Download PDF
Synthesis 2024; 56(04): 549-552
DOI: 10.1055/s-0042-1751470
special topic
Synthetic Development of Key Intermediates and Active Pharmaceutical Ingredients (APIs)

An Alternative Formal Synthesis of (S)-(+)-Vigabatrin

Sudhir P. Chaskar
a   API R & D Centre, Emcure Pharmaceuticals Ltd., ITBT Park, Phase-II, MIDC, Hinjewadi, Pune-411057, Maharashtra, India
,
Ramchandra Honparkhe
a   API R & D Centre, Emcure Pharmaceuticals Ltd., ITBT Park, Phase-II, MIDC, Hinjewadi, Pune-411057, Maharashtra, India
,
Arvind K. Aghao
b   Belbhim College, Department of Chemistry, Beed, Maharashtra, India
,
Rakesh G. Thorat  ∗
a   API R & D Centre, Emcure Pharmaceuticals Ltd., ITBT Park, Phase-II, MIDC, Hinjewadi, Pune-411057, Maharashtra, India
,
Chinmoy Pramanik
a   API R & D Centre, Emcure Pharmaceuticals Ltd., ITBT Park, Phase-II, MIDC, Hinjewadi, Pune-411057, Maharashtra, India
› Author Affiliations
› Further Information
 
 PDF Download Permissions and Reprints All articles of this category


Abstract

An improved synthesis of chirally pure advanced pyrrolidone intermediate of the (S)-(+)-vigabatrin, an antiseizure active pharmaceutical ingredient (API) is reported. The synthesis is developed using commercially available, cheaper amino acid (R)-methionine. Meldrum’s acid served as a two-carbon homologation unit to access the pyrrolidone intermediate in a short synthetic sequence. The sequence to pyrrolidone intermediate is scalable and eludes the use of organometallic pyrophoric reagents on a large scale.


#

Key words

GABA - vigabatrin - decarboxylative cyclization - Meldrum’s acid - seizure - API

Convulsive seizures,[1] major characteristic for epilepsy, are prevented by γ-aminobutyric acid (GABA, 1) (Figure [1]), a very well-known and major inhibitory transmitter. When GABA level decreases below a threshold level in brain, it triggers the initiation of convulsive seizures. It is already demonstrated that by increasing the GABA concentration in brain, convulsive seizures could be reduced or prevented.

Zoom Image
Figure 1 GABA and γ-vinyl-GABA

However, the peripheral admission of GABA is ineffective because GABA (1) cannot cross ‘the blood brain barrier (BBB)’. Vigabatrin (2), 4-amino-5-hexenoic acid (Figure [1]) is an analogue of GABA with a vinyl group at gamma position.[2a] Owing to its more lipophilic nature, ‘γ-vinyl-GABA’ is able to cross the blood brain barrier and it helps to increase the GABA concentration by inhibiting the activity of GABA-T, a GABA degrading enzyme.[2b] By increasing GABA concentration in brain, γ-vinyl-GABA helps to prevent convulsive seizures.[2b]

In this regard, vigabatrin was first approved in the UK as an antiseizure medication in 1989 and later in 2009 in the US.[2b] [c] Vigabatrin became available as a generic medication in 2019.[2d] Although racemic vigabatrin can be used for the clinical practice, only (S)-(+)-vigabatrin is the pharmacologically active (API) enantiomer out of the two. Due to its biological importance and high market demand for its cheaper source, we explored alternative routes for its large scale production.

Among the various literature reported synthesis[3] for chirally pure (S)-(+)-vigabatrin, Knaus et al.[3a] have used (R)-methionine as the starting material in their chiral pool approach (Scheme [1]).

