Lipase-based HPLC stationary phase: Enantioselective synthesis of 2- substituted 1,3-propanediol monoacetates

Article abstract of DOI:10.1016/S0957-4166(99)00478-4

Pseudomonas cepacia lipase (PCL) has been immobilized by coating the enzyme on an epoxysilica HPLC column. The biocatalyst has been successfully used for the preparation of both the enantiomers of 3-acetoxy-2-benzyl- propan-1-ol and of 3-acetoxy-2-methylpropan-1-ol. The immobilized enzyme is active after months of use either in aqueous or in organic media. (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0957-4166(99)00478-4

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 10 (1999) 4455–4462
Lipase-based HPLC stationary phase: enantioselective synthesis
of 2-substituted 1,3-propanediol monoacetates
Carlo Bertucci,a Antonella Petri,a Guy Felix,b Benedetta Perini a and Piero Salvadori a,∗
aCentro CNR per Macromolecole Stereordinate ed Otticamente Attive, Dipartimento di Chimica e Chimica Industriale,
Università degli Studi di Pisa, Via Risorgimento 35, 56126 Pisa, Italy
bENSCPB, University of Bordeaux, Avenue Pey Berland BP 108 3 3402, Talence Cedex, France
Received 22 October 1999; accepted 27 October 1999
Abstract
Pseudomonas cepacia lipase (PCL) has been immobilized by coating the enzyme on an epoxysilica HPLC
column. The biocatalyst has been successfully used for the preparation of both the enantiomers of 3-acetoxy-2-
benzyl-propan-1-ol and of 3-acetoxy-2-methylpropan-1-ol. The immobilized enzyme is active after months of use
either in aqueous or in organic media. © 1999 Elsevier Science Ltd. All rights reserved.
1. Introduction
The asymmetrization of prochiral or meso diols or diesters together with the resolution of racemic
alcohols or esters by lipases represents one of the most useful methods to prepare enantiomerically pure
compounds. The lipase catalyzed enantioselective esterification in the presence of an acyl donor and
enantioselective hydrolysis have been shown to be efficient methods for the preparation of pharmacolog-
ically active compounds or important chiral building blocks with defined stereochemistry.1–4 These
enzymes are available from different sources and are very stable either in aqueous or organic solvents.
Several attempts have been made to immobilize lipases.5,6 It is well known that immobilization can raise
an enzyme’s potential, such as performance at high substrate concentration, increased operating stability,
feasibility of continuous operation, repeated usage and retention of the enzyme in a bioreactor. Among the
different examples reported, the immobilization of lipases on a membrane allows the enzyme to remain
active in the preparation of enantiomerically pure compounds which is important for pharmaceutical large
scale applications.7,8 Recently a new procedure has been reported in which lipases are incorporated in
hydrophobic organic–inorganic hybrid materials.9 These heterogeneous biocatalysts show much higher
activity for esterifications in organic solvents and remarkably long-term stability. Non-covalent binding
of lipase has also been reported using an immobilized artificial membrane stationary phase (IAM-SP):
the enzyme was active and compatible with the on-line syntheses and purification of stereochemically
∗
Corresponding author. Tel: +39 50 918 273; fax: +39 50 918 260; e-mail: psalva@dcci.unipi.it
0957-4166/99/$ - see front matter © 1999 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(99)00478-4
tetasy 3114

