Lipase-catalyzed kinetic resolution of 3-phenyloxazolidin-2-one derivatives: Cascade O- and N-alkoxycarbonylations

Article abstract of DOI:10.1016/j.catcom.2016.04.009

A lipase-catalyzed, cascade kinetic resolution protocol has been established for the synthesis of 3-phenyloxazolidin-2-one derivatives with up to excellent enantioselectivities (95% ee). Candida antarctica lipase B showed high catalytic activity and stere

Full text of DOI:10.1016/j.catcom.2016.04.009

Catalysis Communications 82 (2016) 11–15
Contents lists available at ScienceDirect
Catalysis Communications
journal homepage: www.elsevier.com/locate/catcom
Short communication
Lipase-catalyzed kinetic resolution of 3-phenyloxazolidin-2-one
derivatives: Cascade O- and N-alkoxycarbonylations
a,
Yang Zhang a, Yan Zhang a, Sheng Xie a, Mingdi Yan a,b, , Olof Ramström
⁎
⁎
a
Department of Chemistry, KTH — Royal Institute of Technology, Teknikringen 30, S-10044 Stockholm, Sweden
Department of Chemistry, University of Massachusetts Lowell, 1 University Avenue, MA 01854, Lowell, USA
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 7 March 2016
Received in revised form 15 April 2016
Accepted 16 April 2016
Available online 19 April 2016
A lipase-catalyzed, cascade kinetic resolution protocol has been established for the synthesis of 3-
phenyloxazolidin-2-one derivatives with up to excellent enantioselectivities (95% ee). Candida antarctica lipase
B showed high catalytic activity and stereoselectivity in sequential O- and N-alkoxycarbonylation processes.
© 2016 Elsevier B.V. All rights reserved.
Keywords:
Oxazolidinone
Kinetic resolution
Asymmetric, synthesis
Candida antarctica lipase B
Cascade alkoxycarbonylation
1. Introduction
medically relevant structures, since the N-aryl group plays a crucial
role in antibiotic activity [22]. This challenge has been addressed in
3-Oxazolidinone-2-ones constitute a class of heterocyclic organic
compounds of medical importance. These structures are commonly
used as efficient antimicrobials against both Gram-positive and Gram-
negative pathogens (Fig. 1) [1]. For instance, befloxatone is used clini-
cally for the treatment of dysthymia [2], while linezolid and tedizolid
are commercial drugs to treat serious infections caused by Gram-
positive bacteria, especially for those resistant to many other antibiotics
[3–4]. Oxazolidinone derivatives are also used as Evans auxiliaries for
asymmetric synthesis [5].
the present study, where we present the synthesis of enantiopure 3-
phenyloxazolidin-2-one structures using Candida antarctica lipase B
(CAL-B).
2. Experimental section
2.1. General
To obtain enantioenriched oxazolidinone derivatives, most reported
studies involve chiral transition metal catalysts, organocatalysts or chi-
ral starting materials [6–8]. Meanwhile, biocatalysis has been gaining
importance in asymmetric synthesis over the past decades for its high
chemo-, regio-, and stereoselectivity, mildness of reaction conditions
and environmental friendliness, thus showing advantages over other
catalytic protocols [9–11]. Certain enzymes, such as lipases [12–14],
have also been reported in the asymmetric synthesis of different hetero-
cyclic structures, thus demonstrating the utility of biocatalysis for this
class of compounds [15–20]. In a previous study, an enzyme-catalyzed
kinetic resolution protocol using a Burkholderia (Pseudomonas) cepacia
lipase (BCL) preparation was established for the asymmetric synthesis
of 5-phenyloxazolidin-2-ones with high enantiopurities from 1,2-
aminoalcohols [21]. However, 3-phenyloxazolidin-2-ones are more
Reagents were obtained from commercial suppliers and used as re-
ceived. Lipases from B. (Pseudomonas) cepacia (PS-CI and PS-CII),
C. antarctica (CAL-B), Pseudomonas fluorescens (PFL), Rhizopus niveus
(RNL), Mucor javanicus (MJL), and porcine pancreas, type II (PPL) were
purchased from Sigma-Aldrich. Lipase PS “Amano” IM (EC 3.1.1.3) was
purchased from Amano Enzyme Inc. 1H and 13C NMR data were re-
corded on a Bruker Avance 400 (100) MHz and/or a Bruker Avance
500 (125) MHz, respectively. Chemical shifts are reported as δ values
(ppm) with CHCl3/CDCl3 (1H NMR δ 7.26, 13C NMR δ 77.0) as internal
standards. J values are given in Hertz (Hz). Analytical high performance
liquid chromatography (HPLC) with chiral stationary phase was per-
formed on an HP-Agilent 1110 Series controller and a UV detector,
using Daicel Chiralpak OD-H column (4.6 × 250 mm, 10 μm)/Astec Cel-
lulose DMP column (4.6 × 250 mm, 5 μm)/Daicel Chiralpak OJ column
(4.6 × 250 mm, 10 μm). Solvents for HPLC use were of spectrometric
grade. Thin layer chromatography (TLC) was performed on precoated
Polygram® SIL G/UV 254 silica plates (0.20 mm, Macherey-Nagel),
⁎
Corresponding authors.
