A new highly enantioselective synthesis of both (R)- and (S)-2- mercaptosuccinic acids

Article abstract of DOI:10.1016/S0957-4166(98)00149-9

(R)- and (S)-2-Mercaptosuccinic acids 4 were prepared in five steps in >96% ee and 49-52% overall yield via a new efficient synthesis starting from commercially available L- and D-aspartic acids.

Full text of DOI:10.1016/S0957-4166(98)00149-9

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 1641–1644
A new highly enantioselective synthesis of both (R)- and
(S)-2-mercaptosuccinic acids
Bang-Chi Chen,∗ Mark S. Bednarz, Octavian R. Kocy and Joseph E. Sundeen
Discovery Chemistry, Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543, USA
Received 3 March 1998; accepted 31 March 1998
Abstract
(R)- and (S)-2-Mercaptosuccinic acids 4 were prepared in five steps in >96% ee and 49–52% overall yield via a
new efficient synthesis starting from commercially available L- and D-aspartic acids. © 1998 Elsevier Science Ltd.
All rights reserved.
2-Mercaptosuccinic acid is an important chiral multifunctional intermediate in organic synthesis.
It has been widely employed in the synthesis of various biologically active sulfur containing com-
pounds such as the antileukemic spiro[indoline-3,20 -thiazolidine]-2,40 -diones,1 and the antimicrobial2
and antitubercular3 4-thiazolidinones. More recently, 2-mercaptosuccinic acid has been used as a building
block in the synthesis of novel polyanionic inhibitors of human immunodeficiency virus and other
viruses,4 and as a starting material in the synthesis of isocysteine, an important non-proteinogenic
amino acid in a potent peptide inhibitor of stromelysin.5 In addition, the sodium salt of the anionic
Au(I) complex of 2-mercaptosuccinic acid is an effective antiarthritic drug.6 In all of these studies,
racemic 2-mercaptosuccinic acid was used, undoubtedly due to the lack of availability of its optically
pure enantiomers.
Several methods have been reported for the preparation of 2-mercaptosuccinic acid in enantiomerically
enriched forms. Free radical addition of thioacetic acid to enantiomerically pure menthyl maleate or
fumarate followed by saponification of the resulting adducts afforded 2-mercaptosuccinic acid. However,
the enantiomeric excess was poor (4–21% ee).7 The asymmetric induction was improved to some
degree (up to 30–37% ee) by using amine-based chiral auxiliaries instead of menthol.8 In a more
encouraging approach, freshly prepared cesium thioacetate was reacted with diethyl (S)-2-mesylmalate
in DMF for 5 days at −20°C to give diethyl (R)-2-acetylthiosuccinate in 100% ee. However, significant
racemization occurred in the subsequent hydrolysis, affording (R)-2-mercaptosuccinic acid in only 59%
ee.9 In addition, 2-mercaptosuccinic acid was prepared in an early report from xanthosuccinic acid with
undetermined enantiomeric excess.10 In a program of our combinatorial drug discovery, we required an
∗
Corresponding author. E-mail: chenb@bms.com
0957-4166/98/$19.00 © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00149-9

1642
B.-C. Chen et al. / Tetrahedron: Asymmetry 9 (1998) 1641–1644
easy access to large amounts of both enantiomerically pure (R)- and (S)-mercaptosuccinic acids. Herein,
we describe an efficient synthesis of both of these compounds in high enantiomeric excess.
Considering the ease with which the α-chiral center is racemized during the conventional hydrolysis
of diethyl α-acetylthiosuccinate to 2-mercaptosuccinic acid9 and the sensitivity to oxidation of free
mercaptans in general, we decided to choose α-protected thiosuccinic free acid as the penultimate
intermediate en route to 2-mercaptosuccinic acid. Our initial focus was on the preparation of α-
tritylthiosuccinic acid from the readily available L-aspartic acid 1 (Scheme 1). Thus, treatment of 1
with sodium nitrite in water, using a modified procedure, afforded (S)-bromosuccinic acid 2 in 87%
yield.11 Reaction of (S)-2 with triphenylmethyl mercaptan in DMF at room temperature in the presence
of triethylamine gave (S)-tritylthiosuccinic acid 3 in 25% yield and 50% ee.12,13 Noteworthy is that the
configuration of the α-chiral center is retained due apparently to the neighboring group participation of
the adjacent carboxylic group. Subsequent deprotection of (S)-3 occurred smoothly with Et3SiH/TFA14
giving (S)-4 in 75% yield without racemization. The enantiomeric excess of (S)-3 could be improved to
70% ee by generating the mercaptan anion with NaH and carrying out the reaction at lower temperature
(−60°C). However, attempts to further upgrade the optical purity of 3 by either modifying the reaction
conditions or by partial resolution with chiral non-racemic amines such as benzylmethylamine, quinine,
etc., were unsuccessful.
Scheme 1.
We then turned our attention to ethylxanthosuccinic acid 5 as the penultimate intermediate to mer-
captosuccinic acid (Scheme 2). Thus, reaction of (S)-2 with the commercially available potassium
ethylxanthate in water gave (R)-5 in almost quantitative yield and 70% ee.13,15 Interestingly, the con-
figuration of the major enantiomer from this reaction is R instead of S obtained in the previous reaction
(Scheme 1), indicating that the neighboring group participation was not operative in the xanthate reaction.
Attempts to upgrade the enantiomeric excess of 5 by recrystallizing this crystalline compound from
warm water were not successful. However, it was found that highly enantiomerically pure (R)-5 could
be obtained by carrying out the substitution reaction with 0.5 equiv. sodium carbonate and 2.5 equiv.
of potassium xanthate to afford (R)-5 in 84.6% ee and subsequently resolving this material with (S)-
benzylmethylamine (Table 1, Entry 5). It was also found that it is important to obtain sufficiently
pure (>84% ee) compound 5 prior to an efficient final resolution. Finally, deprotection of (R)-5 with
concentrated ammonium hydroxide in methanol afforded (R)-4 in >96% ee and 90% yield. The optical
purity of 4 was determined by derivatizing (R)-4 with Ph3COH/BF316 to give (R)-312,13 and analyzing the
latter by chiral HPLC. Following the strategy used for the preparation of (R)-4, (S)-2-mercaptosuccinic
acid 4 was prepared in five steps from D-aspartic acid 1 in 50% overall yield and >96% ee.

