Enzymatic resolution of 2-substituted tetrahydroquinolines. Convenient approaches to tricyclic quinoxalinediones as potent NMDA-glycine antagonists

Article abstract of DOI:10.1016/S0957-4166(98)00484-4

Two approaches leading to the enantiomerically pure tricyclic quinoxalinedione class of NMDA-glycine antagonists using enzymatic resolutions are described. An intermediate, racemic methyl 1,2,3,4- tetrahydroquinoline-2-carboxylate 3, was resolved to (S)-3

Full text of DOI:10.1016/S0957-4166(98)00484-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 4295–4299
Enzymatic resolution of 2-substituted tetrahydroquinolines.
Convenient approaches to tricyclic quinoxalinediones as potent
NMDA-glycine antagonists
Seiji Katayama, Nobuyuki Ae and Ryu Nagata ∗
Sumitomo Pharmaceuticals Research Center, 1-98 Kasugade naka 3 chome, Konohanaku, Osaka 554, Japan
Received 2 November 1998; accepted 26 November 1998
Abstract
Two approaches leading to the enantiomerically pure tricyclic quinoxalinedione class of NMDA-glycine anta-
gonists using enzymatic resolutions are described. An intermediate, racemic methyl 1,2,3,4-tetrahydroquinoline-
2-carboxylate 3, was resolved to (S)-3 in 97% ee and 47% yield (E=67) using α-chymotrypsin. In an im-
proved method, hydrolysis of another intermediate, racemic methyl 1,2,3,4-tetrahydroquinoline-2-acetate 4, with
Novozym® 435 provided the desired (S)-4 in high enantioselectivity and yield (93% ee, 50%, E=94). © 1998
Elsevier Science Ltd. All rights reserved.
1. Introduction
There is increasing evidence that over-excitation of the NMDA receptor plays an important role in
neuronal cell death during ischemic or hypoxic conditions such as stroke.1 Several binding sites on the
NMDA receptor, including glutamate, glycine, and channel blocker binding sites, have been identified
and these sites offer a target for the drug treatment of not only stroke but other neurodegenerative
disorders such as Alzheimer’s and Huntington’s diseases.2,3 We have recently synthesized a series of
tricyclic quinoxalinediones as potent antagonists of the glycine site of the NMDA receptor.4–6 Among
them, the six-membered ring fused tricyclic quinoxalinedione 1 (Ki=9.9 nM) and its anilide 2 (Ki=2.6
nM) showed extremely high affinity for the glycine site, as determined by radio ligand binding assay
using [3H] 5,7-dichlorokynurenic acid.
∗
Corresponding author. E-mail: rnagata@sumitomopharm.co.jp
0957-4166/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00484-4

4296
S. Katayama et al. / Tetrahedron: Asymmetry 9 (1998) 4295–4299
(1)
In particular, only the S isomer (Ki=0.96 nM) of 2 retained the affinity and the R isomer (Ki=82
nM) was 85-fold weaker than the S isomer. Consequently, we focused our attention on the synthesis
of the enantiomerically pure S isomer, (S)-2, and its derivatives. Tricyclic quinoxalinediones 1 and 2
have been synthesized from methyl 1,2,3,4-tetrahydroquinoline-2-carboxylate (3) via methyl 1,2,3,4-
tetrahydroquinoline-2-acetate (4) according to the route outlined in Scheme 1.4
Scheme 1.
We report here two approaches for the preparation of the key intermediate (S)-4, leading to enantio-
merically pure (S)-2, as well as its derivatives.
Synthesis of enantiomerically pure 3 has been reported in the literature,7 which describes the con-
version of racemic 3 into diastereomers by coupling with an enantiomerically pure benzoyl-protected
amino acid derivative followed by a fractional recrystallization and hydrolysis. In our hands, it turned
out that this approach was too tedious for multi-gram synthesis. We then attempted to resolve racemic
3 using an enantioselective enzymatic hydrolysis. As a result, we found that treatment of racemic 3
with α-chymotrypsin in 0.1 M phosphate buffer (pH 7.5) afforded (S)-3 in high enantiomeric excess
and chemical yield (97% ee, 46%), and with a high enantiomeric ratio19 (E=67). Enantiomerically pure
(R)-3 could also be obtained as shown in Scheme 2. In this way, we were able to supply a number
of enantiomerically pure derivatives of 2, so that we could identify in vivo active glycine antagonists
including SM-18400.8–10
Scheme 2.
Although this approach was useful, we required a more straightforward and practical route leading
to the key intermediate (S)-1 for extensive pharmacological and toxicological studies of SM-18400

