Synthesis of optically active α-methylene β-lactams through lipase-catalyzed kinetic resolution

Article abstract of DOI:10.1021/jo0003089

A convenient method for the preparation of the hitherto unknown chiral α-methylene β-lactam derivatives 5a,b is reported. The optically active α-methylene β-lactams 5a-c, and their corresponding amino acids 6a-c have been readily made available through lipase-catalyzed kinetic resolution in high enantiomeric purity (up to 99% ee). The N-substituted β-lactam derivatives 4a,b and 10 are not accepted by the lipases and were prepared in optically active form by chemical transformation.

Full text of DOI:10.1021/jo0003089

J . Org. Chem. 2000, 65, 4919-4922
4919
Syn th esis of Op tica lly Active r-Meth ylen e â-La cta m s th r ou gh
Lip a se-Ca ta lyzed Kin etic Resolu tion
Waldemar Adam,*,† Peter Groer,† Hans-Ulrich Humpf,‡ and Chantu R. Saha-Mo¨ller†
Institut fu¨r Organische Chemie and Institut fu¨r Lebensmittelchemie, Universita¨t Wu¨rzburg, Am Hubland,
D-97074 Wu¨rzburg, Germany
E-mail: adam@chemie.uni-wuerzburg.de
Received March 6, 2000
A convenient method for the preparation of the hitherto unknown chiral R-methylene â-lactam
derivatives 5a ,b is reported. The optically active R-methylene â-lactams 5a -c, and their corre-
sponding amino acids 6a -c have been readily made available through lipase-catalyzed kinetic
resolution in high enantiomeric purity (up to 99% ee). The N-substituted â-lactam derivatives 4a ,b
and 10 are not accepted by the lipases and were prepared in optically active form by chemical
transformation.
Sch em e 1. Syn th esis of th e Ra cem ic r-Meth ylen e
â-La cta m s 5a ,b
In tr od u ction
Enzymatic transformations of organic substrates have
become a well-established method for the preparation of
optically active compounds.1 Enzymes efficiently catalyze
a broad spectrum of chemical reactions under mild and
environmentally acceptable conditions, often with very
high selectivity. Among these biocatalysts, lipases play
an outstanding role in enzymatic synthesis, since they
are readily available, easy to handle, and require no
cofactors nor any elaborate experimental setup.
Recently we have reported2 the preparation of optically
active R-methylene â-lactones, versatile building blocks
in organic synthesis,3 through lipase-catalyzed kinetic
resolution. The nitrogen analogues of R-methylene â-lac-
tones, in particular, the unprotected (free NH) R-meth-
ylene â-lactams, are so far not known in optically active
form,4 yet they may also serve as chiral synthons in
organic synthesis. Therefore, there is a need for the
