Poly(3-hydroxyoctanoate) depolymerase from Pseudomonas fluorescens GK13: Catalysis of ester-forming reactions in non-aqueous media

Article abstract of DOI:10.1016/j.molcatb.2012.01.011

Several industrial processes based on lipase catalysis have been established. However, since there are still a vast number of catalytic processes that lack a suitable enzyme, the discovery of new biocatalysts is required to fulfil this purpose. The potential of using the medium-chain-length (mcl)-PHA depolymerase from Pseudomonas fluorescens GK13 in anhydrous media to catalyze ester-forming reactions has been investigated and compared with that of Novozyme 435. The mcl-PHA depolymerase catalyzes the ring-opening polymerization of racemic β-butyrolactone (β-BL), l- and d-lactide (LLA, DLA) with high yield resulting in low molecular weight polymers. On the other hand, ε-caprolactone and pentadecalactone, which show high polymerizability using Novozyme 435 as catalyst, were not polymerized by mcl-PHA depolymerase. Besides, the activity of mcl-PHA depolymerase toward transesterification and esterification of ethyl-3-hydroxyoctanoate, lauric acid, (R,S)-β-BL, LLA and DLA has been studied.

Full text of DOI:10.1016/j.molcatb.2012.01.011

Journal of Molecular Catalysis B: Enzymatic 77 (2012) 81–86
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Poly(3-hydroxyoctanoate) depolymerase from Pseudomonas fluorescens GK13:
Catalysis of ester-forming reactions in non-aqueous media
Marta Santosa,1, Joana Gangoitia,1, María J. Llamaa, Juan L. Serraa, Helmut Keulb,∗, Martin Möllerb
a Enzyme and Cell Technology Group, Department of Biochemistry and Molecular Biology, Faculty of Science and Technology, University of the Basque Country (UPV/EHU), P.O. Box
644, 48080 Bilbao, Spain
b DWI an der RWTH Aachen e. V., and Institute of Technical and Macromolecular Chemistry, RWTH Aachen, Pauwelsstr. 8, D-52056 Aachen, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Several industrial processes based on lipase catalysis have been established. However, since there are
still a vast number of catalytic processes that lack a suitable enzyme, the discovery of new biocatalysts
is required to fulfil this purpose. The potential of using the medium-chain-length (mcl)-PHA depoly-
merase from Pseudomonas fluorescens GK13 in anhydrous media to catalyze ester-forming reactions
has been investigated and compared with that of Novozyme 435. The mcl-PHA depolymerase catalyzes
the ring-opening polymerization of racemic ␤-butyrolactone (␤-BL), l- and d-lactide (LLA, DLA) with
high yield resulting in low molecular weight polymers. On the other hand, ␧-caprolactone and pentade-
calactone, which show high polymerizability using Novozyme 435 as catalyst, were not polymerized by
mcl-PHA depolymerase. Besides, the activity of mcl-PHA depolymerase toward transesterification and
esterification of ethyl-3-hydroxyoctanoate, lauric acid, (R,S)-␤-BL, LLA and DLA has been studied.
© 2012 Elsevier B.V. All rights reserved.
Received 8 November 2011
Received in revised form 12 January 2012
Accepted 12 January 2012
Available online 21 January 2012
Keywords:
Pseudomonas fluorescens GK13
Extracellular mcl-PHA depolymerase
Enzymatic ring-opening polymerization
Novozyme 435
1. Introduction
fluorescens GK13 (PhaZGK13) [6,7]. This enzyme consists of a signal
peptide, an N-terminal substrate binding domain and a C-terminal
Polyhydroxyalkanoate (PHA) depolymerases are microbial
enzymes that degrade polyhydroxyalkanoates, biopolyesters that
are synthesized and accumulated intracellularly by a wide range of
bacteria [1]. PHAs are biopolyoxoesters composed of (R)-3-hydroxy
fatty acids which represent a complex class of storage polyesters
[2]. The study of the PHAs hydrolysis by PHA depolymerases has
attracted much attention not only due to their potential use as bio-
plastics or biomaterials [3], but also because their potential use
in the production of chiral (R)-3-hydroxyalkanoic acids [4]. PHA
depolymerases can be divided into intracellular and extracellular
depolymerases [5]. These enzymes are specific for either short-
chain-length (scl)-PHAs (3–5 carbon atoms, EC 3.1.1.75) or for
medium-chain-length (mcl)-PHAs (6–14 carbon atoms, EC 3.1.1.76)
[5]. Until now, more than 80 extracellular PHA depolymerases have
been purified and characterized from various microorganisms but
most of the isolated enzymes are specific to scl-PHA [1]. In general,
mcl-PHA depolymerases differ considerably from scl-PHA depoly-
merases in terms of primary sequence and polymer-binding [5].
