A short and efficient synthesis of the NMDA glycine site antagonist: (3R,4R)-3-amino-1-hydroxy-4-methyl pyrrolidin-2-one (L-687,414)

Article abstract of DOI:10.1016/j.tetlet.2008.07.172

A short and efficient synthesis of the NMDA glycine site antagonist: (3R,4R)-3-amino-1-hydroxy-4-methyl pyrrolidin-2-one (L-687,414) is described.

Full text of DOI:10.1016/j.tetlet.2008.07.172

Tetrahedron Letters 49 (2008) 6079–6080
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
A short and efficient synthesis of the NMDA glycine site antagonist:
(3R,4R)-3-amino-1-hydroxy-4-methyl pyrrolidin-2-one (L-687,414)
a
b
b
a
Emmanuel Pinard a, , Serge Burner , Philippe Cueni , François Montavon , Daniel Zimmerli
*
a F. Hoffmann-La Roche Ltd, Pharmaceutical Research Basel, Discovery Chemistry, CH-4070 Basel, Switzerland
b F. Hoffmann-La Roche Ltd, Pharmaceutical Research Basel, Chemical Synthesis, CH-4070 Basel, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
A short and efficient synthesis of the NMDA glycine site antagonist: (3R,4R)-3-amino-1-hydroxy-4-
methyl pyrrolidin-2-one (L-687,414) is described.
Received 4 July 2008
Accepted 30 July 2008
Available online 6 August 2008
Ó 2008 Elsevier Ltd. All rights reserved.
L-687,414, (3R,4R)-3-amino-1-hydroxy-4-methyl-pyrrolidin-2-
one (Scheme 1) is an antagonist at the glycine modulatory site of
As a result, L-687,414 has proved to be a key tool for studying
the behavioral relevance of the glycine site on the NMDA receptor.5
However, further in vivo characterization has been hampered by
the lack of efficient synthetic route permitting facile gram quantity
production of this important CNS agent. To our knowledge, three
syntheses of L-687,414 have been reported till now: from
Merck,1,6,7 Pfizer,8 and from the Baldwin group.9 The Merck and
Baldwin routes are lengthy (10 and 13 steps, respectively) whereas
the N-methyl-D-aspartate (NMDA) receptor discovered at Merck
Sharp and Dohme.1 In contrast to many reported glycine site
NMDA receptor antagonists, L-687,414 displays CNS-mediated ac-
tions after systemic administration. Indeed, robust anticonvulsant2
and neuroprotective3 properties of L-687414 injected intraperito-
neally have been reported, suggesting good brain penetration.4
O
O
O
Merck Synthesis
6 steps, 20%
Ref. 1 and 7
O
(iii)
O
(ii)
OEt
(i)
N
OEt
N
O
O
N
H
O
EtO
O
O
Ref. 10
O
O
2
3
1
4
(iv)
(v)
N
H2N
OH
N
N
N
H
O
N
O
+
H
O
O
O
L-687,414
5a
5b
Scheme 1. Reagents and conditions: (i) Isopropenylmagnesium bromide, diethylether-THF, ꢀ70 °C, 20 min, quantitative yield; (ii) O-benzylhydroxylamine hydrochloride,
diisopropylethylamine, DMF-ethylacetate, 0 °C, then 45 °C, 16 h, 40% yield; (iii) (R)-(+)-1-Phenylethylamine, methanol, reflux, 23 h, 95%; (iv) (a) PtO2, H2 (1 atm), acetic acid,
ethylacetate, 0 °C, 15 h; (b) Chiral preparative HPLC (Chiralpak ADÒ), 40% yield of 5a from 4; (v) Pd(OH)2, H2 (1 atm), methanol, acetic acid, rt, 5 h, 86% yield.
* Corresponding author.
E-mail address: emmanuel.pinard@roche.com (E. Pinard).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.07.172

6080
E. Pinard et al. / Tetrahedron Letters 49 (2008) 6079–6080
the Pfizer synthesis which is reported to be shorter (6 steps) starts
from an optically active synthon which is no longer commercially
available. In addition, the Pfizer route lacks efficient stereochemi-
cal control.
