An enantioselective synthesis of (R)-4-piperidinylglycine

Article abstract of DOI:10.1016/S0957-4166(01)00425-6

An efficient process for the synthesis of (R)-N-t-Boc-4-piperidinylglycine 8a, an unnatural amino acid, via enantioselective rhodium-catalyzed hydrogenation of the Cbz-enamide 5a is described. Subsequent deprotection of 8a affords unprotected (R)-4-piperi

Full text of DOI:10.1016/S0957-4166(01)00425-6

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 12 (2001) 2421–2425
An enantioselective synthesis of (R)-4-piperidinylglycine
Wen-Chung Shieh,a,* Song Xue,a Noela Reel,a Raeann Wu,a John Fittb and Oljan Repicˇa
aChemical and Analytical Development, Novartis Institute for Biomedical Research, East Hanover, NJ 07936, USA
bArthritis and Bone Metabolism Chemistry Research, Novartis Institute for Biomedical Research, Summit, NJ 07901, USA
Received 31 July 2001; accepted 18 September 2001
Abstract—An efficient process for the synthesis of (R)-N-t-Boc-4-piperidinylglycine 8a, an unnatural amino acid, via enantioselec-
tive rhodium-catalyzed hydrogenation of the Cbz-enamide 5a is described. Subsequent deprotection of 8a affords unprotected
(R)-4-piperidinylglycine 9 in good yield. © 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
an alternative synthesis of enantiomerically pure (R)-N-
t-Boc-4-piperidinylglycine 8a.
(R)-4-Piperidinylglycine 9, an unnatural amino acid, is
an important building block for the synthesis of novel
pharmaceutical drugs, such as matrix metalloproteinase
inhibitors1 and thrombin inhibitors.2 Racemic
piperidinylglycine2 was synthesized by hydrogenation of
an enamide substrate whose structure is similar to 5a.
Enantiomerically pure (R)-N-t-Boc-4-piperidinylglycine
8a, a derivative of 9, can be prepared in 8 steps by
employing Evans’ asymmetric azidation3 with (R)-(+)-
4-benzyl-2-oxazolidinone 1 as the chiral auxiliary and
trisyl azide as the electrophile (Scheme 1).1 This strat-
egy works very well for the preparation of small quanti-
ties of 8a. On scale up, we found this approach lengthy
and unsafe due to the handling of large quantities of
trisyl azide, which liberated a significant amount of
energy upon heating. To address this safety issue and to
reduce the number of synthetic steps, we investigated
Syntheses of O- and S-containing heterocyclic, unnatu-
ral amino acids with high enantiomeric excess by
rhodium-catalyzed asymmetric hydrogenation have
been reported by Burk.4 In his study, N-acetyl-enam-
ides were used as the hydrogenation precursors. Excel-
lent enantiomeric purity has been reported in the
synthesis of heterocyclic amino acids containing N-Cbz
protected enamides.5,6 To synthesize (R)-N-t-Boc-4-
piperidinylglycine 8a, an N-containing heterocyclic
amino acid, we decided to investigate asymmetric
hydrogenation utilizing N-Cbz-enamide 5a. This would
allow selective deprotection after the asymmetric hydro-
genation and avoid the harshly acidic conditions, which
are incompatible with the Boc-protecting group, but are
required for the N-acetyl group removal.7 Herein,
we report the synthesis of (R)-N-t-Boc-4-piperidinyl-
O
O
O
O
O
H2N
N3
OH
N
O
N
O
trisyl azide
KHMDS
Bn
Bn
N
Boc
N
Boc
N
Boc
2
1
8a
Scheme 1. Evans asymmetric azidation and amino acid synthesis.