Among the mcl-PHA depolymerases characterized, the proto-
type of extracellular mcl-PHA depolymerase is that of Pseudomonas
catalytic domain [6]. Although mcl-PHA depolymerases do not
display significant overall sequence homology to lipases or to
other esterases, these enzymes possess a catalytic triad con-
sisting of serine, histidine, and aspartate residues which have
been demonstrated to form the active site in all known serine
hydrolases. Moreover, the conserved Ser is a part of the lipase con-
sensus sequence motif Gly-X1-Ser-X2-Gly [1,6]. Apart from this,
PhaZGK13 is active for p-nitrophenylacyl esters, which are typi-
cal substrates for the determination of esterase activity. However,
PhaZGK13 does not show significant lipase activity being unable
to catalyze the hydrolysis of triolein and Tween 80 [8]. On the
contrary, some lipases are able to hydrolyze polyesters consist-
ing of an ␻-hydroxyalkanoic acid, such as poly(4-hydroxybutyrate)
and poly(6-hydroxyhexanoate) but not to hydrolyze poly-␤-
hydroxyalkanoates [9]. Apparently, lipases and PHA depolymerases
may follow the same reaction mechanism, but differing in the spa-
tial arrangement of their active sites.
The syntheses of aliphatic polyesters have been extensively
studied by both fermentation and chemical processes in the field
of biodegradable materials science [10]. Nevertheless, microbial
syntheses generally give random copolyesters with relatively high
production costs; and chemical modifications involve the use
of toxic chemicals or catalysts. Alternatively, extensive research
proved that a number of hydrolases can catalyze very efficiently
ester and amide bond formation in non-aqueous media allowing
∗
Corresponding author.
E-mail address: keul@dwi.rwth-aachen.de (H. Keul).
1
Both researchers share the position of the first author.
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2012.01.011

82
M. Santos et al. / Journal of Molecular Catalysis B: Enzymatic 77 (2012) 81–86
the synthesis of a number of functional aliphatic polyesters [11].
Enzymatic polymerization has provided novel synthetic strate-
gies for the production of different types of polymers affording
cleaner products under milder conditions than conventional chem-
ical catalysts [12]. Reactions leading to polymers include step
growth reaction like polycondensation, and chain growth reaction
like ring-opening polymerization [12]. Surprisingly, the majority
of enzymes studied for polymerization reactions are members of
the lipase family with Lipase B from Candida antarctica (CALB)
immobilized on an acrylic resin (Novozyme 435) as the dominant
enzyme [11]. This commercially available biocatalyst has proven to
be highly useful and versatile. Although numerous reactions have
been performed using lipases and related enzymes, it is still a chal-
lenge to identify the optimal biocatalyst and reaction conditions for
a specific reaction [13].
Size exclusion chromatography (SEC) was carried out using
chloroform as eluting solvent. A high pressure liquid chromatogra-
phy pump (Jasco PU-2080-Plus), a Jasco 2031-Plus RI detector and
four MZ-SD Plus gel columns (50 A, 100 A, 1000 A and 10,000 A)
were used in series at 30 ◦C with a flow rate of 1.0 mL/min.
The molecular weights and polydispersities were determined
using narrow distributed poly(methyl methacrylate) (PMMA)
standards.
˚
˚
˚
˚
The catalytic activity of PhaZGK13 was determined at 30 ◦C by
measuring for 5 min the turbidity decrease at 650 nm of P(3HO)
suspensions [18]. The standard assay mixture consisted of 500 ␮g
of P(3HO) in 200 mM Tris–HCl (pH 8.0) in a total volume of 1 mL.
After incubation at 30 ◦C for 10 min, the reaction was started by the
addition of the purified enzyme.
Previous work describing scl-PHA depolymerase activity in non-
aqueous media have been reported [14–17]. Kumar et al. [14]
reported that scl-PHA depolymerase from Pseudomonas lemoignei
has the potential to catalyze ester-forming reactions starting
with cyclic esters and carbonates. Specifically, they reported the
ring-opening of the lactones and propylation of their carboxyl
group with 1-propanol in organic solvents. On the other hand,
Suzuki et al. [15] provided the first report on polymerization of
␤-butyrolactone (␤-BL) into poly(3-hydroxybutyrate), P(3HB), by
scl-PHA depolymerase from Pseudomonas stutzeri. Lately, they sub-
jected unsubstituted lactones of different ring size to ring-opening
polymerization using an extracellular P(3HB) depolymerase from
Alcaligenes faecalis T1 and obtained best results when using small
ring-sized lactones as monomers [17]. However, the use of mcl-PHA
depolymerases to catalyze bond formation in non-aqueous media
has not yet been investigated. Hence, in this work, we prepared
and purified the mcl-PHA depolymerase from P. fluorescens GK13
and studied its activity and specificity to catalyze ester-bond form-
ing reactions in non-aqueous media in comparison to those of the
commercially available immobilized lipase Novozyme 435.
2.3. General procedure for transacylation reactions using
mcl-PHA depolymerase as the catalyst
The enzyme-catalyzed transacylation of lactides, ␤-BL, ethyl-
3-hydroxy-octanoate and lauric acid were performed as follows.
The monomer (100 mg) and the powdered enzyme (1 wt%) were
transferred to a dry flask. An equimolar amount of the alcohol
with respect to the acyl-compound was added. The glass flask
was then placed in an oil bath at 80 ◦C with agitation. Depend-
ing on the monomer, the reactions were performed in bulk or in
toluene. In the absence of enzyme no significant transacylation was
observed.
n-Propyl-3-hydroxy-octanoate (1): 1H NMR (400 MHz, CDCl3)
ı = 0.90 (t, J = 8.0 Hz, CH3CH2), 1.6 (m, J = 7 Hz, CH2CH2CH3), 4.02 (t,
J = 6.8 Hz, OCH2CH2), 2.46; 2.35 (ddd, J = 7 Hz, HOCHCH2CO2), 3.94
(m, J = 7 Hz, CH2CH(OH)CH2), 1.44 (q, J = 7.1 Hz, CH2CH2CH2OH),
1.28 (m, J = 7.1 Hz, CH2CH2CH2CH2), 1.31 (m, J = 7.1 Hz, CH3CH2),
0,88 (t, J = 7.6 Hz, CH3CH2). 13C NMR (101 MHz, CDCl3) ı = 173.1,
67.9, 66.14, 41.6, 36.46, 31.71, 25.12, 21.87, 21.8, 13.95, 10.26.