References and notes
1. Leeson, P. D.; Williams, B. J.; Baker, R.; Ladduwahetty, T.; Moore, K. W.; Rowley,
M. J. Chem. Soc., Chem. Commun. 1990, 22, 1578–1580.
2. Tricklebank, M. D.; Bristow, L. J.; Hutson, P. H.; Leeson, P. D.; Rowley, M.;
Saywell, K.; Singh, L.; Tattersall, F. D.; Thorn, L.; Williams, B. J. Br. J. Pharmacol.
1994, 113, 729–736.
We describe here a short and efficient synthesis of L-687,414. In
our approach, we set out to improve the original Merck route. In
the Merck synthesis, key intermediate pyrrolidine-dione 3 was
prepared in 6 steps with only 20% overall yield (4 chromatogra-
phies on silica gel were required).6 In addition, one of the steps
employed the hazardous oxidizing agent 3-phenyl-2-(phenylsulf-
onyl)-1,2-oxaziridine, making it difficult, for safety reason, to per-
form the reaction on a large scale. We found that 3 could
actually be prepared in 2 steps only starting from diethyloxalate
1 with 40% overall yield, and no chromatographic separation. In-
deed, 1 was first reacted with 1.2 equiv of isopropropenylmagne-
sium bromide in diethylether/THF at ꢀ70 °C as described by
Villieras et al.10 to provide ethyl-3-methyl-2-oxo-but-3-enoate
intermediate 2 in quantitative yield. Remarkably, under these con-
ditions, the Grignard reagent added to only one carbonyl function
and no subsequent addition on the formed keto group (via 1,2-
addition) or on the double bond (via 1,4-addition) of the product
was observed. Pleasingly, the obtained intermediate 2 reacted
smoothly with O-benzylhydroxylamine to provide, after a Michael
addition followed by an in situ cyclization, the desired pyrrolidine-
dione 3 in 40% yield.11 This sequence could be performed easily on
a large scale (>100 g). Product 3 was isolated upon crystallization
of the crude material, and purification via chromatography was
therefore not necessary. From intermediate 3 to the final com-
pound L-687414, we improved the sequence described by Merck:
First, the formation of enamine 4 with (R)-(+)-1-phenylethylamine
proceeded with excellent yield (95%).12 Next, the hydrogenation of
4 with platinum oxide was performed at 0 °C instead of at room
temperature as described by Merck. Under these milder condi-
tions, the cleavage of the benzyl ether protective group was not
observed. Such cleavage was reported by Merck to take place at
room temperature, necessitating a subsequent reprotection of
the hydroxylamine function. In addition, and pleasingly, a signifi-
cant improvement in the diastereoselectivity of the hydrogenation
step in favor of the desired stereoisomer 5a versus 5b was ob-
served when the reaction was performed at 0 °C (4:1 mixture ob-
tained) instead of at room temperature (2:1 mixture obtained).13
Furthermore, in our hands, the separation of this mixture on
Waters PrepLC as described by Merck proved to be extremely inef-
ficient due to poor separation. We found that much better discrim-
ination between the two diastereoisomers was achieved by
preparative chiral HPLC (Chiralpak ADÒ). Finally, deprotection of
the required diastereoisomer 5a with Pearlman’s catalyst under
an atmospheric pressure of hydrogen provided L-687,414 as its
acetic acid salt in excellent yield (86%).14
3. Gill, R.; Hargreaves, R. J.; Kemp, J. A. J. Cerebral Blood Flow Metab. 1995, 15, 197–
204.
4. Leeson, P. D.; Williams, B. J.; Rowley, M.; Moore, K. W.; Baker, R.; Kemp, J. A.;
Priestley, T.; Foster, A. C.; Donald, A. E. Bioorg. Med. Chem. Lett. 1993, 3, 71–76.
5. Kemp, J. A.; Leeson, P. D. Trends Pharmacol. Sci. 1993, 14, 20–25.
6. Baker, R.; Ladduwahetty, T.; Leeson, P. D.; Williams, B. J. E. P. Patent 362941,
1990; Chem. Abstr. 1991, 114, 6289.