* Corresponding author. E-mail: wen.shieh@pharma.novartis.com
0957-4166/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(01)00425-6

2422
W.-C. Shieh et al. / Tetrahedron: Asymmetry 12 (2001) 2421–2425
glycine 8a and (R)-4-piperidinylglycine 9 via a highly
enantioselective, rhodium-catalyzed hydrogenation reac-
tion.
by HPLC) substrate. The high purity ensured the elim-
ination of any contaminants that may poison the
rhodium catalyst.7b Asymmetric hydrogenation of enam-
ide 5a in the presence of [(R,R)-Me-BPE-Rh-COD]OTf
catalyst (2 mol%) at 90 psi generated the (R)-enantiomer9
of protected 4-piperidinylglycine 6a with 94% e.e. in
99.8% yield. Asymmetric hydrogenation of 5a with two
other commercially available rhodium catalysts led to the
amino acid with much lower enantiomeric purity: [(R,R)-
Me-DuPHOS-Rh-COD]OTf gave the (R)-enantiomer
with 60% e.e. and [(R,R)-DiPAMP-Rh-COD]BF4
afforded the (S)-enantiomer with 20% e.e. (Scheme 2).
2. Results and discussion
Cbz-enamide 5a was prepared from commercially avail-
able N-Cbz-phosphonoglycine trimethyl ester 3 and
N-t-Boc-4-piperidone 4a in 61% yield using the Schmidt
protocol.8 Enamide 5a was recrystallized from a mixture
of ethyl acetate and hexane to obtain a highly pure (>99%
O
O
H
N
tetramethyl-
guanidine
O
O
O
O
R
O
P
O
O
O
+
N
O
N
H
R
O
O
O
N
3
4a, R = t-Bu
4b, R = i-Pr
O
O
5a, R = t-Bu
5b, R = i-Pr
Other catalysts used:
P
P
+
O
O
Rh
P
P
P
P
-OTf
+
+
Rh
Rh
[(R,R)-Me-BPE-Rh-COD]OTf
-BF4
-OTf
H2
[(R,R)-DiPAMP-Rh-COD]BF4
[(R,R)-Me-DuPHOS-Rh-COD]OTf
O
H
O
H
O
N
O
N
O
R
OH
R
LiOH
O
O
N
N
O
O
O
O
6a, R = t-Bu
6b, R = i-Pr
7a, R = t-Bu
7b, R = i-Pr
H2/Pd/C
O
O
H2N
+
− H3N
OH
Cl
OH
HCl
N
+
Cl−
H
N
R
H
O
O
8a, R = t-Bu
8b, R = i-Pr
9
Scheme 2. Asymmetric synthesis of (R)-4-piperidinylglycine.

W.-C. Shieh et al. / Tetrahedron: Asymmetry 12 (2001) 2421–2425
2423
4. Experimental
Saponification of the protected piperidinylglycine 6a
with aqueous LiOH solution produced carboxylic acid
7a in 98% yield. Chiral-HPLC analysis of 7a indicated
a slightly lower enantiomeric purity (90% e.e.), which
suggested partial epimerization of amino ester 6a
under the hydrolysis conditions. Further hydrogenoly-
sis of 7a to remove the Cbz-protecting group was
achieved via the standard protocol (H2, Pd/C, 50 psi,
rt) to obtain the crude (R)-N-t-Boc-4-piperidinyl-
glycine 8a in quantitative yield. The enantiomeric
purity of the crude amino acid 8a was enriched by
stirring it in methanol, then collecting the solid by
filtration to furnish the desired (R)-N-t-Boc-4-pipe-
ridinylglycine 8a. This extraction procedure resulted in
excellent enantiomeric purity (98% e.e.) and an 82%
yield. The augmentation of the enantiomeric excess
was caused by a lower solubility of 8a in methanol
than that of its racemate.10 Furthermore, (R)-pipe-
ridinylglycine 9 can be obtained easily in 79% yield by
removing the t-Boc group from 8a with a solution of
HCl generated in situ with trimethylsilyl chloride and
methanol.11
N-a-Cbz-phosphonoglycine trimethyl ester 3 was pur-
chased from Aldrich and N-Boc-4-piperidone 4a from
Lancaster. Chiral rhodium catalysts: [(R,R)-Me-BPE-
Rh-COD]OTf, [(R,R)-Me-DuPHOS-Rh-COD]OTf and
[(R,R)-DiPAMP-Rh-COD]BF4 were purchased from
Strem Chemicals Inc. Proton magnetic resonance spec-
tra were recorded on either a Bru¨ker ARX300,
DPX300 or a DRX500 FT-NMR spectrometer. Chiral
HPLC analyses were performed on a Waters HPLC
system with a 996 PDA detector and Daicel columns.