2-Propyl-3-hydroxy-octanoate (2): 1H NMR (400 MHz, CDCl3)
ı = 1.30 (d, J = 6.8 Hz, OCH(CH3)2), 5.06 (m, J = 6.8 Hz, OCH(CH3)2,
2.46; 2.35 (ddd, J = 7 Hz, HOCHCH2CO2), 3.94 (m, J = 7 Hz,
CH2CH(OH)CH2), 1.44 (q, J = 7.1 Hz, CH2CH2CH2OH), 1.28 (m,
J = 7.1 Hz, CH2CH2CH2CH2), 1.31 (m, J = 7.1 Hz, CH3CH2), 0,88 (t,
J = 7.6 Hz, CH3CH2). 13C NMR (101 MHz, CDCl3) ı = 171.4, 68.9, 68.04,
41.6, 36.46, 31.76, 25.20, 22.58, 21.46, 14.18.
2. Materials and methods
2.1. Materials and reagents
␤-Propiolactone (␤-PL, 98% purity, Aldrich), ␤-butyrolactone
(␤-BL, 98% purity, Aldrich), pentadecanolide (PDL, 99.0% purity,
Fluka), ethyl-3-hydroxyoctanoate (Penta) and lauric acid (99.5%
purity, Acros Organics) were used without further purification. ␧-
Caprolactone (␧-CL, 99% purity, Acros Organics) was distilled over
calcium hydride in vacuo. d-Lactide (DLA, >99.5% purity, a gift from
PURAC) was purified by recrystallization in ethyl acetate and subli-
mation in vacuum at 80 ◦C. l-Lactide (LLA, Boehringer) was purified
by sublimation in vacuum at 80 ◦C.
n-Propyl laurate (3): 1H NMR (400 MHz, CDCl3) ı = 0.91 (t,
J = 8.0 Hz, CH3CH2), 1.26–1.29 (m, J = 7.1 Hz, CH2(CH2)8CH2), 1.31
(m, J = 7.5 Hz, CH3CH2CH2), 1.61 (m, J = 7.1 Hz, CH2CH2CH2COOH),
1.73 (m, J = 7.5 Hz, CH3CH2CH2), 2.3 (t, J = 7.1 Hz, CH2CH2COOH),
4.02 (t, J = 7.0 Hz, OCOCH2CH2). 13C NMR (CDCl3): ı = 173.9, 65.7,
34.4, 31.9, 29.6, 25.1, 22.8, 21.4, 14.1, 10.3.
Propyl-(3R/3S)-3-hydroxybutanoate (4): 1H NMR (400 MHz,
CDCl3) ı = 4.19 (q, J = 6.3 Hz, CH2CHOHCH3), 3.97 (t, J = 6.7 Hz,
CH2CH2OCO), 2.31 (ddd, J = 8.2 and 16.0 Hz, 4.1 and 16.0 Hz,
OCOCH2CHOHCH3), 1.63 (m, J = 6.7 Hz, CH3CH2CH2), 1.19 (d,
J = 6.3 Hz, CH2CHOHCH3), 0.91 (t, J = 6.7 Hz, CH3CH2). 13C NMR
(101 MHz, CDCl3) ı = 172.89, 66.14, 64.50, 44.16, 22.30, 21.87, 10.26.
n-Propyl lactide (5): 1H NMR (400 MHz, CDCl3): ı = 5.18 (q,
J = 7.1 Hz, OCH3CHCO), 4.33 (q, J = 6.9 Hz, OHCHCH3CO), 4.08 (t,
J = 6.8 Hz, OCH2CH2), 1.63 (m, CH2CH2CH3), 1.56 (d, J = 7.1 Hz,
OCH3CHO), 1.48 (d, J = 6.9 Hz, CH3CHOH), 0.92 (t, J = 6.8 Hz, CH2CH3).
13C NMR (101 MHz, CDCl3): ı = 175.06, 170.09, 69.35, 66.94, 66.6,
21.82, 21.35, 16.8, 10.15.
All esterification reactions were carried out under a N2 atmo-
sphere. Nitrogen (Linde, 5.0) was passed over molecular sieves
˚
(4 A) and finely distributed potassium on aluminium oxide. Toluene
was distilled over sodium. Novozyme 435 (10,000 U/g, Novo
Nordisk) was dried at room temperature for 24 h in vacuum
and stored under N2. PhaZGK13 was purified from the culture
broth of cells grown in mineral medium supplied with poly(3-
hydroxyoctanoate) [P(3HO)] as described by Gangoiti et al. [18],
and used after lyophilisation.
n-Propyl lactate (6): 1H NMR (400 MHz, CDCl3): ı = 5.18 (q,
J = 7.1 Hz, O(CH3)CHCO), 4.23 (q, J = 6.9 Hz, HOCHCH3CO), 4.08 (t,
J = 6.8 Hz, OCH2CH2), 1.63 (m, CH2CH2CH3), 1.56 (d, J = 7.1 Hz,
OCH3CHO), 1.39 (d, J = 6.9 Hz, CH3CHOH), 0.92 (t, J = 6.8 Hz, CH2CH3).