Zoom Image
Scheme 1 Knaus et al. approach to (S)-vigabatrin

Initially, acid/amine functionality of (R)-(–)-methionine was protected as its methyl ester and carbamate, respectively, to afford compound 4. One-pot two-carbon homologation of 4 (one-pot reduction homologation process)[4] furnished olefin 5, which on treatment with Mg/MeOH[5] underwent a cascade of double bond reduction, carbamate cleavage, and finally the cyclization to provide the critical pyrrolidone intermediate 6. Oxidation of thioether in 6 followed by elimination and subsequent hydrolysis of pyrrolidone afforded (S)-vigabatrin.

Even though the overall yield in the synthesis was reasonable, the use of pyrophoric reagents like t-BuLi, DIBAL-H, could make this process unviable for industrial manufacturing of (S)-vigabatrin. Recently, Jachak et a.l also reported a synthesis for both enantiomers of vigabatrin, starting from d-methionine aldehyde.[3h] Using activated carbonyl of aldehyde for homologation avoided the use of strong bases for the homologation. However, the reduction of unsaturated ester was effected using catalytic hydrogenation.

While performing literature survey, some of the methods employed pressure reactions and/or organometallic reagents to access the pyrrolidone core (intermediate 6 and 7).[6] Further exploration of literature provided an interesting report where starting from an α-amino acid and Meldrum’s acid, the 5-subsituted pyrrolidone core was accessed in three steps (Scheme [2]).[7]

Zoom Image
Scheme 2 a) Reports for pyrrolidone core synthesis using Meldrum’s acid.[7] b) This work.

Based on these reports, a strategy was devised to use (R)-methionine and Meldrum’s acid[6] to arrive at intermediate 6. Accessing the pyrrolidone 6 would be a much simpler alternative to Knaus et al.’s two-carbon homologation approach. The rationale was that it would eliminate most of the pyrophoric reagents and hazardous reaction conditions. Thus, a synthetic strategy was finalized starting from (R)-methionine (3) and Meldrum’s acid as a two-carbon surrogate.

Accordingly, the amine functionality of (R)-methionine (3) was protected as a Boc-carbamate[7b] (Scheme [3]). The protected acid 8 was then condensed with Meldrum’s acid in the presence of EDAC·HCl to furnish intermediate 9 (Table [1]).

Zoom Image
Scheme 3 Meldrum’s acid route to intermediate 6

Table 1 Reagents Screened for the Conversion of 8 into 9

Reagent

Solvent

Yield (%)a

Isobutyl chloroformate

DCM, –10 to –20 °C

68

Dicyclohexylcarbodiimide

DCM, –5 to –10 °C

60

EDAC·HCl

DCM, –10 °C to rt

80

a Isolated yields.

It was proposed to eliminate the purification of compound 9 from scalability perspective. The reduction of the crude keto(-enol) intermediate 9 was proposed, which would save a purification step as well as yield loss during the purification of intermediate 9, thus improving the efficiency of the process. Accordingly, the keto(-enol) functionality of crude intermediate 9 was reduced (in situ) by using sodium borohydride in the presence of acetic acid to provide compound 10 in very good yield (75% over 2 steps).[7]

By following the protocol of thermal decarboxylation reported by Smrcina et al.[7a] compound 10 first provided the Boc-protected pyrrolidone and the subsequent in situ Boc deprotection[7b] with TFA resulted in cyclization to provide the critical pyrrolidone intermediate 6 (80% yield) (Scheme [3]). The intermediate 6 has been converted to (S)-(+)-vigabatrin by Knaus et al.[3] Thus, the synthesis of intermediate 6 completes the formal synthesis of (S)-(+)-vigabatrin.

