Oxidation reactions using air as oxidant thanks to silica nanoreactors containing GOx/peroxidases bienzymatic systems

Article abstract of DOI:10.1016/j.cattod.2010.02.065

A new kind of enzymes encapsulation method is presented. It combines the sol-gel method with a templating process using bilayers of phospholipids to provide an organized network of phospholipids inside the silica and in the same time protect the embedded

Full text of DOI:10.1016/j.cattod.2010.02.065

Catalysis Today 157 (2010) 94–100
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Oxidation reactions using air as oxidant thanks to silica nanoreactors containing
GOx/peroxidases bienzymatic systems
P. Laveille, L. Truong Phuoc, J. Drone, F. Fajula, G. Renard, A. Galarneau∗
Institut Charles Gerhardt Montpellier, UMR 5253 CNRS/UM2/ENSCM/UM1, ENSCM, 8 rue de l’Ecole Normale, 34296 Montpellier Cedex 5, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 2 April 2010
A new kind of enzymes encapsulation method is presented. It combines the sol–gel method with a tem-
plating process using bilayers of phospholipids to provide an organized network of phospholipids inside
the silica and in the same time protect the embedded enzymes, as if they were entrapped in a biological
membrane supported on silica. This brings a porosity control to the classical sol–gel encapsulation and
therefore increases the accessibility of the substrates to a maximum of enzymes. A first family of hybrid
phospholipids–silica materials, named Sponge mesoporous silica (SMS), has been synthesized, which
features a sponge-like structure. The formation of a negative curvature in the rigid phospholipid bilayers
was induced by adding ethanol (>50%) to the aqueous medium, which provokes the separation of the
head groups and the undulations of the lipid bilayers. Dodecylamine added to stabilize this structure
acts as a catalyst for silica condensation. Lactose was added to avoid direct interaction of the enzymes
with the silica at the interface with the lipids bilayer. Lipase was encapsulated in SMS and shows higher
activity for ester hydrolysis compared to commercial immobilized-lipase. An experiment design plan
was carried out to optimize SMS synthesis using hemoglobin (Hb) as biocatalyst for oxidation reactions
using H2O2. As a result, a new structure was identified, formed by the aggregation of nanoporous silica
capsules, named NPS, with a higher activity compared to SMS biocatalysts. These new biological nanore-
actors were used to encapsulate simulteanously Hb and glucose oxidase, to produce in situ H2O2 from air,
and perform oxidation reactions. This system was used for the elimination of the carcinogenic polycyclic
aromatic hydrocarbon pollutants from water.
Keywords:
Biocatalysis
Immobilized enzymes
Mesoporous silica
Phospholipids
Hemoglobin
Glucose oxidase
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
and the easiest, but is often susceptible to a progressive leaching.
The grafting of enzymes into a support avoid the leaching, but is
Enzymes are very selective catalysts and may represent a suit-
able response for the challenges of the 21st century in catalysis
with respect to improvement of the selectivity of the reactions. For
an industrial application, a challenge is to find a way to increase
the stability of enzymes towards catalytic inactivation by temper-
ature, inhibitors, mechanical treatments, solvents or pH variations
and to recover the biocatalyst from the medium after reaction.
The solution is to heterogenize the biocatalyst by immobilizing it
in an inorganic matrix without deactivating it (Fig. 1). More and
more studies have selected silica as inorganic supports [1–4] and
have shown that mesoporous silicas are very efficient supports to
maintain the activity of the biomolecules [5–7]. Different meth-
ods can be used to immobilize enzymes in inorganic supports such
as covalent binding, adsorption and sol–gel encapsulation (Fig. 1)
[8–11]. All of these methods have advantages and disadvantages.