13C NMR (101 MHz, CDCl3): ı = 175.69, 67.0, 66.6, 21.82, 21.35,
10.15.
2.2. Measurements
1H NMR and 13C NMR spectra were recorded on a Bruker
DPX-400 FT-NMR spectrometer at 400 MHz and 101 MHz, using
deuterated chloroform (CDCl3) as the solvent. Chemical shifts were
reported relative to tetramethylsilane signals.

M. Santos et al. / Journal of Molecular Catalysis B: Enzymatic 77 (2012) 81–86
83
PhaZGK13
2.4. Typical procedure for mcl-PHA depolymerase catalyzed
ring-opening-polymerizations
OH
O
OH
O
Novozyme
O
O
O
O
4
4
1
n-propanol
The general procedure used for the enzymic ring-opening poly-
merization of lactones and l- and d-lactides was as follows. A
mixture of monomer (100 mg) and the enzyme (1 wt%) was stirred
under a N2 atmosphere in a capped vial placed in an oil bath at 80 ◦C.
Unless otherwise specified all reactions were performed for 24 h.
After the reaction, the mixture was dissolved in deuterated chloro-
form (1 mL), and the insoluble enzyme was removed by filtration.
The chloroform was then evaporated in vacuo to quantitatively
obtain the product mixture.
OH
O
PhaZGK13
OH
O
4
Novozyme
2-propanol
4
2
PhaZGK13
O
O
O
OH
Novozyme
8
8
3
n-propanol
The molecular weight and the molecular weight distribution
of the polymer fraction were determined by SEC. Depending on
the monomers used the reactions were performed in bulk or in
toluene.
PhaZGK13
O
OH
O
n-propanol
O
O
n
O
Novozyme
O
O
OH
H3C
Monomer conversion was determined by 1H NMR spectroscopy,
comparing the integration intensities of protons of the monomers
with those of the repeating unit in the polymer. For ␤-PL, ␧-CL
and PDL the position of the methylene protons adjacent to the
oxygen atom in the monomer were ı = 4.2 (C-3 of ␤-PL), 4.23 (C-
6 of ␧-CL) and 4.15 ppm (C-15 of PDL) and in the polymer were
ı = 4.3 [C-3 of P(3HP)], 4.06 [C-6 of P(6HH)] and 4.05 ppm [C-15
of P(PDL)]. In the case of (R,S)-␤-BL the monomer conversion was
determined by comparison of the integration intensities for the
peak at ı = 1.58 ppm corresponding to the methyl protons of the
monomer with the corresponding methyl protons of the repeating
unit of P(3HB) at ı = 1.28 ppm.
n-propanol
4
5
O
Novozyme/
PhaZGK13
O
O
O
O
OH
O
n-propanol
O
O
O
O
PhaZGK13
O
O
OH
OH
O
O
n-propanol
O
O
O
O
5
5
O
O
Novozyme
+
OH
n-propanol
O
6
Conversion of lactides to poly(lactic acid) (PLA) was deter-
mined by comparison of the integration intensities of the peak
at ı = 4.76 ppm corresponding to the methyne protons of the
monomeric lactide with the peak at ı = 5.11 ppm corresponding to
PLA.
Scheme 1. Transacylation reactions.
In all cases, control reactions without enzyme were performed.
the studied conditions, no monomer conversion was achieved
after 48 h. Similarly, PhaZGK13 did not promote the transacylation
of ethyl-3-hydroxy-octanoate with n-propanol and 2-propanol
(Scheme 1) after 2 days of reaction, while 41% and 10% of conver-
sion was obtained, respectively, when the reaction was conducted
by the Novozyme 435.
Additionally, the potential of using PhaZGK13 to catalyze the
esterification of lauric acid was studied. As expected, after 72 h high
conversions (96%) were obtained using Novozyme 435, whereas
3. Results and discussion
The potential of mcl-PHA depolymerase from P. fluorescens GK13
as a new catalyst for esterification reactions was investigated.
In order to evaluate the potential use of this new biocatalyst, a
direct comparison with the reactivity of commercially available
Novozyme 435 for different ester forming reactions was performed.
All reactions were carried out simultaneously under the same con-
ditions using both catalysts. However, the fact that PhaZGK13 was
used as free enzyme will ultimately influence comparison between
these two protein catalysts.
low conversions (5%) were observed in the case of PhaZGK13
.
