Studies directed towards asymmetric synthesis of levobupivacaine

Article abstract of DOI:10.1016/j.tetlet.2004.11.039

We report herein the first catalytic asymmetric synthesis of levobupivacaine. The key step involves the asymmetric alkylation of N-benzylimine glycinamide (2b).

Full text of DOI:10.1016/j.tetlet.2004.11.039

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 19–21
Studies directed towards asymmetric synthesis of levobupivacaine
Sanjeev Kumar and Uma Ramachandran*
Department of Pharmaceutical Technology, National Institute of Pharmaceutical Education & Research (NIPER), Sector-67,
Mohali, Punjab 160 062, India
Received 7 October 2004; revised 3 November 2004; accepted 8 November 2004
Abstract—We report herein the first catalytic asymmetric synthesis of levobupivacaine. The key step involves the asymmetric alkyl-
ation of N-benzylimine glycinamide (2b).
ꢀ 2004 Elsevier Ltd. All rights reserved.
Bupivacaine is currently the most widely used, long act-
ing local anaesthetic agent in both surgery and obstet-
rics. Though it has a good safety record, it can result
in fatal cardiotoxicity if given intravenously by mistake
or if used in excessive dose by other routes.1 Bupivacaine
exhibits stereoisomerism and recent studies comparing
the two enantiomers have shown that R-(+)-bupivacaine
is 3–4 times more likely to cause cardiovascular toxicity
than S-(ꢀ)-bupivacaine in rabbit hearts.2 While the S-
enantiomer of bupivacaine, also known as levobupiva-
caine (1), retains the anaesthetic activity of the racemate,
it produces less CNS side-effects. It also has wider appli-
cation as it has FDA approval for post operative pain
management.3
Our laboratory has been studying asymmetric alkylation
of various imine glycinamides using chiral phase-trans-
fer catalysis.7 Earlier we found that alkylation of imine
glycinamide 2a using chiral phase-transfer catalysts
O(9)-allyl-N-(9-anthracenylmethyl)cinchonidinium bro-
mide (3, Fig. 1), gave very low enantioselectivty (20%,
Table 1, entry 1).
We explained, through molecular modeling studies, that
the reason for this low enantioselectivity was the hydro-
gen bonding possibility between the NH group of the
glycinamide 2a enolate and the nitrogen atom of the
Levobupivacaine is obtained mainly by diastereomeric
resolution of bupivacaine or its synthetic intermediates
like pipecolic acid using D-tartaric acid4 and/or D-di-
benzoyl tartaric acid.5 An alternative method uses
(S)-lysine as a synthon.6 Herein we report a novel and
efficient catalytic route to levobupivacaine.
H
R
Br
Br
N
O
N
H
R
N
Retrosynthetic analysis of levobupivacaine indicates
that it could be synthesized from benzophenone imine
glycinamide 2a (Scheme 1).
4 (R = 3,4,5-F3-Ph)
3
Figure 1.
H
N
H
N
H
N
Ph
Ph
N
N
Bu
X
HN
R'
O
2a
O
1
O
Scheme 1. Retrosynthetic analysis of levobupivacaine.
Keywords: Chiral phase-transfer catalysts; Asymmetric synthesis; Alkylations.
*
Corresponding author. Tel.: +91 1722214682 87; fax: +91 1722214692; e-mail: umaram54@yahoo.com
0040-4039/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.11.039