* To whom correspondence should be addressed. Fax: +49 931 888
4756. Internet: http://www-organik.chemie.uni-wuerzburg.de.
† Institut fu¨r Organische Chemie.
‡ Institut fu¨r Lebensmittelchemie.
(1) (a) Wong, C. H.; Whitesides, G. M. Enzymes in Synthetic Organic
Chemistry; Pergamon: Oxford, 1994. (b) Drauz, K.; Waldmann, H.
Enzyme Catalysis in Organic Synthesis; VCH: Weinheim, 1995. (c)
Faber, K. Biotransformations in Organic Chemistry, 3rd ed.; Springer-
Verlag: Berlin, 1997.
(2) Adam, W.; Groer, P.; Saha-Mo¨ller, C. R. Tetrahedron: Asymmetry
1997, 8, 833-836.
(3) (a) Adam, W.; Albert, R.; Hasemann, L.; Nava Salgado, V. O.;
Nestler, B.; Peters, E.-M.; Peters, K.; Prechtl, F.; von Schnering, H.
G. J . Org. Chem. 1991, 56, 5782-5785. (b) Adam, W.; Nava Salgado,
V. O. J . Org. Chem. 1995, 60, 578-584.
(4) (a) Adlington, R. M.; Barrett, A. G. M.; Quayle, P.; Walker, A.;
Betts, M. J . J . Chem. Soc., Perkin Trans. 1 1983, 605-611. (b) Moriconi,
E. J .; Kelly, J . F. J . Am. Chem. Soc. 1966, 88, 3657-3658. (c) Moriconi,
E. J .; Kelly, J . F. J . Org. Chem. 1968, 33, 3036-3046. (d) Gu¨rtler, S.;
Ruf, S.; J ohner M.; Otto, H.-H. Pharmazie 1996, 51, 811-815. (e)
Fletcher, S. R.; Kay, T. I. J . Chem. Soc., Chem. Commun. 1978, 903-
904. (f) Mori, M.; Chiba, K.; Okita, M.; Ban, Y. J . Chem. Soc., Chem.
Commun. 1979, 698-699. (g) Mori, M.; Ban, Y. Heterocycles 1985, 23,
317-323. (h) Mori, M.; Kagechika, K.; Tohjima, K.; Shibasaki, M.
Tetrahedron Lett. 1988, 29, 1409-1412. (i) Minami, T.; Ishida, M.;
Agawa, T. J . Chem. Soc., Chem. Commun. 1978, 12-13. (j) Palomo,
C.; Aizpurua, J . M.; Urchegui, R.; Iturburu, M. J . Org. Chem. 1992,
57, 1571-1579. (k) Sasaki, T.; Eguchi, S.; Hirako, Y. Tetrahedron Lett.
1976, 17, 541-544. (l) Anklam, S.; Liebscher, J . Tetrahedron 1998,
54, 6369-6384.
preparation of optically active R-methylene â-lactams,
e.g., through the possibility of enzymatic kinetic resolu-
tion.
Herein we report on the synthesis of two hitherto
unknown racemic derivatives (5a ,b) of R-methylene
â-lactam (Scheme 1) and their kinetic resolution through
lipase-catalyzed hydrolysis.
Furthermore, the preparation of the optically active
N-protected derivatives 4a ,b and 10 is described by
chemical transformations, because lipases do not serve
the purpose.
Resu lts a n d Discu ssion
The racemic R-methylene â-lactams 5a ,b were pre-
pared in a multistep synthesis (Scheme 1) by amidation
of the hydroxy acids 1 with p-anisidine as nitrogen source
and dicyclohexylcarbodiimide (DCC) as coupling agent.
The hydroxy amides 2 were converted to the mesyl
10.1021/jo0003089 CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/19/2000