Previous studies described that the scl-PHA depolymerase from
P. lemoignei catalyzed alcoholysis of different cyclic esters in
organic solvents [14] obtaining best results with l-LA (SS), meso-
LA (RS), d-LA (RR) and racemic ␤-BL. Lactides and ␤-BL have
secondary hydroxyl groups attached to the chiral centre that
participate in the formation of ester bonds. Furthermore, the sub-
stituents attached to the chiral centres of both lactide and ␤
-BL are methyl groups. In contrast to ␤-BL, which has the hydroxyl
group in ␤ position to the carbonyl group, lactides have the
hydroxyl group in ␣-position. Motivated by the above, reaction of
n-propanol with racemic ␤-BL, and d- and l-lactides was studied
(Scheme 1). As shown in Table 1, when ␤-BL was used as acyl donor,
19% of n-propanol was transacylated by PhaZGK13 in toluene after
72 h as revealed by the presence of the signal (ı = 3.97 ppm) corre-
sponding to the acylated alcohol in the 1H NMR spectra. However,
63% of ␤-BL conversion was determined. Besides, SEC analysis con-
firmed the presence of a polymer with an estimated Mn of 770 g/mol
indicating that ␤-BL polymerization also occurred. When this reac-
tion was conducted in bulk instead of toluene, 6% of transacylated
product was estimated by 1H NMR and 99% of ␤-BL conversion was
observed. In this case a polymer of a Mn of 1800 g/mol was obtained.
3.1. Preparation of PhaZGK13
The PhaZGK13 was purified as described by Gangoiti et al. [18].
SDS-PAGE analysis showed a major double band at about 25 kDa,
which corresponded to the enzyme (data not shown). The enzyme
solution in phosphate buffer (5 mM, pH 8.0) was then subjected
to lyophilisation. The enzyme retained 87% of its initial activity
after lyophilisation. Routinely, preparations of PhaZGK13 with a spe-
cific activity of approximately 52.7 U/mg for P(3HO) were used as
catalyst in ester-bond forming reactions.
3.2. Transacylation reactions
The natural substrate hydrolyzed by mcl-PHA depolymerases in
aqueous media are enantiopure (R)-P(3HO) or structurally related
polyhydroxyalkanoates. Therefore, ethyl-3-hydroxy-octanoate
was selected as a substrate for the transesterification reaction
to determine the activity of PhaZGK13 for polyester synthesis
via condensation in a non-aqueous medium. However, under

84
M. Santos et al. / Journal of Molecular Catalysis B: Enzymatic 77 (2012) 81–86
Table 1
Enzyme catalyzed esterification of n-propanol using different acyl-donors.a
b
Monomer
Enzyme
Solvent
Time (h)
Conversion (%)
Product (%)
Mn
1800
770
–
␤-BL
PhaZGK13
PhaZGK13
Novozyme
In bulk
Toluene (1.8 mL)
Toluene (1.8 mL)
24
72
72
99
63
54
Polymer (93)
Polymer (44)
4 (54)
LLA
PhaZGK13
Toluene (0.6 mL)
24
99
Oligomer (11)
5 (88)
ndc
–
Novozyme
PhaZGK13
Toluene (0.6 mL)
Toluene (0.6 mL)
24
24
78
99
5 (78)
–
DLA
Oligomer (19)
5 (80)
ndc
–
Novozyme
Toluene (0.6 mL)
5 (74)
–
24
99
6 (25)
–
Numbers 4, 5 and 6 correspond to the products represented in Scheme 1.
a
Reaction was performed at 80 ◦C (mol ratio of alcohol/acyl donor = 1:1) using 1% (w/w) of PhaZGK13 or Novozyme 435.
Molecular weight (Mn) determined by SEC using narrow distributed poly(methyl methacrylate) standards and chloroform as eluent.
nd: not determined.
b
c
As a consequence of ␤-BL polymerization, the number of free ter-
3.3. Ring-opening polymerization reactions
minal carboxyl groups able to react with n-propanol was limited.
Based on the 1H NMR results we supposed we have obtained a
functionalized polymer. We assume that n-propanol acts as initia-
tor and once the ␤-BL ring is opened the rate of polymerization
is higher than the rate of initiation and consequently oligomers
are the preferred product. In contrast, the absence of P(3HB) in the
product mixture when Novozyme 435 catalyzed the transacylation
of n-propanol indicated that the immobilized CALB promoted only
the transacylation of the n-propanol to propyl-3-hydroxybutanoate
reaching a 54% of conversion (Table 1). In the case of Novozyme
435, it is assumed that the rate of initiation is higher and propaga-
tion by the secondary alcohol is inefficient. Similar to immobilized
CALB, Kumar et al. [14] reported that scl-PHA depolymerase of P.
lemoignei catalyzed the n-propylation of racemic ␤-BL leading to
propyl-3-hydroxybutanoate with 50% of conversion.
Based on our previous results, PhaZGK13 might be a potential
catalyst for the ring opening polymerization of the four-membered
␤-methyl substituted lactone, ␤-BL (Scheme 2).
