MeOH (5.0 L) and water (2.0
L) in 2 portions. The cake was dried with heat and vacuum,
1
resulting in 0.97 kg of light-brown powder (88% yield). H
Acknowledgment
NMR (400 MHz, CDCl3) δ 1.59 (s, 9 H) 7.71 (s, 2 H). 13C
NMR (101 MHz, CDCl3) δ 28.16, 83.76, 122.28, 143.98,
150.76, 161.03. IR (MIC) 2977, 1730, 1546 cm-1. Anal. Calcd
for C10H11Cl2NO2: C 48.41, H 4.47, N 5.65; found: C 48.46, H
4.42, N 5.58.
We gratefully acknowledge Nancy Benz, Qunying Zhang,
Michael Oberlander, Jacqueline Knue, and Shelly Keaton for
analytical support, and Chun (Casey) Zhou, Ian Marsden, Steve
Hollis, and Paul West for spectroscopic analysis and interpretation.
Preparation of tert-Butyl 2-(4-amino-4-methylpiperidin-
1-yl)-6-chloroisonicotinate (26). To a mixture of dichloropy-
ridine 25 (1.0 kg, 4.0 mol, 1.0 equiv) and salt 23 (2.03 kg, 4.4
mol, 1.1 equiv) was added NMP (4.1 kg). To the well-stirred
reaction mixture was added K3PO4 (1.8 kg, 8.5 mol, 2.1 equiv)
and the mixture was warmed to 80 °C for 12 h. HPLC analysis
indicated 97% conversion (see HPLC conditions above). After
Supporting Information Available
1H and 13C NMR spectra. This material is available free of
charge via the Internet at http://pubs.acs.org.
Received for review July 30, 2009.
OP900197R
Vol. 13, No. 6, 2009 / Organic Process Research & Development
•
1155