20
S. Kumar, U. Ramachandran / Tetrahedron Letters 46 (2005) 19–21
Table 1. Alkylation with 1-chloro-4-iodobutane catalyzed by 3 or 4
Entry Substrate (R)
Basea
Catalyst
Solvent
CH2Cl2
Temp (ꢁC)
ꢀ50
Time (h) % eeb
Yieldc (%)
1
2
3
4
5
6
2a (H)
2b (Bn)
2b (Bn)
2a (H)
2b (Bn)
2b (Bn)
CsOHÆH2O
CsOHÆH2O
CsOHÆH2O:K2CO3 (2:10) 3 (10mol%)
CsOHÆH2O
CsOHÆH2O
CsOHÆH2O:K2CO3 (2:10) 4 (1mol%)
3 (10mol%)
3 (10mol%)
18
19
25
20
18
22
20
64
64
43
86
96
60, 5a
58, 5b
78, 5b
63, 5a
57, 5b
85, 5b
Toluene:CH2Cl2 (7:3) ꢀ40
Toluene:CH2Cl2 (7:3) ꢀ40
Toluene:CH2Cl2 (7:3) ꢀ40
Toluene:CH2Cl2 (7:3) ꢀ40
4 (1mol%)
4 (1mol%)
Toluene
ꢀ40
a 12equiv.
b Determined using a Chiralcel OD-H column.
c Isolated products.
NHBn
Cl
O
Bn
N
Cl
i
ii, iii
Cl
R
N
Cl
+
R
N
ii
Ph
Ph
O
i
Ph
Ph
O
N
N
O
Bn
N
Bn
N
5
2
iv
Ph
Ph
a) R = H, b) R = Bn.
a) R = H, b) R = Bn.
N
HCl.H2N
O
2b
O
Bn
H
iv
iii
1
N
N
N
N
Scheme 2. Reagents and conditions: (i) 0–30ꢁC, toluene, 4h, 92%; (ii)
NaN3, acetone, 20h, 93%; (iii) (a) H2, Pd–C; (b) Et2O, HCl (dry), 98%;
(iv) Ph2C@NH, CH2Cl2, rt, 93%.
H
O
O
Ph
Ph
7
6
Scheme 3. Reagents and conditions: (i) I(CH2)4Cl, 3 or 4, base,
solvent; (ii) (a) NaCNBH3, THF; (b) NaHCO3, NaI, CH3CN, 92%;
(iii) PdCl2, EtOAc–AcOH (4/1), 40psi, rt, 6h, 95%; (iv) BuBr, K2CO3,
94%.
quinoline nucleus of the catalyst 3, which stabilized the
ion pair leading to formation of the unwanted isomer.7
To circumvent the possibility of hydrogen bonding,
we modified the substrate by introducing a benzyl
group onto the amide nitrogen making it a tertiary
amide as shown in Scheme 2. As anticipated,
reaction with this substrate 2b while using the cata-
lyst O(9)-allyl-N-(9-anthracenylmethyl)cinchonidinium
bromide (3), resulted in an increase in the enantioselec-
tivity of the alkylated product 5b to 64% (Table 1,
entry 2).8
compared to a mixture of solvents such as toluene–
dichloromethane (7/3). This indicates that the catalyst
4 exhibits greater asymmetric induction in nonpolar
medium. Further, we found that a combination of CsO-
HÆH2O and K2CO3 increased the yield of the product
from 57% to 85%. Thus the conditions for the synthesis
of 5b, were optimized to obtain a high yield and excel-
lent stereoselectivity.
We next used (R,R)-3,4,5-trifluorophenyl-NAS bromide
(4, Fig. 1),9 which has a different structural motif, as the
chiral phase-transfer catalyst for the asymmetric alkyl-
ation of the imine glycinamide 2a.
Next, the alkylated intermediate 5b was reduced with
sodium cyanoborohydride in tetrahydrofuran at 0ꢁC,
followed by cyclization at 60ꢁC in the presence of
NaHCO3 and a catalytic amount of NaI in acetonitrile
to afford 6 in 92% yield. Amine deprotection and amide
debenzylation was achieved in one-pot by hydrogenoly-
sis using PdCl2 in EtOAc–AcOH (4/1)11 to give 7 in 95%
yield. This was finally alkylated with n-butyl bromide to
complete the asymmetric synthesis of levobupivacaine
(Scheme 3).12
Binaphthyl derived catalysts such as (R,R)-3,4,5-trifluo-
rophenyl-NAS bromide (4) were reported by Maruokaꢀs
group.10 The advantages of using the above catalysts is
their higher stability, high asymmetric inductions and
the use of only 1mol% of the catalyst. With the catalyst
4 the asymmetric alkylation of imine glycinamide 2a
gave the alkylated product 5a in moderately good
enantioselectivity (43%) as compared to 20% ee with
catalyst 3 (Table 1, entries 1 and 4).
In conclusion, we have achieved the first asymmetric
synthesis of levobupivacaine in high yield and enantio-
purity through a novel catalytic route.
When N-benzylated imine glycinamide 2b was alkylated
in the presence of catalyst 4 and CsOHÆH2O at ꢀ40ꢁC,
the enantiomeric excess of the alkylated product in-
creased steeply to 86% (Table 1). Thus even here, the
N-benzylated imine glycinamide was found to be more
suitable as a substrate for asymmetric alkylation. The
use of toluene as solvent gave higher selectivity (98:2),
Acknowledgements
This work was funded by a grant from the National
Institute of Pharmaceutical Education & Research
(NIPER).