As shown in Table 2, PhaZGK13 catalyzed the ring opening poly-
merization of racemic ␤-BL with a monomer conversion of 98%
to yield P(3HB) with a Mn of 1500 after 24 h. Similarly, the scl-
PHA depolymerase of P. stutzeri catalyzed the polymerization of
␤-BL with 95% of monomer conversion obtaining a P(3HB) poly-
mer with a Mn ∼1250 g/mol. However, in this study the reaction
was performed for 3 days at 80 ◦C in the presence of 5 wt% of
enzyme [15]. Under the same conditions (Table 2), ␤-BL was hardly
polymerized by Novozyme 435 and only 8% of conversion was
observed. The lipase-catalyzed ring-opening polymerization of this
monomer was first reported by Nobes et al. [19]. P(3HB) having
On the other hand, PhaZGK13 catalyzed the n-propylation of LLA
and DLA to n-propyl lactide. After 24 h of reaction using PhaZGK13,
more than 80% of n-propanol was transacylated (Table 1). However,
99% of lactide conversion was observed, indifferently of the stereo-
chemistry of the lactide used. This fact and the presence of a weak
but detectable signal at ı = 4.97 ppm correspondent to unreacted
␣ methyne proton of the carboxyl end group suggested that lac-
tide oligomers have been formed as a side product (Table 1). When
the molar ratio monomer/alcohol was increased from 1:1 to 1:3,
100% of lactide was converted to propyl lactide. On the other hand,
LLA ring-opening and esterification by Novozyme 435 yielded 78%
of propyl lactide as the only product. In contrast, in the case of
DLA esterification catalyzed by Novozyme 435 yielded beside to n-
propyl d-lactide (74%), n-propyl d-lactate (25%), as revealed by the
presence of signal at ı = 4.23 ppm (data not shown). Similar results
were obtained with the scl-PHA depolymerase from P. lemoignei
[14]. However, based on our results, PhaZGK13 is unable to further
cleave the internal ester bond of the n-propyl-lactide to form n-
propyl-lactate indifferently of the stereochemistry of the monomer.
It seems that PhaZGK13 catalyzes ester-bond-formation showing
different reactivity than the one determined for Novozyme 435.
In fact, in contrast to the immobilized CALB, PhaZGK13 is unable
to catalyze the transacylation of ethyl-3-hydroxy-octanoate and to
propylate lauric acid. Both of these substrates present an alkyl moi-
ety that might impede the recognition of their functional group
by the active centre. However, PhaZGK13 catalyzed transacylation
of n-propanol when using lactides as acyl donors producing n-
propyl lactide as the major product with high yield. On the contrary,
PhaZGK13 catalyzed ring opening polymerization of ␤-BL in the
presence of a functional alcohol and not transacylation leading to
propyl-3-hydroxybutanoate, while in the case of Novozyme 435 no
P(3HB) polymer was synthesized.
Scheme 2. Ring opening polymerization reactions. PhaZGK13 catalyzed reactions 1,
2, 3 and 4. Novozyme 435 catalyzed reactions 1, 3, 6 and 7.

M. Santos et al. / Journal of Molecular Catalysis B: Enzymatic 77 (2012) 81–86
85
Table 2
Enzyme catalyzed ring-opening polymerization of cyclic esters: monomer, conversion and data from SEC analysis.a
b
b
Monomer
Enzyme
Solvent
In bulk
Time (h)
24
Conversion (%)
Mn
Mw/Mn
␤-BL
PhaZGK13
Novozyme 435
98
8
1500
760
1.52
1.05
LLA
PhaZGK13
Novozyme 435
Toluene (0.6 mL)
Toluene (0.6 mL)
In bulk
24
24
6
98
0
900
–
1.03
–
DLA
␤-PL
␥-BL
␧-CL
PhaZGK13
Novozyme 435
93.4
14.3
900
ndc
1.01
ndc
PhaZGK13
None (Control)
100
3
9200
960
2
1.15
PhaZGK13
Novozyme 435
In bulk
24
48
48
0
0
–
–
–
–
PhaZGK13
Novozyme 435
In bulk
0
98.1
–
6500
–
1.9
PDL
PhaZGK13
In bulk
0
–
–
a
Reaction was performed with 1 wt% of enzyme to monomer and at 80 ◦C in all cases except of ␤-PL reaction which was carried out at 60 ◦C.
Molecular weight (Mn) and molecular weight distribution (Mw/Mn) determined by SEC using narrow distributed poly(methyl methacrylate) standards and chloroform
b
as eluent.
c
nd: not determined.
weight average molecular weights ranging from 260 to 1050 g/mol
were obtained after several weeks of polymerization. Recent work
reported by Gorke et al. [20] using Novozyme 435 as catalyst in ionic
liquids gave oligomers of ␤-BL with a degree of polymerization
of 5. Consequently, based on these results the mcl-PHA depoly-
merase of P. fluorescens GK13 might be a promising catalyst for
␤-BL polymerization.
It has been reported that, P(3HB) obtained from the lipase
catalyzed ring-opening polymerization of (R,S)-␤-BL presented
three different kinds of structural isomers, cyclic, hydroxyl-
and crotonate-terminated P(3HB) [21]. The 1H NMR spectrum
of P(3HB) revealed small peaks corresponding to the hydrox-
ymethyne (ı = 4.2 ppm) and crotonate (ı = 1.86, 5.75 and 6.9 ppm)
terminal groups of P(3HB). However, by 1H NMR spectroscopy we
were unable to confirm the presence of cyclic oligomers.
application of CALB in the ring-opening polymerization of lactides
have been identified so far. First, CALB is known to be a slow catalyst
when ester substrates have a substituent at ˛-position next to the
carbonyl group [25] and, secondly, CALB is highly enantioselective
for R-secondary alcohols [26].