7. Rowley, M.; Leeson, P. D.; Williams, B. J.; Moore, K. W.; Baker, R. Tetrahedron
1992, 48, 3557–3570.
8. Fray, M. J.; Bull, D. J.; James, K. Synlett 1992, 9, 709–710.
9. Baldwin, J. E.; Adlington, R. M.; Elend, A. S.; Smith, M. L. Tetrahedron 1995, 51,
11581–11594.
10. Rambaud, M.; Bakasse, M.; Duguay, G.; Villieras, J. Synthesis 1988, 564–566.
11. 1-Benzyloxy-4-methyl-pyrrolidine-2,3-dione 3: To
a solution of o-benz-
ylhydroxylamine hydrochloride (119 g, 0.73 mol) in DMF (195 ml) and
ethylacetate (780 ml) was added dropwise N-ethyldiisopropylamine (182 ml,
1.0 mol). After stirring for 1 h at rt, the reaction mixture was cooled to 0 °C, and
2 (136 g, 0.73 mol) was added dropwise. After stirring for 1 h at 0 °C and 4 h at
RT, the reaction mixture was heated at 45 °C, stirred for 16 h, then quenched
with 10% citric acid (1.3l). The aqueous phase was extracted three times with
tert-butylmethylether, the combined organic phases were washed successively
with 1 M H2SO4, sat. NaCl, and satd NaHCO3, then dried over Na2SO4 and
concentrated. The residue was crystallized from tert-butylmethylether (1 l),
dichloromethane (300 ml) and heptane (400 ml) to provide 3 as a white solid
(64 g, 40% yield), mp: 130–132 °C (Ref.: mp: 129–132 °C),7 existing
predominantly at RT as its enol form in CDCl3. 1H NMR (CDCl3) of enol: 1.81
(s, 3H), 3.61 (s, 2H), 5.02 (s, 2H), 6.48 (br s, 1H), 7.3–7.5 (m, 5H). MS: m/z (%):
219 (5%, M+), 91 (100%, Bn).
12. (R)-1-Benzyloxy-4-methyl-3-(1-phenyl-ethylamino)-1,5-dihydro-pyrrol-2-one 4:
A
solution of 3 (40 g, 0.18 mol) and (R)-(+)-1-phenylethylamine (46.9 g,
0.36 mol) in MeOH (500 ml) was refluxed under N2 for 23 h. The solvent was
evaporated in vacuo and the residue was purified by flash chromatography on
silica gel (Eluent: heptane, ethylacetate 9:1) to provide 4 as a light yellow oil
(56 g, 95% yield), ½a D20
ꢁ
+56.5 (c 0.88, CHCl3) (Ref.: ½a D20
ꢁ
+59.7 (c 1.8, CHCl3)).7 1
H
NMR (CDCl3): 1.45 (d, J = 6.7 Hz, 3H), 1.60 (s, 3H), 3.42 (d, J = 16.2 Hz, 1H), 3.50
(d, J = 16.2 Hz, 1H), 4.14 (br s, 1H), 4.64 (q, J = 6.4, 1H), 5.0 (s, 2H), 7.2–7.5 (m,
10H). MS: m/z (%): 323 (100%, M+H+).
13. (3R,4R)-1-Benzyloxy-4-methyl-3-(1-phenyl-ethylamino)-pyrrolidin-2-one 5a: To
a solution of 4 (26.5 g, 82 mmol) in ethyl acetate (772 ml) were added acetic
acid (80 ml) and Platinum (IV) oxid hydrate (1.3 g). The mixture was stirred
with
a mechanical stirrer and hydrogenated at 0 °C under atmospheric
pressure for 15 h. The catalyst was filtered and the solvent evaporated in
vacuo. The residue was azeotroped twice with toluene and then dissolved in
ethyl acetate (200 ml). The solution was washed once with sat. NaHCO3
(100 ml), dried over Na2SO4, filtered, and the solvent was evaporated in vacuo.