4456
C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
pure products.10 Thus, the key to scale-up a lipase catalyzed enantioselective synthesis can be considered
to reside in the immobilization technology.11 This expectation provides the reasons for the development
of so many methods to immobilise these enzymes. No reports, however, describe the immobilization of
lipase on high performance liquid chromatography (HPLC) supports. We were interested in preparing a
lipase-based HPLC column to be used as a bioreactor, i.e. for the enantioselective synthesis of optically
active compounds, starting either from chiral or prochiral substrates.
In order to evaluate reliably the performances of the immobilized enzyme, reactions were selected
which are well known to occur in bulk solution for the preparation of synthetically useful compounds.
2. Results and discussion
The properties of the support material can influence the activity of the immobilized enzyme and
epoxysilica was chosen because it has been successfully used in the immobilization of proteins12
and enzymes.13 The anchoring of lipase can be realised either in bulk solution or in situ and these
supports have been proved to guarantee high chromatographic performances, once they have been
completely derivatized.14 The column used for this investigation has been prepared by in situ coating
of Pseudomonas cepacia lipase (PCL) on the packed epoxysilica matrix. The epoxysilica support was
prepared by reacting activated silica with (3-glycidoxypropyl)trimethoxysilane according to a reported
procedure.15 An HPLC column (5×0.46 cm i.d.) was then conventionally slurry packed. The enzyme
(100 mg suspended in phosphate buffer 50 mM, pH 7.5) was allowed to coat the epoxysilica matrix.
The obtained lipase-based column was finally stabilized by treating the support with glutaraldehyde as a
cross-linking agent.
Indeed, there has been much interest in the synthesis of enantiomerically pure derivatives of 2-
substituted 1,3-propanediols, as they can be used in the preparation of a wide range of natural products
and biologically active compounds, such as muscone, microbial growth factors and renin inhibitors.16–20
Thus, the PCL-based HPLC column was used to prepare both enantiomers of 3-acetoxy-2-benzyl-propan-
1-ol. Compound (R)-3a was obtained by esterification of 2-benzyl-1,3-propan-diol, 1a, using vinylacetate
as the acyl donor, while (S)-3a was obtained by hydrolysis of 1,3-diacetoxy-2-benzyl-propane, 2a, easily
prepared by acetylation of the corresponding diols 1 (Scheme 1 and Table 1). Both reactions occurred
in bulk solution, with high conversion, and essentially enantiomerically pure fractions were obtained, in
agreement with the literature data.18,19
The enzyme-based HPLC column furnished comparable results for the hydrolysis and esterification
reactions (entries 1–4), but over longer reaction times. Thus, the immobilized enzyme maintains its
activity either using aqueous medium or vinylacetate as the mobile phase. An HPLC method was
developed which allowed the simultaneous determination of the conversion of the substrate and of the
enantiomeric excesses (ees) of the product by using a chiral stationary phase. As an example, the analysis
of 3a, obtained by hydrolysis of 2a, is reported in Fig. 1. The target product, 3a, was obtained in an
enantiomerically pure form, and the starting compound 2a and the obtained diol 1a are determined as
well. The absolute configuration of the prevailing enantiomer was assigned by measuring the optical
rotation at 589 nm and comparing its sign with that reported for (R)-3a.18
To enhance characterization of the performances of the immobilized PFL, two other substrates were
investigated in the esterification and hydrolysis reactions both in bulk solution and on the column
(Scheme 1). Esterification of 1b in bulk solution gave (R)-3b with 100% conversion and 24% ee after 4 h;
on the contrary, the conversion was much slower on the column, at only 7% after 48 h. Compound (R)-3b
was obtained by hydrolysis of 2b in bulk solution with very low yield and enantioselectivity; again the
reaction on the column was very slow (3% conversion after 48 h) so we decided not to investigate further
the 2-phenyl derivatives. On the contrary, very high ees (98%) were obtained in the esterification of 1c in

C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
4457
Table 1
Lipase catalyzed enantioselective esterification and hydrolysis
bulk solution, after the substrate had been completely converted and the formation of diacetate 2c from
3c had started. This suggested a very high enantioselectivity in the esterification of the second hydroxyl
group of 3c, in agreement with literature data.22 This behaviour was also observed when the reaction
was performed on the PCL-based column. The reaction was monitored up to 85% conversion, because
of the much longer time needed on the column, and a significant increase in the ee (from 16 to 36%)
Scheme 1.