In contrast, as occurred in lipase PS [23], PhaZGK13 can poly-
merize LLA and DLA, indicating that both enzymes are not
enantiospecific for lactides. This is the first report on in vitro
ring-opening polymerization of l- and d-LA into PLA, by a PHA
depolymerase. Taking into account that PhaZGK13 catalyzed not
only the ring-opening polymerization of ␤-BL, but also of d- and
l-LA monomers, the possibility of copolymerizing these substrates
might be explored. Interestingly, PhaZGK13 does not catalyze the
hydrolysis of P(3HB) and PLA polymers in aqueous media [9]. There-
fore, the enzyme might show different specificity for the hydrolysis
and formation of ester bonds of these substrates.
Additionally, studies were performed to assess the PhaZGK13
activity toward ring-opening polymerization of lactides of different
configuration (Scheme 2). Lactide, a six-membered ring monomer,
is highly reactive in chemical ring-opening polymerization and
results in biocompatible and biodegradable polymers that are used
in a variety of biomedical applications. This makes this monomer
interesting for exploration in PhaZGK13-catalyzed ring-opening
polymerization. As shown in Table 2, PhaZGK13 showed excellent
polymerization activities toward LLA and DLA. In fact, after 24 h
the conversion obtained with LLA and DLA is very high in both
cases. However, the molecular weight of the synthesized polymers
was low (Mn,SEC ∼900). On the other hand, under the same con-
ditions Novozyme 435 did not promote the polymerization of the
monomer in (S)-configuration while 14% conversion was obtained
with DLA. Although enzymes have been successfully used for the
synthesis of many classes of polyesters, yet no enzyme has shown
any high activity for the in vitro polymerization of d,l-lactide. Sev-
eral hydrolases, such as Burkhordelia cepacia lipase (lipase PS),
Humicola insolens cutinase, Candida cylindracea lipase, Proteinase K
and lipase CALB have been evaluated for the ring-opening polymer-
ization of lactides. However, long reaction time, high background
activity and high enzyme concentration have characterized these
studies [22]. By the moment, best results were reported by Mat-
sumura et al. [23] using 3 wt% of lipase PS relative to d,l-lactide.
After 7 days of reaction at 130 ◦C poly(d,l-lactide) with a Mn of
115,000 was obtained. In recent years, Hans et al. [24] reported that
the polymerization of DLA catalyzed by Novozyme 435 resulted
in 80% of conversion at 50 ◦C after 2 days, while no polymer
was obtained when using LLA. Two problems that limited the
In order to understand in detail the interaction between the
enzyme and lactones, the activity and specificity of PhaZGK13
toward ring-opening polymerization of unsubstituted lactones
of various ring-sizes was evaluated and compared with that of
Novozyme 435 under the same conditions. In this study, ␤-PL, ␥-
BL, ␧-CL, and PDL were used as monomers and subjected to the
ring-opening polymerization reaction (Scheme 2). It was found that
lactones showed various characteristic reactivities (Table 2).
PhaZGK13 catalyzed ␤-PL ring-opening polymerization after 6 h
at 60 ◦C with a 100% of monomer conversion. The Mn of the resulting
P(3HP) was 9200 g/mol. Similar results were obtained when 1 wt%
of the wild-type scl-PHA depolymerase of A. faecalis T1 was used as
catalyst. In this case, after 2 days at 60 ◦C a P(3HP) polymer with a
Mn of 8200 g/mol was obtained [17]. On the other hand, as expected,
both PhaZGK13 and Novozyme 435 were not able to polymerize
␥-BL because of the thermodynamic stability of the 5-membered
ring lactone. In fact, there are few reports on the enzyme-catalyzed
ring-opening polymerization of the ␥-BL, except for Nobes et al.
[19] and Dong et al. [27]. However, in both cases enzymatic poly-
merization by using ␥-butyrolactone as monomer only lead to low
molecular weight products and monomer conversions. Indeed, best
results were obtained by Nobes et al. [19] after 18 days of reaction
yielding P(4HB) in up to 42% yield with a Mw of ∼900. Moreover,
␧-CL (Table 2) and PDL [28], which show excellent polymerization
activities by Novozyme 435, were not polymerized by PhaZGK13
(Table 2).
In conclusion, PhaZGK13 is able to catalyze ester-bond-formation
using not only primary, but also secondary alcohols in ␣ and ␤

86
M. Santos et al. / Journal of Molecular Catalysis B: Enzymatic 77 (2012) 81–86
positions. In fact, PhaZGK13 catalyzed the polymerization of ␤-PL,
␤-BL and LLA and DLA, with ring sizes of 4 and 6, respectively.
Interestingly, PhaZGK13 does not show enantioselectivity toward
secondary alcohols, while Novozyme 435 polymerizes only DLA
under the reaction conditions used. According to Uppenberg et al.
[29], the active site of CALB is composed of two channels; one
hosting the acyl- and the other hosting the alcohol-moiety of the
substrate, with the first channel being the more spacious one,
showing a high enantioselectivity toward secondary alcohols and a
broad substrate tolerance. Although, there is no information about
the quaternary structure of mcl-PHA depolymerases, based on our
results it can be suggested that PhaZGK13 may show a different
organization being less restrictive to alcohol moieties.
Furthermore, contrary to Novozyme 435, PhaZGK13 does not
show polymerization activity toward lactones with ring size higher
than 6. Similar results were reported for scl-PHA depolymerase
from A. faecalis T1 by Suzuki et al. [17] who based on a struc-
tural model proposed that these substrates seemed to be over size
limit for the molecular recognition of the narrow active site cleft of
scl-PHA depolymerases.