E-mail addresses: Mingdi_Yan@uml.edu (M. Yan), ramstrom@kth.se (O. Ramström).
http://dx.doi.org/10.1016/j.catcom.2016.04.009
1566-7367/© 2016 Elsevier B.V. All rights reserved.

12
Y. Zhang et al. / Catalysis Communications 82 (2016) 11–15
Fig. 1. Commercial drugs with 3-oxazolidinone-2-one structures.
Scheme 1. Reaction protocol.
visualized with UV-detection. Flash column chromatography was per-
was removed in vacuo. The crude products were purified using column
formed on silica gel 60, 0.040–0.063 mm (SDS).
chromatography (Hexane:EtOAc = 3:1).
2.2. General procedure for the synthesis of 1,2-aminoalcohols (1a-1k)
2.4. General procedure for kinetic resolutions of 3-phenyloxazolidin-2-one
derivatives (4a-4k) [23–27]
LiBr (0.1 mmol) was added to a stirred mixture of respective epoxide
(2 mmol) and aniline (2 mmol) under nitrogen atmosphere, and the so-
lution was stirred at r.t. for 4 h. The reaction mixture was diluted with
water (10 mL) and extracted twice with diethyl ether (15 mL each).
The combined organic layer was dried over MgSO4 and the solvent
was removed in vacuo. The crude products were purified using column
chromatography (Hexane:EtOAc = 3:1).
1,2-Aminoalcohol 1a-1k (0.05 mmol), diphenyl carbonate
(0.15 mmol) and dry toluene (0.6 mL) were added into a 1.5 mL
sealed-cap vial containing CAL-B (30 mg) and 4 Å molecular sieves
(20 mg). CAL-B was dried under vacuum for at least two days before
use. The vial was kept at r.t. without stirring, and 1H NMR was used to
monitor the reaction progress. After a specific time, the reaction mixture
was filtered through a cotton-stoppered pipette. CH2Cl2 was subse-
quently added, and the aqueous layer was extracted three times with
CH2Cl2 (3 mL each). The combined organic layer was dried over
MgSO4 and the solvent was removed in vacuo. The crude products
were purified using column chromatography (Hexane:EtOAc = 3:1).
Intermediates and the recovered starting materials were also collected.
2.3. General procedure for the synthesis of racemic 3-phenyloxazolidin-2-
one derivatives (4a-4k) [23–27]
1,2-Aminoalcohol 1a-1k (0.08 mmol), diphenyl carbonate
(0.1 mmol), NaH (0.16 mmol, 60% in oil) and THF (1 mL) were added
into a 5 mL flask. The solution was stirred at r.t. for 6 h. TLC was used
to monitor the reaction progress. CH2Cl2 was subsequently added, and
the aqueous layer was extracted three times with CH2Cl2 (3 mL each).
The combined organic layer was dried over MgSO4 and the solvent
Table 2
Effects of solvents and additives on the reaction rate and ee.a
Table 1
Screening of lipases with model reaction.a
Entry
Lipase
Additive
Solvent
Time
3a conv.
(eeb)
4a conv.
