An efficient synthesis of (R)-(-)-baclofen

Article abstract of DOI:10.1016/S0957-4166(99)00235-9

In this report we describe an efficient synthesis of (R)-(-)-baclofen, a selective GABA(B) agonist used as an antispastic agent. Our strategy is based on an enantioselective deprotonation of a cyclobutanone easily obtained by [2+2] cycloaddition of 4-chlorostyrene and dichloroketene.

Full text of DOI:10.1016/S0957-4166(99)00235-9

Pergamon
Tetrahedron: Asymmetry 10 (1999) 2113–2118
An efficient synthesis of (R)-(−)-baclofen
Patrícia Resende,a Wanda P. Almeida b and Fernando Coelho a,∗
aInstituto de Química, UNICAMP, PO Box 6154, 13083-970 Campinas SP, Brazil
bInstituto de Ciências da Saúde, Universidade Paulista, Campinas SP, Brazil
Received 6 April 1999; accepted 27 May 1999
Abstract
In this report we describe an efficient synthesis of (R)-(−)-baclofen, a selective GABAB agonist used as an
antispastic agent. Our strategy is based on an enantioselective deprotonation of a cyclobutanone easily obtained by
[2+2] cycloaddition of 4-chlorostyrene and dichloroketene. © 1999 Published by Elsevier Science Ltd. All rights
reserved.
1. Introduction
GABA or γ-aminobutyric acid is the major inhibitory neurotransmitter in the mammalian central ner-
vous system (CNS).1 This simple amino acid has two major receptor subtypes (GABAA and GABAB).2,3
These receptors apparently play an important role both in the central and peripheral nervous system
through ion-channel regulation.4
The GABAB receptor in peripheral and central nervous systems is implicated in many biological
processes including analgesia muscle relaxation, hypertension, increased gastric motility, and inhibition
of the liberation of corticotropin releasing hormone.5a,b In contrast with the several reports concerning
the development of specific agonists or antagonists acting at the GABAA receptor site,2–6 there are only
few examples of those acting at the GABAB receptor site.
Baclofen, or (3R)-4-amino-3-(4-chlorophenyl)butanoic acid (1, Fig. 1), is the only selective and
therapeutically available GABAB agonist known (Lioresal® and Baclon®).7a–c It is used in the treatment
of spasticity, a serious disease characterized by an increased muscle tone usually perceived as muscle
tightness or achiness in the limbs. These symptons are normally associated with multiple sclerosis (MS).
Baclofen is commercialized in its racemic form, however literature observations suggested that the
biological activity of 1 resides in the R enantiomer.8 According to legislation already approved in many
countries of the world concerning the commercialization of pharmaceutical products, drugs such as
∗
Corresponding author. E-mail: coelho@iqm.unicamp.br
0957-4166/99/$ - see front matter © 1999 Published by Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00235-9

2114
P. Resende et al. / Tetrahedron: Asymmetry 10 (1999) 2113–2118
Figure 1.
1 will soon be sold only in their enantiomerically pure form. This requirement justifies the need for
enantioselective strategies leading to the preparation of these compounds, if possible in a simple and
efficient way.
In our studies on the structure–activity relationship of analogs of 1, it has been necessary to prepare
a series of derivatives with different substitution patterns. However none of the already described
syntheses9,10 permitted us to easily achieve our target. In view of these limitations, we have recently
reported a four step synthesis of (ꢀ)-baclofen,11a based on a [2+2]11b cycloaddition reaction between
dichloroketene and a suitable substituted olefin. Now we report herein an asymmetric version of our
original methodology, in which the asymmetry was introduced through an enantioselective deprotonation
step.
2. Results and discussion
Cycloadduct 2 was obtained as the only detectable regioisomer by the reaction between in situ
generated dichloroketene11b and commercial 4-chlorostyrene in 91% yield (Scheme 1). Reductive
dechlorination of 2 in acetic acid/Zn dust furnished the cyclobutanone derivative 3 in 92% yield.
Scheme 1. Reagents and conditions: (a) Zn–Cu, POCl3, CCl3COCl, Et2O, 12 h, rt, 91%; (b) Zn/AcOH, 14 h, rt, 93%
Enantioselective deprotonation of meso or prochiral compounds using lithium amide bases provides
useful chiral intermediates for the synthesis of a wide variety of organic compounds including natural
products.12 The high enantioselectivity obtained by this methodology associated with its operational
simplicity encouraged us to use it as a way to introduce asymmetry into our synthetic approach.
The enantioselective deprotonation of cyclobutanone 2 was carried out by using the chiral base, lithium
(S,S0)-α,α0-dimethylbenzylamide13,14 at −100°C in tetrahydrofuran (THF) and the resulting enolate was
trapped by triethylsilyl chloride to provide the silylenol ether 4 (Scheme 2) in 70% yield. Although the
enantiomeric excess of the silylenol ether 4 could not be determined at this stage, it was converted further
into the γ-butyrolactone 5 by ozonolysis (Scheme 2), followed by sodium borohydride reduction of the
ozonide.15 The optical purity of lactone 5 was determined to be ≥97%, by comparison of its specific
optical rotation [α]20 −43 (c=0.5, CHCl3) with that reported (lit.10c [α]20 −44 (c=0.5, CHCl3)). Since
D
D
none of the transformations could have epimerized the stereogenic carbon of 4, we can assume that its
optical purity should be comparable to that found for lactone 5, i.e. an enantiomeric excess of 98%.