4920 J . Org. Chem., Vol. 65, No. 16, 2000
Adam et al.
Ta ble 1. Lip a se-Ca ta lyzed Resolu tion of â-La cta m s 5
Sch em e 2. Con figu r a tion a l Assign m en t of th e
â-La cta m s 5a ,b
enantiomeric
excess (% ee)b
Ta ble 2. Lip a se-Ca ta lyzed Resolu tion of Hyd r oxy Ester s
7 on th e P r ep a r a tive Sca le
substrate:
enzyme
amino
acid
6d
time convna lactam
substrate (µmol)/(mg)c (h)
(%)
5
Ee
74
>200
>200
5a
5b
5c
315:250
315:250
315:500
86
24
68
52
50
49
96
99
94
90
98
97
a
The % conversion was calculated from the enantiomeric excess
of the starting material (ees) and the product (eep) according to %
enantiomeric
b
excess (% ee)b
yield (%)c
convn ) ees/(ees + eep) (ref 7). The enantiomeric excess (ee) was
time convna
determined by HPLC (Chiralcel OB-H and Chiralpak AS), error
substrated
(h)
(%)
S-7
R-8
S-7 R-8
Ee
75
>200
ca. 1% of the stated values. c Chirazyme L-2 from Candida
d
antartica, carrier-fixed. 6 was derivatized prior to HPLC analysis
7a
7b
19
91
53
50
99
98
87
98
69
83
78
77
(see Supporting Information). e The enantioselectivity (E) was
calculated (ref 7) from the enantiomeric excess of the starting
material (ees) and the conversion according to
a
b
See Table 1. The enantiomeric excess (ee) was determined
by HPLC (Chiralcel OD), error ca. 1% of the stated values. c Based
on 50% conversion. The substrate-enzyme ratio was 19.3 mmol/
d
E ) ln[(1 - convn)(1 - ees)]/ln[(1 - convn)(1 + ees)]
250 mg. e See Table 1.
amides 3 in good yields with mesyl chloride and triethyl-
amine. Treatment of the mesyl amides 3 with potassium
tert-butoxide afforded the â-lactams 4 also in good yields,
despite the considerable ring strain in the R-methylene
â-lactam. The aryl-substituted lactams 4 were oxida-
tively5 deprotected with ammonium cerium(IV) nitrate6
at subambient temperatures. Due to the lability of these
â lactams, especially toward acids, only moderate yields
were achieved.
The racemic lactam 5b was then submitted to enzyme
screening to select the optimal biocatalyst for its enan-
tioselective hydrolysis. From the numerous enzymes
tested, only the lipase Chirazyme L-2 (from Candida
antartica), which has been successfully employed in the
resolution of R-methylene â-lactones,2 accepted well the
â-lactams 5 as substrates. Chirazyme L-1 (from Burkhold-
eria sp.) and Chirazyme L-6 (from Pseudomonas sp.)
showed no activity, even on prolonged reaction times.
With Chirazyme P-1, a protease, and Penicillin G Ami-
dase (PGA 450) also no conversion of the test substrate
5b was observed.
Since the enzymatic hydrolysis of the lactam 5b with
the lipase L-2 proceeded highly selectively but very slowly
at room temperature, the reaction temperature was
raised to 70 °C to enhance the reaction rate. Fortunately,
even at this relatively high temperature, the activity of
the enzyme persisted sufficiently, but it rapidly lost
activity at higher temperatures. To enhance enzyme
stability and allow ease of separation of the enzyme from
the amino acids 6 in the aqueous phase, the carrier-fixed
L-2 enzyme was used. For the zwitterionic amino acids
6 there is no pH change of the solution as they are formed
and, therefore, no buffer was necessary. In a control
experiment without enzyme, the lactam 5b was not
hydrolyzed under these reaction conditions.
In the enzyme-catalyzed enantioselective hydrolysis of
the lactams 5a -c (Table 1) with L-2 lipase under the
above conditions, essentially perfect kinetic resolution
was achieved for the ethyl derivative 5b, for which
enantiomeric excesses (ee) of 99% for the lactam 5b and
98% for the corresponding amino acid 6b were obtained.
In the case of the sterically less-demanding methyl
derivative 5a , a slightly decreased enantioselectivity was
found, with 96% ee for the residual lactam 5a and 90%
ee for the amino acid 6a . Surprisingly, a significant
longer (ca. 60 h) reaction time was necessary for this
substrate in comparison to that of the ethyl derivative
5b to reach a conversion of about 50%. For the bicyclic
lactam 5c, a conversion of 49% required a 2-fold excess
of the L-2 enzyme and 68 h reaction time, which afforded
ee values of 94% for the lactam 5c and 97% for the amino
acid 6c.
The absolute configuration of the lactams 5a ,b was
assessed by chemical correlation (Scheme 2). For this
purpose, the hydroxy esters 7a ,b were prepared as
optically active starting materials by lipase-catalyzed
kinetic resolution (Table 2), which has been already
reported,8 but substantially improved through the use of
the L-1 enzyme instead of Amano AK. In this way, the
hydroxy esters 7a ,b were obtained with 99% and 98%
ee, while the acetates 8a ,b possessed ee values of 87%
and 98%. Hydrolysis of the esters 7 or the acetates 8 gave
the optically active hydroxy acids 1a ,b, which were then
employed in the lactam synthesis described for the
racemic lactams 5a ,b. Due to the inversion at the
stereogenic center during the ring closure, the S-config-
ured lactams 5a ,b were obtained from the R-configured
hydroxy acids 1a ,b.
The absolute configuration of lactam 5c was estab-
lished by the CD exciton chirality method.9 For this
(5) Romo, D.; Yang H. W. J . A reductive deprotection method with
benzyl hydroxylamine as nitrogen source for lactams was reported
recently. Org. Chem. 1999, 64, 7657-7660.
(6) Kronenthal, D. J .; Han, C. Y.; Taylor, M. K. J . Org. Chem. 1982,
47, 2765-2768.
(7) Chen, C.-S.; Fujimoto, Y.; Girdaukas, G.; Sih, C. J . J . Am. Chem.
Soc. 1982, 104, 7294-7299.
(8) (a) Burgess, K.; J ennings, L. D. J . Org. Chem. 1990, 55, 1138-
1139. (b) Burgess, K.; J ennings, L. D. J . Am. Chem. Soc. 1991, 113,
6129-6139.