(eeb)
Entry
Lipase
4a conv. (%)b
1c
2
3
4
5
6
7
8
PS-IM
CAL-B
CAL-B
CAL-B
CAL-B
CAL-B
CAL-B
CAL-B
SiO2
–
TBME
TBME
46 h
48 h
5 h
33 h
5.5 h
7.5 h
5 h
55 (69)
15 (34)
0
55 (88)
0
8 (82)
10 (25)
0
45 (57)
48 (50)
51 (77)
45 (15)
50 (70)
50 (71)
46 (86)
0
1
2
3
4
5
6
7
8
PS-CI
PS-CII
PS-IM
CAL-B
PFL
RNL
MJL
PPL
52
0
32
71
15
0
–
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
NEt3
1-methylimidazole
4-methylmorpholine
SiO2
ZnBr2
28 h
0
0
a
Reaction conditions: Compound 1a (0.05 mmol), compound 2 (0.15 mmol), additives
(0.025 mmol), lipase preparation (PS-IM: 50 mg; CAL-B: 30 mg), 4 Å molecular sieves (20
mg) in toluene (0.6 mL), 30 °C.
a
Determined by 1H NMR.
b
Reaction conditions: 1a (0.05 mmol), 2 (0.15 mmol), lipase preparation (30 mg), 4 Å
molecular sieves (20 mg) in TBME (0.6 mL), 30 °C for 3 d.
b
Determined by HPLC analysis using Chiralpak OD-H chiral column.

Y. Zhang et al. / Catalysis Communications 82 (2016) 11–15
13
Table 3
Kinetic resolution of different substrates.a
Entry
1
Substrate
Time
5 h
Conv.b (%)
56
Yieldc (%)
46
eed (%)
1
3
4
67
25
86
2
3
4
5
4.5 h
6 h
47
63
51
67
47
45
51
50
34
68
65
92
–
81
76
84
67
58
–
7 h
10 h
44
6
7
4 d
6 h
N99
43
41
–
79
24
95
41
28
–
8
9 h
43
43
29
–
93
9
6 d
5 d
25
43
25
43
8
–
–
34
68
10
29
11
5 d
0
0
–
–
–
a
Total conversion of starting material 1. Determined by 1H NMR spectroscopy.
Yield from starting material 1 to product 4.
Determined by HPLC analysis using Chiralpak OD-H or OJ chiral column; (−) no compound observed at this conversion.
Reaction conditions: Compound 1 (0.05 mmol), compound 2 (0.15 mmol), SiO2 (20 mg), CAL-B (30 mg), 4 Å molecular sieves (20 mg) in toluene (0.6 mL), 30 °C.
b
c
d
3. Results and discussion
carbonate as acyl donor as indicated in Scheme 1. 1-
(Phenylamino)propan-2-ol (1a) was chosen as the model substrate to
evaluate the catalytic efficiency of preparations of lipases from B. cepacia
(PS-CI, PS-CII and PS-IM), C. antarctica (CAL-B), P. fluorescens (PFL),
The protocol for the synthesis of enantiopure oxazolidinones was
based on 1.2-aminoalcohols as starting materials and diphenyl

14
Y. Zhang et al. / Catalysis Communications 82 (2016) 11–15
Table 4
entries 2 and 3). Additives were next applied to improve the
enantioselectivity. Weak Lewis bases, for example, 1-methylimidazole
and 4-methylmorpholine lowered the reaction rate and the selectivity
slightly, while NEt3, a relatively stronger organic base, resulted in
much slower transformation, accumulation of the singly acylated inter-
mediate 3a (see SI) and production of oxazolidinone 4a of low
enantiopurity. SiO2 has also proven to be an efficient additive, resulting
in improved ee while leaving the conversion rate unaffected (Table 2;
entry 7) [21]. Optimization of the amount of SiO2 and temperature did
not display further improvement in ee (Table S1). However, when the
Lewis acid ZnBr2 (0.5 eq.; Table 2, entry 8) was added to the system, nei-
ther product 4a nor intermediate 3a was detected.
Subsequently, the substrate scope of the protocol was evaluated.