The adsorption of enzymes in an inorganic support is the cheapest
more time consuming as it implies a previous functionalization
of the support before to react with a part of the protein, which
can sometimes lead to the denaturation of proteins. Silica sol–gel
entrapment represents a good compromise. However, the direct
interaction of proteins with silanols should be minimized to avoid
protein deformation, and, soluble excipients are added to stabi-
lize the correctly folded protein conformation. Additives help also
to stabilize proteins against the denaturing stresses encountered
upon sol–gel entrapment. Silica sol–gel has allowed to maintain
activity of enzymes [12] or bacteria [13–15] by using additives
such as sugars, glycerol, charged polymers (poly-vinylimidazole,
-ethyleneimine, -ethyleneglycol) or gelatin. For lipases encapsu-
lated in sol–gel, poly(vinyl alcohol) [16,17] has been used and the
synthesis has been further developed commercially by Fluka. Nev-
ertheless, in sol–gel encapsulation, the lack of controlled porosity
limits the diffusivity of the substrates. It is why we have devel-
oped a new concept of sol–gel synthesis [9,10]. It combines the
sol–gel method with a templating process using bilayers of phos-
pholipids to provide an organized network of phospholipids inside
the silica and in the same time protect the embedded enzymes
∗
Corresponding author.
E-mail address: anne.galarneau@enscm.fr (A. Galarneau).
0920-5861/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2010.02.065

P. Laveille et al. / Catalysis Today 157 (2010) 94–100
95
Fig. 1. Schematic representation of the different ways to immobilize enzyme.
(Fig. 1), as if they were entrapped in a biological membrane sup-
ported on silica. This brings a porosity control to the classical
sol–gel encapsulation and therefore increases the accessibility of
the substrates to a maximum of enzymes. Different structures of
biomaterials have been obtained and have been used for ester
hydrolysis and oxidation reactions with lipase and hemoglobin
(Hb), respectively. A bienzymatic system composed of glucose oxi-
dase and Hb allows to perform oxidation reactions from air by
producing in situ H2O2. The removal of carcinogenic polycyclic aro-
matic hydrocarbon (PAH) pollutants from water has been realized
thanks to this new nanobioreactor.
cylamine in 5.2 g ethanol, until a homogeneous emulsion was
formed. 1.35 g TEOS was then added slowly under vigorous stir-
ring for 15 min, and stir until a homogeneous emulsion was formed,
then the mixture was left for 24 h in static conditions at 37 ◦C. The
resulting powder was then centrifuged and washed 5 times with
10 mL of phosphate buffer followed by centrifugation. The samples
were then dried by lyophilization.
2.3. Synthesis of nanoporous silica capsules (NPS)
NPS synthesis uses the same reactants as for SMS synthesis,
with different molar ratios. For a typical NPS (NPS6), the molar
ratios used were: 1 TEOS/0.2 lecithin/0.06 dodecylamine/0.03 lac-
tose/83 H2O/9 ethanol, which corresponds to 0.05 g lactose, 7.2 mL
phosphate buffer containing the desired amount of enzyme, 0.7 g
lecithin, 0.05 g dodecylamine, 2 g ethanol and 1 g TEOS.
2. Experimental
2.1. Materials
Sponge mesoporous silica (SMS) and nanoporous silica cap-
sules (NPS) have been synthesized using tetraethoxysilane (TEOS)
from Aldrich, lecithin from egg yolk (l-␣ Lecithin) (Fluka) and
␤-d-lactose (Aldrich). Egg yolk lecithin is a natural and low cost
phospholipid. Lipase from Mucor Meihei and glucose oxidase
from Aspergillus Niger were purchased from Fluka. Bovine met-
hemoglobin (Hb) was obtained from Sigma–Aldrich (ref H2625).
H2O2 was obtained as a 35% solution from Sigma–Aldrich. Chemi-
cals for the buffers were reagent grade (Fluka). PAHs (naphthalene,
acenaphtene, phenanthrene, anthracene, fluoranthene, pyrene,
benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene,
benzo(g,h,i)perylene, indeno(1,2,3,cd)pyrene) and 2,2ꢀ-azino-
bis(3-ethylbenzthiazoline-6-sulphonic acid) diammonuim salt
(ABTS) were purchased from Sigma–Aldrich (98–99.8% purity).
Acetonitrile was of HPLC grade (SDS).