Education and Science (BIO2007-28707-E) and UPV/EHU
(GIU07/55 and GIU11/25). M.S. and J.G. were the recipients of
scholarships from the Spanish Ministry of Education. We thank
Dr. M.A. Prieto (CIB-CSIC, Madrid, Spain) for kindly providing us
with a strain of P. fluorescens GK13. P(3HO) substrate was kindly
supplied by Biopolis, S.A. (Valencia, Spain) and CPI (Newcastle,
United Kingdom).
References
[1] D.Y. Kim, H.W. Kim, M.G. Chung, Y.H. Rhee, J. Microbiol. 45 (2007) 87–97.
[2] L.L. Madison, G.W. Huisman, Microbiol. Mol. Biol. Rev. 63 (1999) 21–53.
[3] G.Q. Chen, Adv. Biochem. Eng. Biotechnol. 125 (2011) 29–45.
[4] G.Q. Chen, Q. Wu, Appl. Microbiol. Biotechnol. 67 (2005) 592–599.
[5] D. Jendrossek, R. Handrick, Annu. Rev. Microbiol. 56 (2002) 403–432.
[6] A. Schirmer, D. Jendrossek, J. Bacteriol. 176 (1994) 7065–7073.
[7] L.I. de Eugenio, J.L. García, P.L. García, M.A. Prieto, J.M. Sanz, Eng. Life Sci. 8
(2008) 260–267.
[8] A. Schirmer, D. Jendrossek, H.G. Schlegel, Appl. Environ. Microbiol. 59 (1993)
1220–1227.
[9] K.E. Jaeger, A. Steinbüchel, D. Jendrossek, Appl. Environ. Microbiol. 61 (1995)
3113–3118.
[10] R.A. Gross, M. Ganesh, W. Lu, Trends Biotechnol. 28 (2010) 435–443.
[11] A.C. Albertsson, R.V. Srivastava, Adv. Drug Deliv. Rev. 60 (2008) 1077–1093.
[12] R.A. Gross, A. Kumar, B. Kalra, Chem. Rev. 101 (2001) 2097–2124.
[13] U.T. Bornscheuer, C. Bessler, R. Srinivas, S.H. Krishna, Trends Biotechnol. 20
(2002) 433–437.
4. Conclusions
[14] A. Kumar, R.A. Gross, D. Jendrossek, J. Org. Chem. 65 (2000) 7800–7806.
[15] Y. Suzuki, T. Ohura, K.I. Kasuya, K. Toshima, Y. Doi, S. Matsumura, Chem. Lett.
29 (2000) 318–319.
[16] Y. Suzuki, S. Taguchi, T. Saito, K. Toshima, S. Matsumura, Y. Doi, Biomacro-
molecules 2 (2001) 541–544.
[17] Y. Suzuki, S. Taguchi, T. Hisano, K. Toshima, S. Matsumura, Y. Doi, Biomacro-
molecules 4 (2003) 537–543.
[18] J. Gangoiti, M. Santos, M.J. Llama, J.L. Serra, Appl. Environ. Microbiol. 76 (2010)
3554–3560.
[19] G.A.R. Nobes, R.J. Kazlauskas, R.H. Marchessault, Macromolecules 29 (1996)
4829–4833.
[20] J.T. Gorke, K. Okrasa, A. Louwagie, R.J. Kazlauskas, F. Srienc, J. Biotechnol. 132
(2007) 306–313.
[21] Y. Osanai, K. Toshima, S. Matsumura, Macromol. Biosci. 1 (2001) 171–177.
[22] M. Takwa, M.W. Larsen, K. Hult, M. Martinelle, Chem. Commun. 47 (2011)
7392–7394.
This report demonstrates the potential of using mcl-PHA
depolymerase of P. fluorescens GK13 as a novel catalyst for ester-
bond forming reactions in non-aqueous media. This is the first time
that a member of the family of mcl-PHA depolymerases has been
identified as a new enzyme for polymer synthesis showing dif-
ferent specificity and other valuable attributes in comparison to
Novozyme 435. The introduction of this new family of enzymes for
bond formation in anhydrous media might give the opportunity to
explore reactions that thus far were not successfully performed.
However, further research is required to evaluate the possible ben-
efits of the mcl-PHA depolymerase over other extensively studied
enzymes.
[23] S. Matsumura, K. Mabuchi, K. Toshima, Macromol. Rapid. Commun. 18 (1997)
477–482.
Acknowledgments
[24] M. Hans, H. Keul, M. Moeller, Macromol. Biosci. 9 (2009) 239–247.
[25] P.O. Syrén, K. Hult, ChemBioChem 11 (2010) 802–810.
[26] F. Haeffner, T. Norin, K. Hult, Biophys. J. 74 (1998) 1251–1262.
[27] H. Dong, H.D. Wang, S.G. Cao, J.C. Shen, Biotechnol. Lett. 20 (1998) 905–908.
[28] M.L. Focarete, M. Scandola, A. Kumar, R.A. Gross, J. Pol. Sci. 39 (2001) 1721–1729.
[29] 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.
This work was carried out in the framework of the IP project
“Sustainable Microbial and Biocatalytic Production of Advanced
Functional Materials” (BIOPRODUCTION, NMP-2-CT-2007-026515)
funded by the European Commission, the Spanish Ministry of