Table 3 shows the enantiomeric excesses of products 4, as well as inter-
mediates 3, and the recovered starting material when the formation of
compound 4 reached ~50% conversion. Substitutions on the para-
positions of the phenyl ring showed good compatibility for both
electron-withdrawing (−Cl, 1c; −F, 1d) and electron-donating
(−OMe, 1b) groups. The decrease in ee, compared to the unsubstituted
compound 1a, indicate that electronic effects influenced the process in
addition to steric constraints. Fluorine-substituted derivatives were fur-
ther used to evaluate the positional effects of the substituents (Table 3;
entries 4–6). Although the rate dropped significantly in the series from
the para-F structure (7 h, 51% conversion), through the meta-F structure
(10 h, 50% conversion) to the ortho-F compound (4d, 43% conversion),
the meta-F product 4e could be isolated with a synthetically useful
enantiopurity of 67% ee. The ortho-substitution seemed to influence
the enzymatic N-alkoxycarbonylation significantly, since the ee was re-
duced from 79% (3f) to 24% (4f).
Extension of the alkyl chain from a methyl- to an ethyl group (1g)
led to a slight decrease in reaction rate but a surprisingly excellent ee
(95%; Table 3, entry 7). Moreover, an excellent enantioselectivity (93%
ee, 4h) remained for the para-Cl-substituted starting material. The sub-
strate with a chloromethyl group (1j), which has a similar size as the
ethyl substrate (1g), showed a much slower conversion, however
resulting in a synthetically useful ee. Product 4j is of special interest,
since it constitutes an important intermediate towards many pharma-
ceutical drugs [28]. A longer alkyl chain than an ethyl group resulted
in very slow conversion (Table 3, entry 9). Furthermore, substrate 1k,
carrying a phenyl group, resulted in no reaction with CAL-B, although
a good ee (90%) could be obtained with PS-IM [21].
Effects of enzymes and additives on reaction rate and ee during the cyclization step.a
Entry
Additive
Time
4a eeb
3a eeb
(Conv.)
Recovered
1
2
3
4
5
6
7
–
5 d
60 (41)
75 (47)
80 (43)
76 (40)
65 (45)
64 (46)
61 (42)
57
50
47
49
41
53
42
CAL-B
PS-IM
PS-CI
PFL
SiO2
NEt3
40 min
41 h
23 h
26 h
27 h
41 h
a
Determined by HPLC analysis using Chiralpak OD-H chiral column.
Reaction conditions: Compound 3 (0.05 mmol), 4 Å molecular sieves (20 mg), en-
b
zyme (30 mg)/SiO2 (20 mg)/NEt3 (0.5 eq) in 0.6 mL toluene, r.t.
R. niveus (RNL), M. javanicus (MJL) and porcine pancreas (PPL) (Table 1).
Among these tested lipase preparations, only PS-Cl, PS-IM, CAL-B and PFL
could catalyze the transformation of 1,2-aminoalcohols. In contrast to the
synthesis of 5-phenyloxazolidin-2-ones [21], the lipases from B. cepacia
however performed ineffectively when applied to 3-phenyloxazolidin-
2-ones, even after extensive optimization (57% ee; Table 2, entry 1).
CAL-B was therefore chosen for further optimization, giving rise to higher
conversion and lower degree of byproducts obtained over time, as moni-
tored by 1H NMR spectroscopy, compared to the other lipases.
Several parameters, including effects from solvents, additives and
temperature, were next evaluated to get optimal conditions of this ki-
netic resolution system (Table 2). Compared with the use of tert-butyl
methyl ether (TBME) as solvent, the reaction in toluene proceeded sig-
nificantly faster and gave higher enantiomeric excess (77% ee; Table 2,
The moderate ees of the recovered starting amino alcohols and inter-
mediates indicated a complex enzymatic process, and further experi-
ments were undertaken to delineate the role of the enzyme in the
catalytic process. It was observed that O-acylated intermediate 3 was
formed with several substrates in the presence of CAL-B in TBME, how-
ever not all of them. Moreover, the O-acylated intermediates were
found to undergo further N-cyclization towards oxazolidinones 4. Con-
trol experiments in the absence of enzyme gave no transformation to ei-
ther intermediates 3 or cyclization products 4, indicating a catalytic
effect of the lipase in the O-alkoxycarbonylation step similar to previous
results with intra- and intermolecular processes [21,29–30]. The ab-
sence of N-alkoxycarbonylated intermediates may be attributed to the
lower nucleophilicity of the aniline nitrogen than the hydroxyl group
under the present conditions. The O-alkoxycarbonylation process also
proved stereoselective, albeit only moderately.