P. Resende et al. / Tetrahedron: Asymmetry 10 (1999) 2113–2118
2115
Scheme 2. Reagents and conditions: (a) lithium (S,S0)-α,α0-dimethylbenzylamide, THF, TESCl, −100°C, 15 min, 70%; (b) (i)
O3, CH2Cl2, −78°C, 40 min; (ii) Me2S, −78°C -→ rt, 12 h; (iii) NaBH3CN, AcONH4, 12 h; (iv) 6N HCl, 100°C, 70% overall
yield (one-pot sequence); (c) O3, MeOH, −78°C, 45 min, then NaBH4, then dilute HCl (70%)
At this stage we had envisaged the use of the reductive amination conditions described by Brenna et
al.10i to directly transform the ozonide into (−)-1. Unfortunately, we were unable to isolate any reductive
amination product, even by changing the ratio of hydride:substrate, the solvent and the reaction time.
To overcome this problem, we decided to reduce the ozonide in the presence of dimethyl sulfide and
then the crude material was treated with sodium cyanoborohydride and ammonium acetate. This simple
modification had permitted us to prepare (−)-baclofen (1) in 70% yield with an enantiomeric excess of
97.5% ([α]20 −1.95 (c, 0.6, H2O), lit.5j–i [α]20 −2.0 (c, 0.6, H2O)).
D
D
3. Conclusion
In summary, the synthesis of (R)-(−)-baclofen was developed by a four-step sequence in 40% overall
yield from 4-chlorostyrene. As several substituted styrenes are commercially available or could be easily
prepared,16 this approach should also provide convenient access to enantiomerically pure analogs.
4. Experimental
4.1. General procedures
1H NMR and 13C NMR spectra were recorded on a Varian GEMINI BB-300 at 300 MHz and 75.1
MHz, respectively. The mass spectra were recorded on an HP model 5988A GC–MS. The melting
points were measured in open capillary tubes using an Electrothermal apparatus model 9100, and are
uncorrected. The [α] values are corrected and were measured in Polamat A. Column chromatography
D
was performed on silica gel (70–230 or 200–400 mesh). Ether and THF were distilled from benzophenone
ketyl under nitrogen. Trichloroacetyl chloride and phosphorus oxychloride (both Aldrich) were distilled