2.4. ABTS activity
The ABTS reaction is considered as a model reaction for the
peroxidase activity of enzymes [18]. For the evaluation of the
immobilized-Hb biomaterials, a home-made continuous flow cell
maintained at 25 ◦C, consisting of a peristaltic pump, a mechan-
ical stirrer, and a bottom opened flask equipped by a fritted
glass was used. Typically, 19.6 mL of phosphate buffer pH = 6
(I = 50 mM) are loaded to the flask together with 700 ␮L of ABTS
solution at a concentration of 75 mM (50 ␮mol ABTS) and 10–20 mg
of Hb-biomaterial. The pump and the stirrer are activated and
the reaction starts by adding 700 ␮L of H2O2 at a concentra-
tion of 75 mM (50 ␮mol H2O2). The formation of the colored
product (ABTS+, oxidized ABTS) is followed at 414 nm with an
UVIKON 930 spectrophotometer from Kontron Instrument poly-
stat 22/86602 Bioblock integration software. Activity of free Hb was
1015 ␮mol/min/g Hb.
2.2. Sponge mesoporous silica (SMS) synthesis
SMS synthesis was performed by using the following reac-
tants: TEOS, dodecylamine, lecithin, ␤-d-lactose. For a typical SMS
synthesis, the molar ratios used were: 1 TEOS/0.11 lecithin/0.08
dodecylamine/0.068 lactose/56 H2O/17 ethanol. A first solution
of 0.15 g lactose and 7 mL phosphate buffer (pH 5 for Hb and
GOx, and pH 8 for lipase), containing the desired amount of
enzyme (300 mg Hb, 50 mg GOx, 10 mg lipase/g TEOS), was pre-
pared at 37 ◦C and added slowly under vigorous stirring to a second
solution, prepared at 37 ◦C, containing 0.5 g lecithin and 0.1 g dode-
2.5. PAH degradation with NPS nanobioreactors
Studies of PAH oxidation were conducted in a 100 mL reaction
mixture containing 6 g/L Hb/GOx–NPS (or 0.8 g/L GOx–NPS and
6 g/L Hb–NPS), 300 nM of each PAH, citrate buffer pH = 5 (I = 50 mM),
150 mM glucose and 1% (v/v) acetonitrile. The NPS synthesis cor-
responds to the experiment 6 in the experiment design plan. The
kinetic of the PAH oxidation was followed by ultrafast liquid chro-

96
P. Laveille et al. / Catalysis Today 157 (2010) 94–100
2.6. Characterization
Powder X-ray diffraction (XRD) patterns were recorded on a
Bruker AXS D8 diffractometer by using Cu K␣ radiation and Ni fil-
ter. The adsorption–desorption isotherms for nitrogen at 77 K were
measured using a Micromeritics ASAP 2010 instrument on samples
previously calcined at 550 ◦C in air for 8 h to study the poros-
ity of the material after removal of phospholipids and enzymes.
Examination of the particles morphology was achieved using a
Hitachi S-4500 I scanning electron microscope (SEM). Transmis-
sion electron microscopy (TEM) was performed on a JEOL 1200 EX
II microscope operating at 120 keV. The particles were trapped in a
resin (LR White) and cut into slices 70 nm thick by ultramicrotomy
before to be imaged.
3. Results and discussion
Fig. 2. Schematic representation of sponge mesoporous silica (SMS) formation, from
phospholipids and lipids bilayer undulation.
3.1. Sponge mesoporous silica (SMS) encapsulation method
“Sponge mesoporous silica” (SMS) are synthesized using
lecithin/dodecylamine/lactose as templates in an ethanol/aqueous
media [9,10–19,20], and are suitable for enzyme encapsulation.
Lecithin belongs to the phospholipids family, which forms the
lipid matrix of biological membranes. As a biocompatible surfac-
tant, it is widely used in every day life, including human and
animal food, medicine, cosmetics and manifold industrial applica-
tions [21]. Because of its nearly cylindrical molecular shape, lecithin
cannot grow micelles by itself in aqueous media. Its tendency to
curve, described in terms of its spontaneous curvature, is very low,
and therefore it induces the formation of lamellar bilayer phospho-
lipids structures, described as membranes, vesicles or liposomes.
The effect of the addition of ethanol into biological membranes
in aqueous solution is a subject of debate and controversy. A very
recent study of structure simulation concerning the interaction of
matography (UFLC). PAH degradation was calculated from the
decreasing area of chromatogram peaks between reaction time
t = 0 and t = x min. Results of total degradation of PAH were taken
after 2 h. With this method, the detection limit is 15 7.5 nM.