In addition to O-alkoxycarbonylation, lipase-catalyzed intermolecu-
lar N-alkoxycarbonylation has been reported [31], whereas intramolec-
ular cyclization remained to be shown. To further study the
intramolecular N-alkoxycarbonylation cyclization in the present pro-
cess, the isolated O-alkoxycarbonylated intermediate 3a (Table 2,
entry 1) with 69% ee was evaluated under the conditions displayed in
Table 4. The oxazolidinone 4a was isolated in yields of 40–50%, and
the enantiopurities of compounds 4a and recovered 3a were analyzed.
Acid and base accelerated the conversion, but the most significant rate
enhancement was observed with lipases. CAL-B showed the fastest
Scheme 2. Proposed mechanism.

Y. Zhang et al. / Catalysis Communications 82 (2016) 11–15
15
cyclization rate with 47% conversion to product within 20 min, com-
References
pared to the other lipases (Table 4, entries 2–5). This intramolecular
N-cyclization is considerably faster than corresponding intermolecular
N-acylations [32–35]. More interestingly, the 75% ee obtained, which
is higher than 69% ee of the starting material, together with lower ees
of recovered compound 3, indicated stereoselectivity of CAL-B in the
secondary N-alkoxycarbonylation cyclization rather than a statistic ef-
fect of two enantiomers of substrates. This effect was also confirmed
with PS-IM and PS-CI. In the presence of acid or base, regeneration of
starting material 1, i.e. hydrolysis of compound 3, could be observed in
small amounts, which explained the slight decrease in the ee of product
4.
Based on these results, a lipase-catalyzed double alkoxycarbonylation
mechanism can be proposed (Scheme 2). Specifically, 1,2-aminoalcohols
1 first undergo lipase-catalyzed alkoxycarbonylation at oxygen with car-
bonate diester 2 to get intermediate 3. The enzymes then catalyze a very
fast secondary stereoselective N-alkoxycarbonylation-cyclization to pro-
vide 3-phenyloxazolidin-2-one products.
[1] D.J. Diekema, R.N. Jones, Lancet 358 (2001) 1975–1982.
[2] D. Warot, I. Berlin, A. Patat, G. Durrieu, I. Zieleniuk, A.J. Puech, J. Clin. Pharmacol. 36
(1996) 942–950.
[3] J. Weigelt, K. Itani, D. Stevens, W. Lau, M. Dryden, C. Knirsch, The Linezolid CSSTI
Study Group, Antimicrob. Agents Chemother. 49 (2005) 2260–2266.
[4] P. Prokocimer, P. Bien, C. Deanda, C.M. Pillar, K. Bartizal, Antimicrob. Agents
Chemother. 56 (2012) 4608–4613.
[5] M.M. Heravi, V. Zadsirjan, Tetrahedron Asymmetry 24 (2013) 1149–1188.
[6] S. Jegham, A. Nedelec, P. Burnier, Y. Guminski, F. Puech, J.J. Koenig, P. George, Tetra-
hedron Lett. 39 (1998) 4453–4454.
[7] C. Larksarp, H. Alper, J. Am. Chem. Soc. 119 (1997) 3709–3715.
[8] V.B. Birman, H. Jiang, X. Li, L. Guo, E.W. Uffman, J. Am. Chem. Soc. 128 (2006)
6536–6537.
[9] G. Torrelo, U. Hanefeld, F. Hollmann, Catal. Lett. 145 (2015) 309–345.
[10] P. Vongvilai, R. Larsson, O. Ramström, Adv. Synth. Catal. 350 (2008) 448–452.
[11] L. Hu, F. Schaufelberger, Y. Zhang, O. Ramström, Chem. Commun. 49 (2013)
10376–10378.
[12] U.T. Bornscheuer, R.J. Kazlauskas, Hydrolases in Organic Synthesis: Regio- and
Stereoselective BiotransformationsWILEY-VCH 2006.
[13] J. Uppenberg, N. Oehrner, M. Norin, K. Hult, G.J. Kleywegt, S. Patkar, V. Waagen, T.