S. Katayama et al. / Tetrahedron: Asymmetry 9 (1998) 4295–4299
4297
Table 1
and related compounds. We therefore attempted to resolve racemic 4, since 4 was readily available
by a simple two-step sequence from quinoline-N-oxide.11–13 More than 30 lipases and proteases were
screened to determine whether enantioselective hydrolysis of racemic 4 could be achieved. In all cases,
the hydrolyses were carried out at a substrate concentration of 0.05 M with a given enzyme (25–250
wt%) in 0.1 M phosphate buffer (pH 7.5) at 30°C and the enantiomeric excess of the residual substrate
was analyzed by HPLC with a chiral stationary phase.15 Selected examples are listed in Table 1. Among
the enzymes screened, SP-525 (entry 6, available from Novo Nordisk Ltd) afforded the desired (S)-
4 with high enantioselectivity (95% ee at 67% conversion, E=9.3) and showed remarkable reactivity.
An immobilized form of SP-525 is available as Novozym® 435, which provided a somewhat lower
enantioselectivity compared with SP-525 (58% ee at 51% conversion, E=6.2), but was still acceptable.
Considering the ease of handling and the commercial availability, we decided to optimize the reaction
conditions further, using Novozym® 435.
We changed the reaction medium from phosphate buffer to an organic solvent containing a constant
amount of water, since such solvent systems have been successfully employed in enzymatic hydrolysis
to improve the enantioselectivity.15–18 The results are summarized in Table 2.
Fortunately, some of the solvents increased the enantioselectivity remarkably (E>37) as well as the
chemical yield. In particular, polar solvents such as THF (entry 4), acetone (entries 5 and 6) and dioxane
(entry 3) gave favorable results. The amount of water in the organic solvent dramatically affected the
enantioselectivity. For example, an increase in the water content in acetone from 5% to 10% improved the
enantioselectivity (entry 5: 89% ee, E=78; entry 6: 94% ee, E=115), but an increase to 30% remarkably
decreased it (entry 7: 76% ee, E=14).
The typical procedure used for this enzymatic hydrolysis is as follows: to a solution of racemic 4
(9.00 g, 43.9 mmol) in THF (400 ml) containing 5% water (20 ml) was added Novozym® 435 (1.60
g). The mixture was gently stirred at 30°C for 72 h, when the progress of the reaction stopped (usually

4298
S. Katayama et al. / Tetrahedron: Asymmetry 9 (1998) 4295–4299
Table 2
ca. 50% conversion). The reaction mixture was filtered to remove the catalyst beads. After evaporation
of the filtrate, the residue was partitioned between ethyl acetate and aqueous NaHCO3 solution, from
which the (R)-carboxylic acid could also be recovered. The organic layer was washed with water and
brine, dried, and evaporated. The residue was purified by flash chromatography to give (S)-4 (4.53 g,
93% ee, 50% yield, E=94) as a colorless oil. Compound (S)-4 formed a salt with d-camphorsulfonic
acid. Recrystallization of the salt from ethyl acetate, followed by desalination enhanced the enantiomeric
purity (97.5% ee, [α]D20 +104.7 (c=1.0, CHCl3)).
Acknowledgements
We gratefully acknowledge Dr. H. Yamaga and Mr. K. Shimago of our laboratories for valuable
discussions and encouragement. We thank Dr. A. Inoue of Sumitomo Chemical Co. Ltd for his generous
gift of enzyme samples and helpful suggestions.
References
1. Review: Rothman, S. M.; Olney, J. W. Trends Neurosci. 1995, 18, 57.
2. Reviews: Lipton, S. A.; Rosenberg, P. A. N. Engl. J. Med. 1995, 330, 613.
3. Reviews: Muir, K. W.; Lees, K. R. Stroke 1995, 26, 503.
4. Nagata, R.; Tanno, N.; Kodo, T.; Ae, N.; Yamaguchi, H.; Nishimura, T.; Antoku, F.; Tatsuno, T.; Kato, T.; Tanaka, H.;
Nakamura, M.; Ogita, K.; Yoneda, Y. J. Med. Chem. 1994, 37, 3956.
5. Nagata, R.; Ae, N.; Tanno, N. Bioorg. Med. Chem. Lett. 1995, 5, 1527.
6. Nagata, R.; Kodo, T.; Yamaguchi, H.; Tanno, N. Bioorg. Med. Chem. Lett. 1995, 5, 1533.
7. Paradisi, M. P.; Romeo, A. J. Chem. Soc., Perkin Trans. 1 1976, 596.
8. SM-18400:
(S)-9-chloro-5-[p-aminomethyl-o-(carboxymethoxy)phenylcarbamoylmethyl]-6,7-dihydro-1H,5H-pyrido
[1,2,3-de]quinoxaline-2,3-dione hydrochloride.
9. Nagata, R.; Tanno, N.; Yamaguchi, H.; Kodo, T.; Ae, N.; Tanaka, Y.; Nakamura, M.; Ogita, K.; Yoneda, Y. 210th Am.
Chem. Soc. National Meeting, Chicago 1995, Medi 150.

S. Katayama et al. / Tetrahedron: Asymmetry 9 (1998) 4295–4299
4299
10. Tanaka, Y.; Yasuda, H.; Kato, T.; Maruoka, Y.; Kawabe, A.; Kawashima, C.; Ohtani, K.; Nakamura, M. J. Cereb. Blood.
Flow. Metabol. 1995, 15, suppl. 1, S431.
11. Iwao, M.; Kuraishi, T. J. Heterocycl. Chem. 1978, 15, 1425.
12. Katayama, S.; Yamaga, H.; Nagata, R. JP Patent 8-322591, December 10, 1996.
13. Gilchrist, T. L.; Rahman, A. J. Chem. Soc., Perkin Trans. 1 1998, 1203.
14. HPLC analysis was performed with a Chiralpak AD column (Daicel) using 2-propanol:n-hexane, 1:10 as the eluent.
15. Rakels, J. L. L.; Straathof, A. J. J.; Heijinen, J. J. Tetrahedron: Asymmetry 1994, 5, 93 and references cited therein.
16. Mizuguchi, E.; Achiwa, K. Tetrahedron: Asymmetry 1993, 4, 2303.
17. Mizuguchi, E.; Achiwa, K.; Wakamatsu, H.; Terao, Y. Tetrahedron: Asymmetry 1994, 5, 1407.
18. Kang, S.-K.; Jeon, J.-H.; Yamaguchi, T.; Hong, R.-K.; Ko, B.-S. Tetrahedron: Asymmetry 1995, 6, 97.
19. Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J. J. Am. Chem. Soc. 1982, 104, 7294.