Reactions were performed in flasks protected by aluminum foil to
avoid PAH photo-oxidation cross-reaction. To ensure reproducible
results, after reaction and before UFLC analysis, samples have been
diluted in 50% (v/v) acetonitrile and filtrated on a regenerated cellu-
lose membrane syringe filter from Macherey-Nagel (25 mm, 0.2 ␮m
pore diameters) to eliminate solid particles and to stop the reac-
tion. Blank reactions without glucose have also been performed
to estimate the proportion of PAH adsorbed on the biomateri-
als.
Fig. 3. (a) TEM, (b) XRD, (d) SEM of as-synthesized SMS and (c) nitrogen sorption at 77 K of calcined SMS.

P. Laveille et al. / Catalysis Today 157 (2010) 94–100
97
Table 1
Experiment design plan to optimize SMS synthesis. For 7.2 mL phosphate buffer
pH 6, the studied compositions are: Hb (−1) = 50 mg, (+1) = 300 mg; lactose (La)
(−1) = 50 mg, (+1) = 300 mg; lecithin (Lec) (−1) = 300 mg, (+1) = 700 mg; dodecy-
lamine (C12NH2) (−1) = 50 mg, (+1) = 150 mg; ethanol (−1) = 2 g, (+1) = 10 g; TEOS
(−1) = 1 g, (+1) = 3 g.
N
Hb
La
Lec
C12NH2
EtOH
TEOS
1
−1
1
−1
−1
1
−1
−1
−1
−1
1
1
1
1
−1
−1
−1
−1
1
1
1
1
−1
−1
−1
−1
1
1
1
1
−1
−1
−1
−1
1
1
1
1
−1
−1
−1
−1
−1
−1
−1
−1
1
1
1
1
1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
−1
1
1
1
1
1
1
1
1
1
−1
1
1
3
4
5
6
7
−1
1
1
−1
−1
1
−1
−1
1
1
−1
−1
1
−1
8
9
1
1
−1
1
−1
−1
1
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
−1
−1
1
−1
1
1
−1
1
−1
−1
1
−1
1
1
1
1
−1
1
1
1
−1
−1
1
−1
−1
1
−1
−1
−1
−1
−1
−1
−1
−1
1
1
1
1
1
1
−1
−1
1
−1
1
1
Fig. 4. Activity towards ester hydrolysis per g of biomaterials of lipase immobilized
in SMS, in sol–gel or in resin, in function of ester chain length.
−1
1
−1
−1
1
−1
1
−1
1
1
1
−1
−1
1
ethanol with phospholipid bilayers helps to understand the effect
of ethanol on bilayer structures [22]. The addition of ethanol gives
rise to an increase in the head group area leading to: (i) a disorder-
ing of the lipid acyl chains; (ii) a diminution of the bilayer thickness
by interdigitation of the lipid chain; and (iii) an expansion of the
membrane, creating a reduction of its rigidity and an increase of its
fluidity, responsible of an enhanced permeability. At high ethanol
content (50% in volume), ethanol molecules penetrates into the
hydrophobic core of the bilayer increasing the hydrophilicity of the
membrane interior, and a wave motion (Fig. 2) of the lipid bilayer
arises, creating positive and negative curvatures in the membrane.
The insertion of dodecylamine into the phospholipids/EtOH/H2O
system stabilizes the structure of the resulting mesophase by
maintaining the space between lipid heads. In the same time, dode-
cylamine catalyzes the polymerization of the alkoxyde of silica
(TEOS) around this mesophase to lead to the SMS materials. Lac-
tose is present at the interface between the membrane and the
silica to minimize the direct interaction of the enzyme with silica.
Resulting SMS materials (Fig. 3) show a three-dimensional “sponge
mesoporous” porosity with a pore diameter of about 5–6 nm, in
accordance to the length of a lecithin bilayer, evidenced by XRD
and by nitrogen adsorption of calcined SMS. Surface area and pore
volume of calcined SMS increases with the amount of dodecylamine
used in the synthesis from 500 to 700 m2/g and from 0.4 to 0.7 mL/g
for dodecylamine/TEOS molar ratio of 0.05–0.13, respectively [20].
SMS features spherical particles of 3 ␮m as observed in SEM images
(Fig. 3).