All the elemental analyses were performed by Robert-
son Microlit Labs (Madison, NJ).
4.1. 4-(Benzyloxycarbonylamino-methoxycarbonyl-meth-
ylene)-piperidine-1-carboxylic acid tert-butyl ester 5a
A 12-L, four-necked, round-bottomed flask equipped
with a mechanical stirrer and 1-L pressure equalizing
addition funnel was charged with N-a-Cbz-phosphono-
glycine trimethyl ester 3 (470 g, 1.42 mol) and EtOAc
(2.5 L) under nitrogen purge. Tetramethylguanidine
(214 g, 1.84 mol) was added and the solution was
stirred for 30 min at rt. A solution of piperidone 4a
(424 g, 2.13 mol) in EtOAc (2.5 L) was added via the
addition funnel over 60 min. The solution was stirred
at rt for 42 h until 3 was consumed (HPLC). The
mixture was cooled to 10°C and 5% aqueous citric
acid solution (2.8 L) was added. The mixture was
stirred for 15 min and the aqueous layer was sepa-
rated. The organic solution was washed with 0.25N
HCl (2.8 L), H2O (2.0 L), dried over MgSO4, filtered
and concentrated under vacuum to give an oil. The oil
was dissolved in EtOAc (0.5 L) and stirred with hex-
ane (2.0 L) in order to initiate precipitation. The crude
product was collected by filtration and recrystallized
from a mixture of ethyl acetate:hexane (1:4, 2.0 L) to
yield 5a as a white solid (353 g, 61%): mp 101.5–
102.6°C; IR (KBr) 3312, 2972, 1725, 1703, 1684, 1512,
1477, 1451, 1426, 1365, 1327, cm−1; 1H NMR (500
MHz, CDCl3): l 7.49–6.30 (m, 5H), 6.17 (br s, 1H),
5.13 (s, 2H), 3.76–3.68 (m, 3H), 3.56–3.40 (m, 4H),
2.92–2.78 (m, 2H), 2.39 (t, J=5.9 Hz, 2H), 1.47 (s,
9H); 13C NMR (125 MHz, CDCl3): l 165.4, 155.1,
147.6, 136.4, 128.9, 128.8, 128.7, 128.6, 120.6, 80.2,
67.8, 53.3, 44.2, 43.43, 30.8, 30.1, 28.9, 28.8. Anal.
calcd for C21H28N2O6: C, 62.36; H, 6.98; N, 6.93.
Found: C, 62.43; H, 7.07; N, 6.83%.
Asymmetric hydrogenation of the isopropyl carbamate
analogue 5b using [(R,R)-Me-BPE-Rh-COD]OTf at
45°C and a hydrogen pressure of 70 psi produced the
protected amino acid 6b with 84% e.e. Subsequent
deprotection of the methyl ester and Cbz groups
afforded
N-4-isopropoxycarbonyl-piperidinylglycine
8b. An attempt to enrich the enantiomeric purity of 8b
by stirring it in methanol (same procedure as used for
8a) was unsuccessful and resulted in a much lower e.e.