Synthesis of Optically Active R-Methylene â-Lactams
J . Org. Chem., Vol. 65, No. 16, 2000 4921
Sch em e 3 Der iva tiza tion of th e La cta m 5c for CD
Sp ectr oscop y
purpose, the benzoyl group, which has already been
employed in the configurational assignment of allylic and
homoallylic alcohols,10 was introduced as second chro-
mophore (Scheme 3). Therefore, the optically active
lactam 5c and amino acid 6c, which were obtained by
kinetic resolution of racemic 5c, were converted first into
the methyl ester with hydrogen chloride in methanol and
then with benzoylimidazole into the benzoic amide 9, the
latter being directly used for the CD measurements.
The UV and CD spectra of both enantiomers of 9 are
shown in Figure 1. The CD spectrum of S-9 revealed a
positive Cotton effect at 233 nm (∆ꢀ ) +8.8), and the R
enantiomer showed a negative one also at 233 nm (∆ꢀ )
-10.5). The CD of amide 9 results from the exciton
coupling between the R,â-unsaturated ester (ca. 210 nm)
1
and the La transition band of the benzoate (ca. 230 nm).
As observed for the optically active amide S-9, the first
Cotton effect at 233 nm is positive if the axes of the two
chromophores possess the sense of a right-handed screw.
Since the exciton chirality depends on the conformation
of the molecule, calculations with Macromodel 5.011
(modified Allinger MM2 force field) were performed to
assess the preferred conformation. From these calcula-
tions it is seen that the two chromophores are oriented
in a clockwise sense to each other in the S enantiomer
and counterclockwise in the R enantiomer (Figure 1),
which results in a positive first Cotton effect for S-9 and
a negative one for R-9, which is in very good agreement
with our experimental data.
F igu r e 1. CD and UV spectra of the amino-ester derivatives
R- and S-9 in acetonitrile and the favored conformation
assessed by computational studies with Macromodel 5.0.
Sch em e 4. P r ep a r a tion of th e Op tica lly Active
N-Ben zyl-Su bstitu ted â-La cta m 10
The results of the CD spectroscopy are also confirmed
by the established enantioselectivity of the enzyme, since
in all our cases the S enantiomer is accepted with high
preference and hydrolyzed, while the optically active R
enantiomer is left behind. Furthermore, the elution order
for S and R enantiomers of the lactams 5a -c on the
chiral HPLC is also in agreement with the conformation
assigned to the 5c derivative.
In the enzymatic resolution of the R-methylene â-lac-
tams, the ring-nitrogen atom of the lactam must not be
substituted for lipase acceptance. This is evidenced by
the fact that neither the para-methoxyphenyl-substituted
lactams 4 nor the benzyl-substituted derivative 10 are
converted under the reaction conditions used for the
lactams 5a -c. Nevertheless, the benzyl-protected lactam
10 may be synthesized from optically active amide 11
according to the route depicted in Scheme 1. The latter
was made available by kinetic resolution with the L-1
enzyme to afford the amide S-11 with an enantiomeric
excess of 99% and the acetate R-12 with 68% ee at a
conversion of 59% (Scheme 4). Alternatively, the amide
S-11 is also accessible by coupling of benzylamine with
the hydroxy acid S-1b.
(9) Harada, N.; Nakanishi, K. Circular Dichroic Spectroscopy -
Exciton Coupling in Organic Stereochemistry; University Science
Books: Mill Valley, 1983.
Con clu sion s
(10) (a) Harada, N.; Iwabuchi, J .; Yokota, Y.; Uda, H.; Nakanishi,
K. J . Am. Chem. Soc. 1981, 103, 5590-5591. (b) Gonnella, N. C.;
Nakanishi, K.; Martin, V. S.; Sharpless, K. B. J . Am. Chem. Soc. 1982,
104, 3775-3776. (c) Humpf, H.-U.; Berova, N.; Nakanishi, K. J . Org.
Chem. 1995, 60, 3539-3542. (d) Hoch, U.; Schreier, P.; Saha-Mo¨ller,
C. R.; Adam, W. Chirality 1997, 9, 69-74.
Optically active R-methylene â-lactams 5, which are
attractive chiral building blocks for organic synthesis,
have been prepared for the first time by lipase-catalyzed
kinetic resolution in nearly enantiomerically pure form.
Although the â-lactams 5 are good substrates for lipase
L-2, the N-protected derivatives 4 are not converted by
(11) All calculations were carried out with MM2 force field in CHCl3;
at least 1000 conformers were searched for each simulation.

4922 J . Org. Chem., Vol. 65, No. 16, 2000
Adam et al.
this enzyme. The latter substrates are accessible in high
optical purity from the corresponding optically active
hydroxy acids 1, which are again readily prepared by
lipase-catalyzed kinetic resolution. Thus, a variety of
optically active R-methylene â-lactams have been made
available for synthetic applications.
der Chemischen Industrie. We also thank Roche Diag-
nostics for the generous gift of enzymes.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails and the full characterization of the amides 2a ,b, the
mesylates 3a ,b, the N-protected â-lactams 4a ,b, the R-meth-
ylene â-lactams 5a ,b, the amino acids 6a ,b, and the ester
amide 12 are given. This material is available free of charge
in the Internet under http://pubs.acs.org.
Ack n ow led gm en t. This work was supported by the
Deutsche Forschungsgemeinschaft (SFB-347, “Selektive
Reaktionen Metall-aktivierter Moleku¨le”) and the Fonds
J O0003089