2116
P. Resende et al. / Tetrahedron: Asymmetry 10 (1999) 2113–2118
before use. TLC visualization was achieved by spraying with 5% ethanolic phosphomolybdic acid and
heating. 4-Chlorostyrene was purchased from Aldrich and used without purification.
4.2. Racemic 2,2-dichloro-3-(40-chlorophenyl)cyclobutanone 2
To a stirred suspension of Zn–Cu17 (3.10 g, 45.48 mmol) and 4-chlorostyrene (3.00 g, 21.66 mmol)
in dry ether (44 mL) was added dropwise, during 1 h at room temperature, a solution of phosphorus
oxychloride (4.98 g, 3.03 mL, 32.49 mmol) and trichloroacetyl chloride (5.90 g, 3.63 mL, 32.49 mmol)
in dry ether (22 mL). The suspension was stirred for 20 h at room temperature. The mixture was
filtered through a pad of Celite and washings were performed with hexane (3×100 mL). The filtrate
was concentrated in vacuo to one third of the original volume. The residue was diluted with hexane (100
mL) and again concentrated to one third. This operation was repeated twice more. The final concentrate
(∼150 mL) was washed with cold water (200 mL), saturated NaHCO3 (100 mL) and brine (100 mL),
dried over anhydrous sodium sulfate and filtered. Concentration under reduced pressure followed by flash
chromatography on silica gel with hexane:ethyl acetate (8:2) gave a solid residue. Recrystallization from
dichloromethane gave dichlorocyclobutanone 2 as a white solid (4.90 g, 91%). M.p. 73–75°C; IR (KBr,
λ
max): 2928, 1812, 1493, 1403, 1094, 1014, 827 cm−1; MS (70 eV, m/e): 250 (M++2), 248 (M+), 208,
1
206, 138; H NMR (300 MHz, CDCl3) δ 3.51–3.73 (m, 2H), 4.21 (t, J=10.2 Hz, 1H), 7.17–7.43 (dd,
J=8.4 and 8.8 Hz, 4H); 13C NMR (75.1 MHz, CDCl3) δ 29.8, 46.0, 50.2, 129.4, 129.9, 192.2. Anal.
calcd for C10H7Cl3O: C, 48.39; H, 2.84. Found: C, 48.37; H, 2.83.
4.3. 3-(40-Chlorophenyl)cyclobutanone 3
To a stirred solution of 2 (4.0 g, 16.09 mmol) in acetic acid (100 mL) was added Zn dust (6.43 g, 99.0
mmol). The suspension was stirred at room temperature for 14 h and then the mixture was diluted with
ether (300 mL) and filtered. The solid was washed with ether (100 mL) and the combined organic extracts
washed with a saturated solution of NaHCO3 (3×150 mL), brine (200 mL) and dried over anhydrous
sodium sulfate. Concentration under reduced pressure followed by flash chromatography on silica gel
with hexane:ethyl acetate (9:1) gave compound 3 (2.70 g, 93%) as a yellow-tinged oil. IR (neat, λmax):
2924, 1787, 1493, 1378, 1013, 821 cm−1; 1H NMR (300 MHz, CDCl3) δ 3.16–3.26 (m, 2H), 3.44–3.55
(m, 2H), 3.60–3.69 (m, 1H), 7.21–7.34 (m, 4H); 13C NMR (75.1 MHz, CDCl3) δ 27.9, 54.6, 127.9,
128.7, 142.0, 206.0; MS (70 eV, m/e): 180 (M+), 110, 138, 103, 89, 77. Anal. calcd for C10H9ClO: C,
66.65; H, 5.04. Found: C, 66.63; H, 5.03.
4.4. 1-[(3S)-3-(4-Chlorophenyl)cyclobut-1-enyloxyl]-1,1-diethyl-1-silapropane 4
To a stirred solution of (S,S0)-α,α0-dimethylbenzylamine (1.28 g, 5.2 mmol) in dry THF (35 mL)
was added n-butyllithium (3.72 mL, 1.4 M solution in hexanes, 5.2 mmol) at −78°C under nitrogen
and the resulting solution was allowed to warm to room temperature over 10 min. The solution was
cooled to −100°C and triethylsilyl chloride (0.8 mL, 0.762 g, 4.8 mmol) was added, followed by addition
of a solution of cyclobutanone 3 (0.722 g, 4 mmol) in THF (10 mL) at the same temperature. After
stirring for 10 min, triethylamine (5 mL) and saturated sodium carbonate were added and the mixture was
concentrated in vacuo. The residue was extracted with hexane, washed with saturated sodium hydrogen
carbonate and brine, and then dried over Na2SO4. Evaporation of the solvent gave a residue which was
purified by column chromatography on silica gel. Elution with hexane:ethyl acetate (99:1, v/v) afforded
the silyl enol ether 4 (1.178 g, 70%) as a colorless fluid oil. [α]20 −4.25 (c 2.0, CHCl3); IR (neat, λmax):
D