4458
C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
Fig. 1. HPLC analysis of (a) racemic monoacetate 3a; (b) hydrolysis reaction mixture containing 3a (Chiracel OD, hex-
ane:2-propanol 95:5, flow 1 mL/min, λ 212 nm)
was observed at higher conversion. The chemical yield was comparable at the same conversion values
for both the reactions in bulk solution and on the column (entries 5 and 6).
High enantioselectivity was obtained at about 40% conversion in the hydrolysis of 2c, both in bulk
solution (after 0.9 h) and on the column (after 45 h) (entries 7 and 8). A decrease of the enantioselectivity
was observed at higher conversion of the substrate: for example, at about 60% conversion, the ee of (R)-
3c was 92% in bulk solution (after 2 h) and 72% on the column (after 96 h). A significant difference in

C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
4459
the times needed to perform the reaction in bulk solution or on the column has also been observed with
2-methyl derivatives.
The conversion of the substrate (1c or 2c) was followed by GC analysis, while the enantiomeric excess
of the product was determined by HPLC analysis of the benzoate derivative 4, obtained by reaction of
3c with benzoyl chloride (Scheme 2). This allowed easier monitoring and a reliable determination of the
enantiomeric composition.
Scheme 2.
As an example, the HPLC analysis of a sample of the benzoate derivative 4 is reported in Fig. 2.
The absolute configuration of the prevailing enantiomer has been assigned by measuring the optical
rotation and comparing its sign with that reported for (R)-3c; the elution order of the derivative 4 on
Chiralcel OD has been determined by HPLC analysis of a sample of known absolute configuration.‡
3. Conclusion
The PCL-based HPLC column has been proved to be efficient for the enantioselective synthesis of 2-
substituted 1,3-propanediol monoacetates. The enantioselectivity of the enzymatic processes on column
and in bulk solution are comparable, the conversion being generally slower for the immobilized enzyme.
The PCL-based column can be used in organic and aqueous media and it is stable for months after
continuous use. This makes the column useful for preparative purposes.
4. Experimental
4.1. General
1H NMR (200 MHz) spectra were recorded in CDCl3 with a Varian Gemini 200 spectrometer using
tetramethylsilane as internal standard. Optical rotations were measured on a Jasco DIP 360 automatic
polarimeter. Analytical TLC was performed on 0.2 mm silica gel plate Merck 60 F-254. GC (FI detector)
analysis was performed on a SGE BP1 column (25 m×0.53 mm i.d., 1.0 µm film). The enantiomeric
excesses were determined by HPLC analysis of the reaction mixture on an HPLC system consisting
of a Jasco PU-980 pump and a Jasco MD 910 multiwavelength detector (Jasco, Tokyo, Japan). The
chromatographic retention of the solutes were monitored by measuring the absorption at 212 nm (3a, 3b).
The enzymatic reactions were carried out by eluting the reaction mixture on an HPLC system consisting
of a Jasco 880-PU pump and a Jasco 875-UV–vis detector (Jasco, Tokyo, Japan).
4.2. Starting materials, reagents and solvents
Lipase PS from Amano was kindly provided by Prof. P. Crotti, University of Pisa, Italy.
‡
A sample of compound (S)-3c with 98% ee was derivatized with benzoyl chloride and analyzed by optical rotation, obtaining
a value of [α]D25=−0.8 (c 1, CHCl3) to be compared with the value22 of [α]D25=−0.8 (c 0.6, CHCl3) for the enantiomerically
pure (R) compound.

4460
C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
Fig. 2. HPLC analysis of (a) racemic benzoate 4; (b) benzoate 4 obtained by derivatization of a hydrolysis reaction mixture
containing 3c (Chiracel OD, hexane:2-propanol 99:1, flow 0.5 mL/min, λ 230 nm)
Acetic anhydride, benzoyl chloride, diethyl ether, methylene chloride, pyridine, tetrahydrofuran,
triethylamine were refluxed and distilled under an inert atmosphere. 2-Methyl-1,3-propanediol 1c, 2-
benzyldiethyl malonate, 2-phenyldiethyl malonate and vinyl acetate were commercially available. Diols
1a–b were prepared by LiAlH4 reduction of 2-benzyldiethyl malonate and 2-phenyldiethyl malonate
1
following the procedure described in Ref. 20. The obtained products were characterized by H NMR.
Diacetates 2a–c were obtained by reaction of diols 1a–c with 2.5 equiv. of acetic anhydride and 2.5
equiv. of pyridine. After stirring overnight, the mixture was diluted with toluene and evaporated under
reduced pressure to give the products in quantitative yields, which were characterized by 1H NMR.
Racemic monoacetate 3a–c were prepared as described in Ref. 20. The products were characterized
by 1H NMR and HPLC on a chiral column.