(57%). On the other hand, enriched enantiomeric
purity (95% e.e.) was found in the mother liquor of
8b, suggesting higher solubility of the enantiomer 8b in
methanol. These results demonstrate that a small vari-
ation in functionality within piperidinylglycine causes
a reversal in its solubility, and consequently its enan-
tiomeric excess. The nature of the enantiomeric mix-
ture determines whether or not it can be extracted
with solvent to increase its enantiomeric purity.10
3. Conclusion
In conclusion, we have developed a highly efficient
process for the synthesis of (R)-N-t-Boc-4-piperidinyl-
glycine 8a, via rhodium-catalyzed asymmetric hydro-
genation, in four steps and 49% overall yield starting
from commercial N-a-Cbz-phosphonoglycine trimethyl
ester 3 and N-Boc-4-piperidone 4a. The enantiomeric
purity of 8a can be enriched from 90 to 98% e.e. with
excellent recovery (82%) by a simple extraction proce-
dure.10 For the synthesis of amino acid 9 via asymmet-
ric hydrogenation, we found t-butoxycarbonyl
enamide 5a to be the preferred prochiral substrate
because it provided a much higher enantiomeric purity
than isopropoxycarbonyl enamide 5b. Our synthetic
sequence is four steps shorter than the previously
reported1 synthesis and has been employed effectively
for the synthesis of a large quantity (ꢀ500 g) of
amino acid 8a.
4.2. 4-((R)-Benzyloxycarbonylamino-methoxycarbonyl-
methyl)-piperidine-1-carboxylic acid tert-butyl ester 6a
A 5-L high-pressure bottle was charged with enamide
5a (60 g, 148 mmol) and previously degassed MeOH
(3.0 L) under a nitrogen purge. This colorless solution
was charged with [(R,R)-Me-BPE-Rh-COD]OTf cata-
lyst (2 g, 3 mmol). The resulting mixture was subjected
to a vacuum, then refilled with nitrogen for three
cycles. The mixture was placed under vacuum and
refilled with hydrogen for an additional three cycles.
The solution was stirred under 90 psi of hydrogen gas
at rt for 72 h. The mixture was flushed with nitrogen

2424
W.-C. Shieh et al. / Tetrahedron: Asymmetry 12 (2001) 2421–2425
and concentrated under vacuum. The residue was dis-
solved in ethyl acetate (500 mL) and filtered through
silica gel (50 g) to remove the catalyst. The cake was
rinsed with ethyl acetate (500 mL). The organic solu-
tion was concentrated under vacuum to afford 6a as an
oil (60 g 100%): Rf=0.36 (hexane/EtOAc 1/1); [h]D25
−20.7 (c=1.05, CHCl3); IR (KBr) 3323, 2950, 1691,
2 h, cooled to 0°C, and stirred for an additional 1 h.
The solid was collected by filtration and rinsed with
cold (5°C) MeOH (200 mL). The solid was dried at
60°C under vacuum (2 mm Hg) to obtain 40 g (82%) of
1
8a, as an off-white solid: [h]2D5 −4.2 (c=0.51, H2O); H
NMR (500 MHz, D2O): l 4.18 (m, 2H), 3.68 (d, J=4.9
Hz, 1H), 2.76–2.92 (m, 2H), 2.09–2.21 (m, 1H), 1.74–
1.83 (m, 1H), 1.64–1.72 (m, 1H), 1.48 (s, 9H), 1.25–1.53
(m, 2H); 13C NMR (125 MHz, CDCl3): l 173.2, 156.4,
81.7, 58.9, 43.6, 37.0, 27.6, 26.9. Anal. calcd for
C12H22N2O4: C, 55.80; H, 8.58; N, 10.84. Found: C,
55.67; H, 8.35; N, 10.79. Chiral HPLC: 98% e.e.