The work presented above, marks first formal synthesis of (S)-(+)-vigabatrin using Meldrum’s acid as a two-carbon homologation unit. The route to access compound 6, in turn compound 7 can be explored further for its synthetic applications. The pyrrolidone 7 is found in core structures of some other natural products such as (–)-sessiliofoliamide C, (–)-8-epi-etemoamide, and (–)-martinellic acid (Figure [2]), to name a few.[8]

In conclusion, an improved process for (S)-(+)-vigabatrin was developed, starting from chirally pure (R)-methionine by taking advantage of use of Meldrum’s acid for the 2-carbon homologation.[9] The process involves simplified reagents, avoids the use of any hazardous and pyrophoric reagents for homologation and affords good overall yield [56% over 4 steps from (R)-methionine]. This process is being evaluated for commercial production of vigabatrin.

Zoom Image
Figure 2 Natural products with fragments derived from intermediate 7 (blue-colored fragments)

Unless specified otherwise, all materials were purchased from commercial suppliers and were used as received. 1H and 13C NMR spectra were recorded on a Varian 400 MR spectrometer with CDCl3 and DMSO-d 6 as solvents. The Applied Biosystems API-2000LCMS mass spectrometer was used to record the mass spectra.


#

(R)-2-tert-Butoxycarbonylamino-4-thiomethylbutanoic Acid (8)

To a suspension of (R)-methionine (3; 50.0 g, 335.12 mmol) in THF (125 mL) at rt was added aq 20% NaOH (100 mL) followed by Boc2O (77.5 g, 355 mmol). The reaction mixture was maintained at rt for 12 h. After completion of the reaction (monitored by TLC), H2O was added and the mixture was acidified using aq 0.1 N HCl. The mixture was extracted with CH2Cl2 (4 × 250 mL) and the combined organic layers were dried (Na2SO4) and concentrated in vacuum providing 8 as a clear thick oil; yield: 78.5 g (94%). It was used further without any purification.

1H NMR (400 MHz, CDCl3): δ = 8.61 (br s, 1 H), 6.64 (s, 1 H), 5.16 (s, 1 H), 4.45–4.34 (br m, 1 H), 2.58 (t, J = 7.48 Hz, 2 H), 2.25–2.14 (br m, 1 H), 2.11 (s, 3 H), 2.05–1.94 (m, 1 H), 1.46 (s, 9 H).[7e]

13C NMR (100 MHz, CDCl3): δ = 176.7, 176.1, 155.6, 80.4, 53.50, 52.7, 31.8, 29.9, 28.3, 15.4.


#

tert-Butyl (S)-(1-(2,2-Dimethyl-4,6-dioxo-1,3-dioxan-5-yl)-4-(methylthio)butan-2-yl)carbamate (10)

To a round-bottom flask containing a solution of 8 (60 g, 240 mmol) in DCM (1200 mL) was added Meldrum’s acid (38.4 g, 266 mmol) and DMAP (44.16 g, 361 mmol). The reaction mixture was cooled to –5 to –10 °C. Solid EDAC·HCl (115.2 g, 603 mmol) was added to the mixture at –5 to –10 °C and the reaction mixture as stirred at –5 to –10 °C for 4 h. After complete consumption of the starting material (by TLC), the mixture was warmed to rt, diluted with DCM (1200 mL) and dil HCl (0.1 N, 1200 mL). The layers were separated and the organic layer was washed with brine, dried and concentrated under reduced pressure. The residue was dissolved in a mixture of DCM (1200 mL) and AcOH (150 mL) and was cooled to –5 to –10 °C. To the cooled reaction mixture, NaBH4 (22.2 g, 600 mmol) was added portionwise. The reaction mixture was stirred at –5 to 0 °C for 6 h. The mixture was warmed to rt and stirred for 12 h. After completion of the reaction (monitored by TLC), the mixture was diluted slowly with H2O (420 mL) and was stirred for 15 min. The organic layer was separated and washed with H2O (3 × 600 mL) and brine (2 × 600 mL). The organic layer was dried and concentrated under reduced pressure to provide 10 as a light brown colored residue. Flash chromatography (cyclohexane/EtOAc 85:15) purification of the residue provided 10 as a thick light yellow oil; yield: 65.3 g (75% over two steps); [α]D +31 (c 1.0, EtOH).