The residue was purified by flash chromatography on silica gel (Eluent:
heptane, ethylacetate 8:2) to provide 14.8 g of a yellow oil containing a 4/1
mixture of the diastereoisomers 5a and 5b. This mixture was separated by
preparative chiral HPLC (column: Chiralpak ADÒ, eluent:heptane, isopropanol
85:15) to provide 5a (less polar) as a light yellow oil (10 g, 40% yield) ½a D20
ꢁ
+104.5 (c 1.16, CHCl3). (Ref.: ½a D20
ꢁ
+103 (c 1.6, CHCl3)).7 1H NMR (CDCl3): 0.83
(d, J = 6.9 Hz, 3H), 1.38 (d, J = 6.6 Hz, 3H), 1.67 (br s, 1H), 1.90–1.98 (m, 1H),
2.75 (dd, J = 1.8 and 8.7 Hz, 1H), 3.16 (d, J = 7.5, 1H), 3.22 (dd, J = 6.0 and 8.7 Hz,
1H), 4.18 (q, J = 6.6, 1H), 4.95 (d, J = 10.8 Hz, 1H), 4.99 (d, J = 10.8 Hz, 1H), 7.22–
7.41 (m, 10H). MS: m/z (%): 325 (100%, M+H+).
14. (3R,4R)-3-Amino-1-hydroxy-4-methyl-pyrrolidin-2-one acetic acid salt-L-
687,414: To a solution of 5a (9.8 g, 30 mmol) in methanol (150 ml) and acetic
acid (3.2 ml) was added palladium hydroxide on charcoal 20% (2.4 g). The
reaction mixture was hydrogenated under atmospheric pressure at rt for 5 h.
The catalyst was filtered and the solvent was evaporated in vacuo. The residue
was azeotroped three times with toluene. The colorless oil was crystallized
from methanol (30 ml) and ether (30 ml). The white solid was stirred in ether
(30 ml) for 2 h, filtered, and dried under high vacuum to provide L-687,414 as
In summary, a short (5 steps) and efficient synthesis of the
NMDA glycine site antagonist L-687,414 has been established by
improving the original Merck route. Key in our approach was the
development of an extremely facile access to the intermediate pyr-
rolidine-dione 3 via a Michael-in situ cyclization protocol. With
this synthetic route, multigram quantity of L-687,414 could be pre-
pared making possible further in vivo characterization of this
important CNS agent.
its acetic acid salt (4.92 g, 86%). mp: 110–113 °C. ½a D20
ꢁ
+17 (c 0.932, MeOH). 1
H
NMR (D2O): 1.08 (d, J = 7.3 Hz, 3H), 1.85 (s, 3H), 2.80–2.84 (m, 1H), 3.25 (dd,
J = 3.2 Hz and 9.7 Hz, 1H), 3.80 (dd, J = 6.8 Hz and 9.7 Hz, 1H), 4.13 (d, J = 8.3 Hz,
1H), 4.71 (s, 4H). MS: m/z (%): 130 (5%, M+), 113 (100%, M-NH3). Microanalysis:
calcd (1/1 acetic acid salt): C: 44.20, H: 7.42, N: 14.73, found: C: 44.17, H: 7.31,
N: 14.87. The free base was obtained as a white foam (97.4% yield) after
chromatography on Si–tosic acid silica gel (Silicycle R60430B, eluent:
Acknowledgments
methanol, then NH3 in methanol, 2N): ½a D20
ꢁ
+17.8 (c 0.81, MeOH)). (Ref.: ½a D20
ꢁ
+16.5 (c 0.48, MeOH)).7 1H NMR (D2O): 1.00 (d, J = 7.2 Hz, 3H), 2.64 (m, 1H),
3.13 (dd, J = 3.0 Hz and 9.9 Hz, 1H), 3.68 (dd, J = 7.2 Hz and 10.2 Hz, 1H), 3.80
(d, J = 8.1 Hz, 1H), 4.71 (s, 3H). For practical reason, L-687,414 was used as its
acetic salt for its evaluation in biological testing.
We would like to thank our colleagues from the Molecular
Structure Research Department for NMR, MS and optical rotation
measurement.