P. Resende et al. / Tetrahedron: Asymmetry 10 (1999) 2113–2118
2117
1
3070, 2966, 2914, 1641, 1626, 1574, 1300, 1284, 1215, 999, 844 cm−1; H NMR (300 MHz, CDCl3)
δ 0.7 (t, 6H), 0.9 (q, 9H), 2.3 (d, J=16 Hz, 1H), 3.02 (dd, J=4 Hz; J=16 Hz, 1H), 3.5 (d, J=4 Hz, 1H),
4.83 (d, J=0.7 Hz, 1H), 7.0 (s, 4H, aromatic); 13C NMR (75.1 MHz, CDCl3) δ 6.0, 7.0, 48.4, 54.2, 128.3,
128.6, 128.7, 129.9, 141.0, 151.0. Anal. calcd for C16H23ClOSi: C, 65.167; H, 7.861. Found: C, 65.150;
H, 7.852.
4.5. (R)-(−)-Baclofen, hydrochloride 1
A solution of (−)-4 (0.8 g, 2.71 mmol) in methylene chloride:methanol (1:1, 15 mL) was treated with
ozone at −78°C for 40 min. After addition of dimethyl sulfide (5 mL) at −78°C, the temperature was
allowed to rise over 12 h with stirring. Evaporation of the solvents was accompanied by the addition
of methanol (10 mL), ammonium acetate (0.256 g, 4.065 mmol) and sodium cyanoborohydride (0.313
g, 4.065 mmol). The reaction mixture was stirred for 12 h at room temperature. Then a solution of 2
M sodium hydroxide was added and the reaction was stirred for an additional 2 h, concentrated under
reduced pressure, diluted with water and extracted with ethyl acetate. The organic phase was dried over
sodium sulfate, and the solvent was removed under reduced pressure. The residue was treated with
aqueous 6N HCl (5 mL) and heated at 100°C for 12 h. The solvent was removed in vacuo and the residue
was triturated in isopropanol yielding crystalline baclofen (0.37 g, 70%). M.p. 195°C (lit.9a 195°C);
[α]20 −1.95 (c 0.6, H2O), lit.5 [α]20 −2.0 (c 0.6, H2O); IR (KBr, λmax): 3000–2500, 1620, 1550, 1490,
D
D
1090 cm−1; 1H NMR (300 MHz, DMSO-d6): δ 2.65–2.91 (AB part from ABX, JAB=16.6 Hz, JAX=6.9
Hz, JBX=7.7 Hz, 2H), 3.10–3.39 (AB part from ABX, JAB=12.8 Hz, JAX=6.0 Hz, JBX=8.9 Hz, 2H),
3.64–3.72 (m, 1H), 7.41–7.43 (m, 4H); 13C NMR (75.1 MHz, DMSO-d6): δ 176.0, 141.8, 131.2, 128.9,
128.8, 48.5, 37.7.
Acknowledgements
We thank the Brazilian National Research Council (CNPq) for F.C.’s fellowship and FAPESP for
financial support and P.R.’s fellowship (# 97/08571-2).
References
1. Bloom, F. E. In Neurotransmission and The Central Nervous System; Hardman, J. E.; Limbird, L. E.; Molinoff, P. E.;
Ruddon, R. W.; Gilman, A. G., Eds.; Goodman & Gilman’s The Pharmacological Basis of Therapeutics, 9th ed.; McGraw-
Hill: New York, 1996; pp. 267–282.
2. (a) Bowery, N. G.; Hudson, A. L.; Price, G. W. Neuroscience 1987, 20, 365–383; (b) Nicoll, R. A. Science 1988, 241,
545–551; (c) Hill, D. R.; Bowery, N. G. Nature 1981, 290, 149–152.
3. Mann, A.; Boulanger, T.; Brandau, B.; Durant, F.; Evrard, G.; Heaulme, M.; Desaules, E.; Wermuth, C.-G. J. Med. Chem.
1991, 34, 1307–1313.
4. (a) Bowery, N. G. Trends Pharmacol. Sci. 1982, 3, 400–403; (b) Malcangio, M.; Bowery, N. G. Clin. Neuropharmacology
1995, 18, 285–305.
5. Bowery, N. G.; Pratt, G. D.; Knott, C. In GABAB Receptors in Mammalian Function; Bowery, N. G.; Bittiger, H.; Olpe,
H.-R., Eds.; John Wiley & Sons: Chichester, 1990; p. 3.
6. Johnston, G. A. R. Benzodiazepine/GABA receptors and chloride channels. In Receptor Biochemistry and Methodology;
Olsen, R. W.; Veuter, J. C., Eds.; A. R. Liss: New York, 1986; Vol. 5, pp. 57–71.
7. (a) Kerr, D. I. B.; Ong, J. Med. Res. Revs. 1992, 12, 593–636; (b) Berthelot, P.; Vaccher, C.; Flouquet, N.; Debaert, M.;
Luyckx, M.; Brunet, C. J. Med. Chem. 1991, 34, 2557–2560; (c) Kerr, D. I. B.; Ong, J.; Doolette, D. J.; Abbenante, J.;
Prager, R. H. Eur. J. Pharmacol. 1993, 96, 239–245.