(Crownpak CR+, perchloric acid pH 1.5/MeOH 85/15,
flow rate 1 mL/min; (R)-enantiomer, Rt=14.0 min;
(S)-enantiomer, Rt=23.4 min).12
1
1529 cm−1; H NMR (300 MHz, CDCl3) 7.32–7.42 (m,
5H), 5.32 (d, J=8.7 Hz, 1H), 5.12 (s, 2H), 4.33–4.44
(m, 1H), 4.08–4.21 (m, 2H), 3.77 (s, 3H), 2.57–2.76 (m,
2H), 1.85–2.02 (m, 1H), 1.40–1.70 (m, 2H), 1.46 (s, 9H),
1.19–1.39 (m, 2H); 13C NMR (125 MHz, CDCl3) 171.8,
156.0, 154.6, 136.0, 128.5, 128.2, 128.1, 79.5, 67.1, 57.8,
52.3, 43.4, 39.6, 28.3, 27.1. Anal. calcd for C21H30N2O6:
C, 62.05; H, 7.44; N, 6.89. Found: C, 61.99; H, 7.09; N,
7.04. Chiral HPLC: 94% e.e. (Chiralcel OD column,
hexane/IPA/TFA 9/1/0.1%, flow rate 1.5 mL/min; (S)-
enantiomer, Rt=7.21 min; (R)-enantiomer, Rt=10.0
min).12
4.5. (R)-Amino-piperidin-4-yl-acetic acid dihydrochlor-
ide 9
A 50-mL flask equipped with a magnetic stirrer was
charged with Boc-protected amino acid 8a (0.16 g, 0.6
mmol) and MeOH (15 mL). Trimethylsilyl chloride (2.0
g, 18.4 mmol) was added to the suspension in one
portion. The resulting solution was stirred at rt for 3 h.
The reaction mixture was concentrated at 25°C under
vacuum (2 mm Hg) to give 9 as a foamy white solid
(0.11 g, 79%): [h]D25 −18.3 (c=1.05, H21O); IR (KBr)
3416, 2926, 1740, 1597, 1512, 1215 cm−1; H NMR (300
MHz, D2O): l 3.80 (d, J=4.8 Hz, 1H), 3.33–3.48 (m,
2H), 2.84–3.01 (m, 2H), 2.10–2.29 (m, 1H), 1.81–2.01
(m, 2H), 1.58–1.78 (m, 1H), 1.38–1.57 (m, 1H); MS:
m/z 157 (M+−1). Anal. calcd for C7H16Cl2N2O2: C,
36.38; H, 6.98; N, 12.12. Found: C, 36.96; H, 7.38; N,
11.63%.
4.3. 4-((R)-Benzyloxycarbonylamino-carboxy-methyl)-
piperidine-1-carboxylic acid tert-butyl ester 7a
A 5-L flask equipped with a mechanical stirrer was
charged with amino ester 6a (151 g, 371 mmol) and
MeOH (1.7 L). The solution was cooled to 5°C. A
solution of 1N LiOH (557 mL, 557 mmol) was added
and the mixture was allowed to warm to rt, then stirred
for 20 h. The reaction mixture was neutralized to pH 7
with 1N KHSO4 solution and concentrated under vac-
uum. The resulting aqueous solution was adjusted to
pH 2 with 2N KHSO4 and extracted with EtOAc
(3×800 mL). The combined organic layers were washed
with brine (1 L), dried over MgSO4, filtered through
Celite, and concentrated under vacuum to give 7a as a
foamy white solid (142 g, 98%): [h]2D5 −18.6 (c=1.07,
CHCl3); IR (KBr) 3327, 2977, 2931, 1695, 1531, 1479,
1
Acknowledgements
1367, 1243, 1164 cm−1; H NMR (500 MHz, CDCl3): l
8.30–9.20 (br s, 1H), 7.30–7.40 (m, 5H), 5.49 (d, J=8.7
Hz, 1H), 5.13 (s, 2H), 4.34–4.50 (m, 1H), 4.03–4.30 (m,
2H), 2.57–2.80 (m, 2H), 1.95–2.09 (m, 1H), 1.46 (s, 9H),
1.22–1.78 (m, 4H); 13C NMR (125 MHz, CDCl3): l
175.1, 156.7, 155.4, 136.4, 129.0, 128.7, 128.5, 80.5,
67.7, 58.1, 44.0, 39.7, 28.8, 27.3. Anal. calcd for
C20H28O6N2: C, 61.21; H, 7.19; N, 7.14. Found: C,
60.98; H, 6.93; N, 7.14%. Chiral HPLC: 90% e.e.