B.-C. Chen et al. / Tetrahedron: Asymmetry 9 (1998) 1641–1644
1643
Scheme 2.
Table 1
Preparation of 2-ethylxanthosuccinic acids 5 in water at 0–5°C
In summary, a highly enantioselective method for the preparation of both (R)- and (S)-2-
mercaptosuccinic acids, 4, has been developed. This method is expected to be adaptable to the
synthesis of other racemization prone 2-mercaptocarboxylic acids. In addition, the ready availability of
both enantiomerically pure (R)- and (S)-4 will make it possible to synthesize the above-mentioned, 2-
mercaptosuccinic acid derived, biologically active compounds,1–6 as well as others, in chiral nonracemic
forms. Furthermore, considering the ease with which both (R)-/(S)-trithylmercaptosuccinic acids 3 and
(R)-/(S)-ethylxanthosuccinic acids 5 were prepared and deprotected under mild acidic or basic conditions
with high enantiomeric excesses, it is also expected that these intermediates, 3 and 5, will become
valuable chiral nonracemic, multifunctional sulfur-containing building blocks in asymmetric synthesis.

1644
B.-C. Chen et al. / Tetrahedron: Asymmetry 9 (1998) 1641–1644
References
1. Rajopadhye, M.; Popp, F. D. J. Heterocyclic Chem. 1987, 24, 1637.
2. Vashi, B. S.; Mehta, D. S.; Shah, V. H. Indian J. Chem. 1995, 34B, 802. Desai, K.; Baxi, A. J. J. Indian Chem. Soc. 1992,
69, 212.
3. Dave, M. P.; Patel, J. M.; Langalia, N. A.; Thaker, K. A. J. Indian Chem. Soc. 1984, 61, 891.
4. Leydet, A.; Jeantet-Segonds, C.; Bouchitte, C.; Moullet, C.; Boyer, B.; Roque, J. P.; Witvrouw, M.; Este, J.; Snoeck, R.;
Andrei, G.; Clercq, E. D. J. Med. Chem. 1997, 40, 350.
5. Pires, R.; Burger, K. Tetrahedron Lett. 1996, 37, 8159.
6. Nomiya, K.; Yokoyama, H.; Nagano, H.; Oda, M.; Sakuma, S. Bull. Chem. Soc. Jpn 1995, 68, 2875.
7. Nozaki, K.; Yoshihara, M.; Matsubara, Y.; Maeshima, T. Phosphorus and Sulfur 1985, 22, 1. Okuda, Y.; Yoshihara, M.;
Maeshima, T.; Amaya, N.; Murata, Y. Chem. Express 1989, 4, 13; Chem. Abstr. 1989, 111, 232021.
8. Ninoi, T.; Yoshihara, M.; Maeshima, T.; Fujii, M.; Aida, T. Yukagaku 1988, 37, 1044; Chem. Abstr. 1989, 111, 114715.
Nozaki, K.; Yoshihara, M.; Matsubara, Y.; Maeshima, T. Yukagaku 1984, 33, 797; Chem. Abstr. 1985, 102, 113911.
9. Strijtveen, B.; Kellogg, R. M. J. Org. Chem. 1986, 51, 3664.
10. Levene, P. A.; Mikeska, L. A. J. Chem. Soc. 1924, 60, 685.
11. Volkmann, R. A.; Kelbaugh, P. R.; Nason, D. N.; Jasys, V. J. J. Org. Chem. 1992, 57, 4352.
12. Compound 3 has the following physical and spectroscopic properties: mp 190°C dec.; [α]D +63.7 (c 1.13, MeOH) for
(R)-3 with 96.1% ee;13 1H NMR (CD3OD) δ 7.20–7.58 (m, 15H), 4.89 (bs, 2H), 3.22 (dd, J=3.7, 11.1 Hz, 1H), 2.41 (dd,
J=11.1, 17.3 Hz, 1H), 1.42 (dd, J=17.3, 3.7 Hz, 1H).
13. The enantiomeric excesses of 3 and 5 were determined by chiral HPLC using a Chiralcel OJ-R column from Daicel
Chemical Industries, Ltd.
14. For deprotection of tritylmercaptans, see: Pearson, D. A.; Blanchette, M.; Baker, M. L.; Guindon, C. A. Tetrahedron Lett.
1989, 30, 2739.
15. Compound 5 has the following physical and spectroscopic properties: mp 129–130°C; [α]D +72.9 (c 1.43, MeOH) for
(R)-5 with 96.2% ee;13 1H NMR (CD3OD) δ 4.96 (bs, 2H), 4.68 (m, 2H), 2.75–3.1 (m, 3H), 1.42 (t, J=7.2 Hz, 3H).
16. Hiskey, R. G.; Tucker, W. P. J. Am. Chem. Soc. 1962, 84, 4794.