C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
4461
4.3. Monoacetate 3a
1H NMR (200 MHz, CDCl3, δ ppm): 7.36–7.10 (m, 5H); 4.22–3.99 (m, 2H); 3.66–3.43 (m, 2H);
2.93–2.58 (m+br s, 3H); 2.22–1.99 (s+m, 4H). HPLC: Chiralcel OD, hexane:2-propanol 95:5, flow rate
1mL/min.
4.4. Monoacetate 3b
1H NMR (200 MHz, CDCl3, δ ppm): 7.40–7.19 (m, 5H); 4.37 (d, 2H); 3.83 (d, 2H); 3.23–3.05 (q,
4H); 2.16 (br s, 1H); 2.04 (s, 1H). HPLC: Chiralcel OB, hexane:2-propanol 85:15, flow rate 0.8 mL/min.
4.5. Monoacetate 3c
1H NMR (200 MHz, CDCl3, δ ppm): 4.07 (d, 2H); 3.52 (br d, 2H); 2.58 (br s, 1H); 2.16–1.89 (m+s,
4H); 0.95 (d, 3H).
4.6. Synthesis of the benzoate derivative 4
To a solution of 0.1 mmol of monoacetate 3c in 2 mL of anhydrous CH2Cl2, 1.2 equiv. of Et3N and 1.2
equiv. of benzoyl chloride were added at 0°C. The reaction was followed by GC. Upon disappearence of
the substrate, the mixture was worked up with 10% HCl, 10% NaHCO3 and saturated NaCl. The organic
phase was dried on Na2SO4 and evaporated under reduced pressure. The residue was characterized by
GC and HPLC analysis and by measuring the optical rotation.22 1H NMR (200 MHz, CDCl3, δ ppm):
8.22–8.0 (m, 2H); 7.64–7.40 (m, 3H); 4.28 (d, 2H); 4.11 (d, 2H); 2.48–2.11 (m, 1H); 2.05 (s, 3H); 1.09
(d, 3H). HPLC: Chiralcel OD, hexane:2-propanol 99:1, flow rate 0.5 mL/min.
4.7. Synthesis of the lipase-based HPLC stationary phase
The silica gel (Kromasil 200 Å, 5µm, Eka Nobel, Sweden) was heated for 16 h at 180°C under vacuum.
Distilled (3-glycidoxypropyl)trimethoxysilane (1.9 g) in 5 mL of dry toluene was added dropwise to 5
g of activated silica in 60 mL of dry toluene. The mixture was refluxed for 12 h. After filtration the
product was washed twice with dry toluene and acetone. The derivatized silica was then dried for 15 h
at 80°C under vacuum. Epoxysilica (1 g) was slurry packed into a 50×4.6 mm i.d. HPLC column under
conventional conditions, using acetone as pumping solvent. The column was washed with acetone and
dried at 80°C under helium. After equilibration of the column with the coupling solution, i.e. 0.05 M
KH2PO4, pH=7.5, containing 2 M ammonium sulfate, the coupling solution (35 mL) containing 100
mg of PCL was applied to the column at 0.5 mL/min. The column was back flushed every 30 minutes
until the solution was exhausted. The column was disconnected, plugged and allowed to stand in the
refrigerator all night. The column was then washed with 100 mL of 0.05 M KH2PO4, pH=6, and the
remaining epoxide groups blocked with a solution (30 mL) of glycine 1 M in 0.05 M KH2PO4, pH=7,
and finally equilibrated with a solution (30 mL) of 0.05 M KH2PO4, pH=6.
4.8. General procedure for enzymatic hydrolysis
Procedure A: Diacetate 2a–c (0.16 mmol) was dissolved in 0.5 mL of n-propanol and added to a
suspension of 100 mg of lipase in 29.5 mL of 100 mM phosphate buffer (pH 7). The mixture was
magnetically stirred at room temperature and the reaction course was followed by HPLC analysis for
diacetates 2a–b and by GC analysis for diacetate 2c. The reaction was interrupted by filtration of the