(Chiralcel OD, hexane/IPA/TFA 9/1/0.1%, flow rate
1.5 mL/min; (S)-enantiomer, Rt=5.72 min; (R)-enan-
tiomer, Rt=8.54 min).12
We are indebted to our colleagues R. Goldstein, W.
Macchia, and L. Zhou for providing a variety of valu-
able samples during our chiral HPLC methodology
development. We thank Drs. M. Prashad and L.
Waykole for helpful discussions on asymmetric hydro-
genation. We are grateful to Ms. L. Liu and Ms. M.
Mehta for obtaining optical rotations and IR spectra.
1
The high-field H NMR spectra acquired on a Bruker
DRX500 (500 MHz) were performed by Mr. R. Beve-
ridge. We thank Mr. R. DuVal for performing a safety
assessment of trisyl azide.
4.4. 4-((R)-Amino-carboxy-methyl)-piperidine-1-car-
boxylic acid tert-butyl ester 8a
A 2.5-L Parr bottle was charged with 5% Pd/C (50%
wet, 16.0 g) under a nitrogen atmosphere. A solution of
amino acid 7a (75 g, 190 mmol) in EtOH:H2O (2:1, 1.5
L) was added under a nitrogen purge. The mixture was
hydrogenated under 50 psi hydrogen gas at rt for 16 h.
The mixture was flushed with nitrogen and filtered. The
catalyst cake was rinsed with EtOH (400 mL). The
organic solution was concentrated under vacuum until
a grey solid was obtained (49 g, 100%). The grey solid
was suspended in MeOH (750 mL), stirred at 60°C for
References
1. Nantermet, P.; Parker, D. T.; MacPherson, L. J. Certain
alpha-azacycloalkyl substituted arylsulfonamido acetohy-
droxamic acids, US Patent 5,817,822.
2. Plummer, J. S.; et al. Bioorg. Med. Chem. Lett. 1998,
3409–3414.
3. Evans, D. A.; Evrard, D. A.; Rychnovsky, S. D.; Fru¨h,
T.; Whittingham, W. G.; DeVries, K. M. Tetrahedron
Lett. 1992, 33, 1189.

W.-C. Shieh et al. / Tetrahedron: Asymmetry 12 (2001) 2421–2425
2425
4. Burk, M. J.; Gross, M. F.; Martinez, J. P. J. Am. Chem.
Soc. 1995, 117, 9375–9376.
5. Adamczyk, M.; Akireddy, S. R.; Reddy, R. E. Org. Lett.
2000, 2, 3421–3423.
6. Tiffin, P. D.; Jones, S. W.; Palmer, C. F.; Paul, J. M.
Tetrahedron Lett. 1999, 40, 1211–1214.
7. (a) Burk, M. J.; Feaster, J. E.; Nugent, W. A.; Harlow, R.
L. J. Am. Chem. Soc. 1993, 115, 10125–10138; (b) Krause,
H.-W.; Taudien, S.; Schinkowski, K. Tetrahedron: Asym-
metry 1993, 4, 73–84.
8. Schmidt, U.; Griesser, H.; Leitenberger, V.; Liebreger, V.;
Mangold, R.; Meyer, R.; Riedl, B. Synthesis 1992,
487–490.
9. Absolute configuration was assigned by comparing the sign
of the optical rotation of 8a, synthesized from 6a, with that
of an authentic sample of (R)-N-t-Boc-4-piperidinylglycine
8a prepared according to the literature procedure.1
10. For a discussion of techniques in enantiomeric purity
enrichment, see: Wilen, S. H.; Collet, A.; Jacques, J.
Tetrahedron 1977, 33, 2725–2736.
11. Kaiser, Sr., E.; Tam, J. P.; Kubiak, T. M.; Merrifield, R.
B. Tetrahedron Lett. 1988, 29, 303.
12. HPLC conditions used for enantiomer separation of
racemic product. The retention times of individual enan-
tiomers were assigned by comparison with an authentic
sample of 8a.