S. Kumar, U. Ramachandran / Tetrahedron Letters 46 (2005) 19–21
21
dichloromethane (3 · 25mL) and the combined organics
dried over MgSO4. The extract was concentrated in vacuo
and the residue was passed through a short pad of silica
gel to afford 5b as an oil (0.513g, 85%). ½aꢁD : +56.53 (c 0.5,
References and notes
1. Kopacz, D. J.; Allen, H. W. Anesth. Analg. 1999, 89,
1027–1029.
2. Mazoit, J. X.; Boico, O.; Samii, K. Anesth. Analg. 1993,
77, 477–482.
3. (a) Gristwood, R.; Bardsley, H.; Baker, H.; Dickens, J.
Exp. Opin. Invest. Drugs 1994, 3, 1209–1212; (b) McLure,
H. A.; Rubin, A. P. Anaesthesia 1998, 53, 1160–
1164.
4. Tuller, B. F. J. Med. Chem. 1971, 14, 891–892.
5. Federsel, H.; Jaksch, P.; Sanberg, R. Acta Chem. Scand.
1987, B41, 757–761.
6. (a) Hutton, G. E. W.O. Patent 9 611 181, 1996; Chem.
Abstr. 1996, 125, 115135q; (b) Langston, M.; Skead, B. M.
W.O. Patent 9 612 699, 1996; Chem. Abstr. 1996, 125,
114491r; (c) Adger, B.; Dyer, U.; Hutton, G.; Woods, M.
Tetrahedron Lett. 1996, 37, 6399–6402.
25
CHCl3); ee 96%; IR (KBr): 3053, 2958, 2932, 2854, 1655,
1444, 1400, 1312, 1286, 1254, 1233, 1188, 1079, 772; 1H
NMR d (CDCl3, 300MHz): 1.02–1.07 (m, 2H), 1.31 (m,
1H), 1.49 (s, 3H), 1.48 (m, 2H), 1.92 (s, 3H), 2.3 (m, 1H),
3.34 (t, J = 6.61Hz, 2H), 3.64 (dd, J = 2.79Hz, 7.02, 1H),
4.22 (d, J = 13.49Hz, 1H), 5.34 (d, J = 13.49Hz, 1H), 6.18
(d, J = 6.93Hz, 2H), 6.98 (t, J = 6.69Hz, 2H), 7.12–7.55
(m, 14H); 13C NMR d (CDCl3, 75MHz): 17.8, 18.2, 23.7,
32.2, 32.5, 44.5, 52.6, 63.4, 127.1, 127.5, 127.8, 128.0,
128.2, 128.4, 128.6, 129.3, 130.1, 130.2, 135.2, 137.1, 137.0,
137.6, 138.9, 139.1, 169.1, 172.0; MS (APCI): m/z (%) 525
(M+ +2, 42), 523 (100). HPLC conditions [Chiralcel OD-
H, RT 8.9min for the S-enantiomer, 11.40 for the R-
enantiomer isopropanol–hexane (15/85), flow rate: 0.5mL/
min, k = 254nm].
7. Kumar, S.; Ramachandran, U. Tetrahedron: Asymmetry
2003, 14, 2539–2545.
9. [(R)-3,30-di(300,400,500-trifluorophenyl)-1,10-binaphthyl-2,20-
dimethylammonium]spiro[(R)-1,10-binaphthyl-2,20-dimeth-
yl amine]bromide.
10. (a) Ooi, T.; Kameda, M.; Maruoka, K. J. Am. Chem. Soc.
1999, 121, 6519–6520; (b) Ooi, T.; Kameda, M.; Maruoka,
K. J. Am. Chem. Soc. 2003, 125, 5139–5151.
11. Keck, G. E.; Boden, E.; Sonnewald, U. Tetrahedron Lett.
1981, 22, 2615–2618.
12. Ramachandran, U.; Kumar, S. Patent application no. IP/
1667/Del 2004.
8. Formation of 5b: under argon, to a cooled (ꢀ40ꢁC) and
stirred solution of imine glycinamide 2b (0.500g,
1.15mmol), catalyst 4 (0.011mmol), CsOHÆH2O (0.388g,
2.31mmol) and potassium carbonate (1.59g, 11.5mmol) in
toluene (10mL) was added 1-chloro-4-iodobutane
(0.424mL, 3.46mmol) dropwise via a syringe pump. The
contents were stirred for 22h. The reaction mixture was
diluted with water (5mL) and the contents were allowed to
warm to rt. The aqueous phase was extracted with