1H NMR (400 MHz, CDCl3): δ = 4.56 (br d, J = 8.0 Hz, 1 H), 4.0 (br s, 2 H), 2.63–2.50 (m, 2 H), 2.34–2.23 (m, 1 H), 2.22–2.15 (m, 1 H), 2.12 (s, 3 H), 1.97–1.85 (m, 1 H), 1.81 (s, 3 H), 1.78 (s, 3 H), 1.41 (s, 9 H).

13C NMR (100 MHz, CDCl3): δ = 165.7, 165.5, 156.5, 105.0, 79.7, 49.2, 44.40, 35.5, 31.9, 30.5, 28.6, 28.2, 25.9, 15.6.

MS-ESI: m/z for C16H27NO6S (M+): 361.5; found: 361.6.

Anal. Calcd for C16H27NO6S: C, 53.17; H, 7.53; N, 3.88; S, 8.87. Found: C, 53.01; H, 7.52; N, 4.02; S, 8.72.


#

(5S)-5-(2-Methylthioethyl)-2-pyrrolidone (6)

To a round-bottom flask was added compound 10 (20 g, 55.4 mmol) and toluene (400 mL). The reaction mixture was refluxed for 8 h. The mixture was cooled to 0 °C and TFA (5 mL) was added dropwise. The mixture was warmed to rt and stirred for 12 h. Then the mixture was distilled under vacuum to provide a residue. The residue was dissolved in DCM (800 mL) and washed with sat. aq NaHCO3, followed by brine. The DCM layer was dried and concentrated under reduced pressure to afford 6 as light yellow oil (it may solidify upon cooling); yield: 7.0 g (80%); [α]D +10.3 (c 2.5, CHCl3) {Lit.[3a] [α]D +10.3 (c 2.5, CHCl3­)}.

1H NMR (400 MHz CDCl3): δ = 6.45 (br s, 1 H), 3.81–3.74 (m, 1 H), 2.61–2.50 (m, 2 H), 2.39–2.25 (m, 3 H), 2.12 (s, 3 H), 1.84–1.78 (m, 2 H), 1.77–1.70 (m, 1 H).

13C NMR (100 MHz CDCl3): δ = 178.10, 53.7, 35.6, 30.9, 29.90, 27.3 15.60.

MS-ESI: m/z for C7H13NOS (M+): 159.2, found (M + H+): 160.2.

Anal. Calcd for C7H13NOS: C, 52.8; H, 8.23; N, 8.80; S, 20.13. Found: C, 52.6; H, 7.9; N, 8.02; S, 19.9.

The spectral data of 6 were in complete agreement with the literature data.[3a]


#
#

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

The authors wish to thank Mr. Samit Mehta, The Management, and ARD group of Emcure Pharmaceuticals Ltd. for their willing support and constant encouragement.

Supporting Information

    Supporting information for this article is available online at https://doi-org.accesdistant.sorbonne-universite.fr/10.1055/s-0042-1751470.
  • Supporting Information

Corresponding Author

Rakesh G. Thorat
API R & D Centre, Emcure Pharmaceuticals Ltd.
ITBT Park, Phase-II, MIDC, Hinjewadi, Pune-411057, Maharashtra
India   

Publication History

Received: 17 March 2023

Accepted after revision: 07 June 2023

Article published online:
03 July 2023

© 2024. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


   Permissions and Reprints All articles of this category
Zoom Image
Figure 1 GABA and γ-vinyl-GABA
Zoom Image
Scheme 1 Knaus et al. approach to (S)-vigabatrin
Zoom Image
Scheme 2 a) Reports for pyrrolidone core synthesis using Meldrum’s acid.[7] b) This work.
Zoom Image
Scheme 3 Meldrum’s acid route to intermediate 6
Zoom Image
Figure 2 Natural products with fragments derived from intermediate 7 (blue-colored fragments)