−1
1
−1
−1
1
1
1
1
1
1
1
1
−1
1
1
1
−1
−1
1
−1
−1
1
1
1
1
1
−1
−1
1
−1
1
1
immobilization and the amount of enzyme is not given by the
supplier, so its efficiency is unknown. However in real catalytic
application, the activity per g of biocatalyst is the most relevant
data and lipase-SMS is the most active. It was demonstrated that
lactose was the more efficient sugar to preserve enzyme-SMS
activity. The high catalytic activity was explained by the use of
lecithin/dodecylamine assembly structure permitting facile diffu-
sion of substrates, through the generation of a 3D mesoporosity in
silica sol–gel, and also to the use of natural and biocompatible sur-
factants. However, a change in typoselectivity of the enzyme was
noticed for lipase encapsulated in SMS materials. Indeed, whereas
the free lipase hydrolyzes preferentially long chain esters with
the highest selectivity for chains with 8 carbons, the lipase-SMS
exhibits the highest selectivity for shorter chains with 2 or 3 car-
bons, that free lipases are not able to hydrolyze. A change in the
conformation of the lipase has occurred during the SMS encapsu-
lation changing the selectivity of the enzyme.
In the presence of lipase from Mucor miehei during the syn-
thesis, the lipase-SMS exhibited higher catalytic activity per g
of biocatalyst (Fig. 4), compared to the same lipase immobi-
lized in commercially available supports: lipase encapsulated
in silica sol–gel (Fluka) or post-synthesis resin immobilization
(“Lipozyme”-Fluka) [9,10]. The highest efficiency of SMS encap-
sulation was calculated by comparing the relative activity of the
immobilized enzyme to the activity of the free enzyme and was
of 154% for lipase-SMS and 4% lipase-sol–gel for the hydrolysis
of p-nitrophenol propionate. This calculation was done thanks to
the dissolution of the silica matrix in basic solution and by mea-
suring the amount of resulting protein by the Bradford technique.
Unfortunately this dissolution method was unsuccessful for resin
3.2. Nanoporous silica capsules (NPS) for enzyme encapsulation
In order to increase the performance of SMS materials, an exper-
iment design plan of 32 experiments (Table 1) has been performed
by varying the reactants molar ratio and using hemoglobin (Hb)
as active biocatalyst for oxidation reactions in presence of H2O2
[23–25]. The ABTS oxidation test [18] was performed to evalu-
ate rapidly the activity of the biomaterials. In Fig. 5, only the
results for the higher amount of Hb (300 mg Hb/g TEOS) in the bio-
materials have been presented for more clarity, as biomaterials
prepared with the lower amount of Hb (50 mg/g TEOS) features
much lower activity. From the experiment design plan, one can
calculate the favorable factors responsible for a higher activity of

98
P. Laveille et al. / Catalysis Today 157 (2010) 94–100
Fig. 5. ABTS activities per g of biomaterials of encapsulated hemoglobin synthesized from the experiment design plan described in Table 1 (for 300 mg Hb/g TEOS).
the biomaterials: Hb (+4.2) > lecithin (+0.9) > lactose (+0.3), and the
unfavorable factors: dodecylamine (−2.17) > ethanol (−1.8) > TEOS
(−0.2). Unfortunately, this experiment design plan could not be
totally exploited as a new material family, named nanoporous sil-
ica capsules (NPS), was formed at lower amount of ethanol (25%
in volume) characterized by a larger d-spacing (dXRD = 8 nm) than
SMS materials (dXRD = 6 nm). Most of the members of this new NPS
family (as the one coming from experiments number 2, 4, 6, 8)
exhibits a higher activity than SMS biomaterials. The characteriza-
tion of the NPS resulting from experiment number 6 (named NPS6)
shows that NPS biomaterials are built of silica nanocapsules of 7 nm
aggregated in larger cells of 20 nm as raspberries. Each nanocapsule
is connected to the others by an opening of around 1 nm (Fig. 6).
The SEM morphology of the NPS materials is totally different of the
one of SMS materials, as the NPSs feature an aggregation of nano-
metric particles and SMSs present large sphere of 2–3 ␮m. After
Fig. 6. (a) TEM, (b) XRD, (d) SEM of as-synthesized NPS6 and (c) nitrogen sorption at 77 K of calcined NPS6.