2118
P. Resende et al. / Tetrahedron: Asymmetry 10 (1999) 2113–2118
8. Olpe, H.-R.; Demieville, H.; Baltzer, V.; Bencze, W. L.; Koella, W. P.; Wolf, P.; Haas, H. L. Eur. J. Pharmacol. 1978, 52,
133–136.
9. For some examples of the racemic synthesis of baclofen and derivatives see: (a) Kerberle, H.; Faigle, J. W.; Wilhelm, M.
Swiss Patent 499,046, Chem. Abstr. 1968, 69, 106273f; (b) Wall, G. M.; Baker, J. K. J. Med. Chem. 1989, 32, 1340–1348;
(c) Pifferi, G.; Nizzola, R.; Cristoni, A. Il Farmaco 1994, 49, 453–455; Chem. Abstr. 1995, 122, 204918; (d) Ibuka, T.;
Schoenfelder, A.; Bildstein, P.; Mann, A. Synth. Commun. 1995, 25, 1777–1782.
10. For some examples of the asymmetric synthesis of (−)-baclofen and derivatives see: (a) Chênevert, R.; Desjardins, M.
Tetrahedron. Lett. 1991, 32, 4249–4250; (b) Herdeis, C.; Hubmann, H. P. Tetrahedron: Asymmetry 1992, 3, 1213–1221;
(c) Schoenfelder, A.; Mann, A.; Le Coz, S. Synlett 1993, 63–64; (d) Chênevert, R.; Desjardins, M. Can. J. Chem. 1994, 72,
2312–2317; (e) Prager, R. H.; Schafer, K.; Hamon, D. P. C.; Massy-Westropp, R. A. Tetrahedron 1995, 51, 11465–11472;
(f) Yoshifuji, S.; Kaname, M. Chem. Pharm. Bull. 1995, 43, 1302–1306; (g) Langlois, N.; Dahuron, N.; Wang, H.-S.
Tetrahedron 1996, 52, 15117–15126; (h) Mazzini, C.; Lebreton, J.; Alphand, V.; Furstoss, R. Tetrahedron Lett. 1997,
38, 1195–1196; (i) Brenna, E.; Carraccia, N.; Fuganti, C.; Fuganti, D.; Graselli, P. Tetrahedron: Asymmetry 1997, 8,
3801–3805.
11. (a) Coelho, F.; de Azevedo, M. B. M.; Boschiero, R.; Resende, P. Synth. Commun. 1997, 27, 2455–2465; (b) Tidwell, T.
T. Ketenes; Wiley-Intersciences: New York, 1995.
12. Honda, T.; Kimura, N.; Tsubuki, M. Tetrahedron: Asymmetry 1993, 4, 1475–1478 and references cited therein.
13. The chiral base was prepared according to a procedure described by Eleveld, M. B.; Hogeveen, H.; Schudde, E. P. J. Org.
Chem. 1986, 51, 3635–3642; see also, Marshall, J. A.; Lebreton, J. J. Am. Chem. Soc. 1988, 110, 2925–2931.
14. The [α]D value reported for this amine by Hogeveen et al. (see above) was −157 (c 2.4, EtOH). We found a value of [α]D20
−156.
15. Honda, T.; Kimura, N.; Sato, S.; Kato, D.; Tominaga, H. J. Chem. Soc., Perkin Trans. 1 1994, 1043–1046.
16. Borka, L. Acta Pharm. Suec. 1979, 16, 345–348; Chem. Abstr. 1980, 92, 82536r.
17. (a) Smith, C. W.; Norton, D. G. Organic Synthesis, Coll. IV 1963, 348–350; (b) McCaney, C. C.; Ward, R. S. J. Chem.
Soc., Perkin Trans. 1 1975, 1600–1603; (c) for the preparation of the Zn–Cu couple see: Brady, W. T.; Liddell, H. G.;
Vaughan, W. L. J. Org. Chem. 1966, 31, 626–628.