Anthonsen, T.A. Jones, Biochemistry 34 (1995) 16838–16851.
[14] J. Ottosson, L. Fransson, K. Hult, Protein Sci. 11 (2002) 1462–1471.
[15] M. Sakulsombat, Y. Zhang, O. Ramström, Chem. Eur. J. 18 (2012) 6129–6132.
[16] L. Hu, Y. Zhang, O. Ramström, Org. Biomol. Chem. 12 (2014) 3572–3575.
[17] L. Hu, Y. Ren, O. Ramström, J. Org. Chem. 80 (2015) 8478–8481.
[18] Y. Zhang, F. Schaufelberger, M. Sakulsombat, C. Liu, O. Ramström, Tetrahedron 70
(2014) 3826–3831.
[19] Y. Zhang, O. Ramström, Chem. Eur. J. 20 (2014) 3288–3291.
[20] D. Heine, T. Bretschneider, S. Sundaram, C. Hertweck, Angew. Chem. Int. Ed. 53
(2014) 11645–11649.
[21] Y. Zhang, Y. Zhang, Y. Ren, O. Ramström, J. Mol. Catal. B Enzym. 122 (2015) 29–34.
[22] G. Zappia, C. Ingallina, F. Ghirga, B. Botta, 13. Oxazolidin-2-Ones: Antibacterial Activ-
ity and Chemistry, 2014 247–266.
[23] T. Baronsky, C. Beattie, R.W. Harrington, R. Irfan, M. North, J.G. Osende, C. Young, ACS
Catal. 3 (2013) 790–797.
[24] K.S. Gates, R.B. Silverman, J. Am. Chem. Soc. 111 (1989) 8891–8895.
[25] X. Zhang, W. Chen, C. Zhao, C. Li, X. Wu, W.Z. Chen, Synth. Commun. 40 (2010)
3654–3659.
[26] C. Beattie, M. North, RSC Adv. 4 (2014) 31345–31352.
[27] P. ten Holte, L. Thijs, B. Zwanenburg, Org. Lett. 3 (2001) 1093–1095.
[28] B. Mallesham, B.M. Rajesh, P.R. Reddy, D. Srinivas, S. Trehan, Org. Lett. 5 (2003)
963–965.
[29] F. Morís, V. Gotor, Tetrahedron 49 (1993) 10089–10098.
[30] L.F. García-Alles, V. Gotor, J. Mol. Catal. B Enzym. 6 (1999) 407–410.
[31] R. Morán-Ramallal, V. Gotor-Fernández, P. Laborda, F.J. Sayago, C. Cativiela, V. Gotor,
Org. Lett. 14 (2012) 1696–1699.
4. Conclusions
A kinetic resolution process has been developed for the synthesis of
enantiopure 3-phenyloxazolidin-2-one derivatives using CAL-B. This
protocol showed good substrate tolerance and enabled access to phar-
maceutically potentially important oxazolidinone intermediates. Fluori-
nated oxazolidin-2-ones, for example, improve the antibiotic activity of
oxazolidinone derivatives [22]. The reaction process proceeded step-
wise, where the O-acylated intermediate was detected in the first
alkoxycarbonylation step and various additives were compared for
their effects on the N-alkoxycarbonylation/ring closing process. Both
alkoxycarbonylation processes were stereoselectively catalyzed by the
enzyme, thus representing a unique case in lipase catalysis. Studies
are ongoing in developing a dynamic kinetic resolution process from
this protocol.
Acknowledgements
[32] M.A.J. Veld, K. Hult, A.R.A. Palmans, E.W. Meijer, Eur. J. Org. Chem. 2007 (2007)
5416–5421.
[33] P. Vongvilai, O. Ramström, J. Am. Chem. Soc. 131 (2009) 14419–14425.
[34] K.P. Gustafson, R. Lihammar, O. Verho, K. Engström, J.-E. Bäckvall, J. Org. Chem. 79
(2014) 3747–3751.
This work was in part supported by the Royal Institute of Technol-
ogy. Yan Z., Yang Z. and S.X. thank the China Scholarship Council for spe-
cial scholarship awards.
[35] Y. Kim, J. Park, M.J. Kim, Tetrahedron Lett. 51 (2010) 5581–5584.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.catcom.2016.04.009.