4462
C. Bertucci et al. / Tetrahedron: Asymmetry 10 (1999) 4455–4462
enzyme: the filtrate was extracted with Et2O, dried on Na2SO4 and evaporated under reduced pressure.
The residue was characterized by GC or HPLC analysis.
Procedure B: The reaction mixture was prepared as in procedure A, except lipase, and it was
continuously eluted through the lipase column. The reaction was followed as indicated above.
Upon completion of the reaction, the column was flushed with clean buffer and re-used.
4.9. General procedure for enzymatic esterification
Procedure A: To a solution of 2.5 mmol of 1a–c in 8 mL of vinyl acetate, 100 mg of lipase were added.
The suspension was magnetically stirred at room temperature and the reaction course was followed by
HPLC analysis for 2a–b and by GC analysis for 1c. The reaction was interrupted by filtration of the
enzyme and the filtrate was evaporated under reduced pressure. The residue was characterized by GC or
HPLC analysis.
Procedure B: The reaction mixture was prepared as in procedure A, except lipase, and it was
continuously eluted through the lipase column. The reaction was followed as indicated above.
Upon completion of the reaction, the column was flushed with clean vinyl acetate and re-
used.
References
1. Schoffers, E.; Golebiowski, A.; Johnson, C. R. Tetrahedron 1996, 52, 3769.
2. Azerad, R. Bull. Soc. Chim. Fr. 1995, 132, 17.
3. Roberts, S. M. J. Chem. Soc., Perkin Trans. 1 1998, 157.
4. Xie, Z. Tetrahedron: Asymmetry 1991, 2, 733.
5. Boland, W.; Frößl, C; Lorenz, M. Synthesis 1991, 1049.
6. Bovara, R.; Carrea, G.; Ferrara, L.; Riva, S. Tetrahedron: Asymmetry 1991, 2, 931.
7. Bommarius, A. S.; Schwarm, M.; Drauz, K. Chemistry Today, 1996, 14, 61–64.
8. Bommarius, A. S.; Schwarm, M.; Drauz, K.; Groeger, U.; Wandrey, C. In Chirality and Industry; Collins, A. N.; Sheldrake,
G. N.; Crosby, J., Eds.; Membrane Bioreactors for the Production of Enantiomerically Pure Amino Acids; John Wiley &
Sons: London 1992; pp. 371–397.
9. Reetz, M. T. Adv. Mat. 1997, 9, 943.
10. Zhang, X.-M.; Wainer, I. W. Tetrahedron Lett. 1993, 34, 4731.
11. Cowan, D. Trends Biotechnol. 1996, 14, 177.
12. Felix, G.; Liu, M. Biosciences 1986, 8, 2.
13. (a) Marle, I.; Karlsson, A.; Pettersson, C. J. Chromatogr. 1992, 604, 185. (b) Felix, G.; Descorps, V. Chromatographia
1999, 49, 595.
14. (a) Hollins, D.; Shorr, R. Beckman Instruments Inc., Bulettin 1985, 5933. (b) Descorps, V. PhD dissertation, University of
Bordeaux, France, 1996.
15. Chang, S. H.; Gooding, K. M.; Regnier, F. E. J. Chromatogr. 1976, 120, 321.
16. Ramos Tombo, G. M.; Schar, H. P.; Fernandez, X.; Busquets, I.; Ghisalba, O. Tetrahedron Lett. 1986, 27, 5707.
17. Mori, K.; Chiba, N. Liebigs Ann. Chem. 1989, 957.
18. Tsuji, K.; Terao,Y.; Achiwa, K. Tetrahedron Lett. 1989, 30, 6189.
19. Atsuumi, S.; Nakano, M.; Koike, Y.; Tanaka, S.; Ohkubo, M.; Yonezawa, T.; Funabashi, H.; Hashimoto, J.; Morishima, H.
Tetrahedron Lett. 1990, 31, 1601.
20. Guanti, G.; Narisano, E.; Podgorski, T.; Thea, S.; Williams, A. Tetrahedron 1990, 46, 7081.
21. Santaniello, E.; Ferraboschi, P.; Grisenti, P. Tetrahedron Lett. 1990, 31, 56.
22. Grisenti, P.; Ferraboschi, P.; Manzocchi, A.; Santaniello, E. Tetrahedron 1992, 48, 3827.