P. Laveille et al. / Catalysis Today 157 (2010) 94–100
99
Fig. 7. Schematic representation of porous nanoporous silica capsules (NPS) forma-
tion, from phospholipids and formation of micelles of phospholipids inside lipids
bilayer.
calcination, the XRD peak of NPS disappears; NPS structure is not
stable upon calcination in contrary to SMS materials. By nitrogen
adsorption only some residual porosity can be detected. The for-
mation of NPS materials is explained by the feasibility of micelles
of phospholipids when lower ethanol amount and larger amount of
phospholipids are used in comparison to SMS materials. In simula-
tion study [22], it has been shown that ethanol facilitate the water
penetration in the membrane interior as well as the migration of
phospholipids, giving rise to clusters of water molecules (in blue in
Fig. 7) surrounded by lipids head groups in the membrane interior,
forming micelles of 7 nm large and 4 nm high inside the undu-
lated bilayer of 7 nm high, with possible connections between the
micelles and the bilayer. The formation of these micelle-like struc-
tures is on line with our TEM observation of NPS materials, where
silica takes the place of the water in the interior of the bilayer and
polymerize around the micelles of phospholipids, which are then
expulsed from the distorted membrane.
Fig. 8. ABTS activities per g of biomaterials for different amount of Hb encapsu-
lated in NPS6 with addition of H2O2 or simultaneous with GOx (50 mg/g TEOS) with
addition of glucose (without peroxide).
3.3. NPS as nanobioreactors for oxidation reaction from air
The choice of the biocatalyst for an oxidation reaction is cru-
cial and depends on its activity, stability, and its economic impact
on the process [26]. Enzymes such as oxygenases and oxidases
(P450 cytochromes [27] and laccases [28]), which directly use O2
to oxidize their substrates, can be used. The disadvantage to this
biotechnology is the high cost of the enzymes due to expen-
sive extraction and purification processes that are needed. Bovine
hemoglobin, a non-enzymatic member of the hemoprotein fam-
ily, presents the advantage to be a low cost waste product of the
food industry, and is able to perform oxidation reactions through a
peroxidase-like pathway using H2O2 [23–25]. However, in indus-
trial processes, H2O2 is difficult to handle and oxidation from
air is preferred. In order to use directly air as oxidant, a bienzy-
matic system has been built thanks to another industrially used
enzyme, the glucose oxidase (GOx), which produces in situ H2O2
from oxygen dissolved in water and glucose (Fig. 8). GOx and
Hb have been encapsulated in the same NPS6 nanobioreactor
with 50 mg GOx/g TEOS and different amount of Hb from 50 to
300 mg Hb/g TEOS. The activity of Hb/GOx–NPS towards ABTS oxi-
dation has been compared to the direct addition of H2O2 to Hb–NPS
(Fig. 8). Results show that the activity of the biomaterials increases
linearly with the amount of Hb and that Hb/GOx–NPS nanobioreac-
tors are less active than direct addition of H2O2 on Hb–NPS, but are
enough efficient to perform oxidation reactions without adding any
peroxides. The enzyme–NPS biocatalysts has been stored at 4 ◦C and
after 60 days they show the same activity has the free enzyme in the
same conditions (maintain 70% of activity). Reuse of enzyme–NPS
has not been checked as we are working in batch condition with
H2O2 (added or in situ produced), which is a suicide substrate able
to destroy enzyme. Only continuous flow reaction could allow to
test recyclability, but as NPS are nanoparticles, columns are impos-
sible to fill, therefore only batch reactions are considered. In the
following, the more active biomaterial (synthesized from NPS6
conditions with 50 mg GOx and 300 mg Hb/g TEOS) has been con-
sidered.
Recently, we have shown that free Hb was able to oxidize strong
carcinogenic pollutants such as polycyclic aromatic hydrocarbon
(PAH), such as anthracene (ANT), in water (with 1% acetonitrile)
with a large excess of H2O2 and Hb compared to ANT, with molar
ratios of [H2O2]/[ANT] > 1000 and [Hb]/[ANT] > 1 [25]. We have also
shown that 65% of oxidized anthracene remained covalently bind
to the protein [25]. Hb allows in the same time to oxidize and
to fix the oxidation products. Heterogenizing Hb by NPS encap-
sulation will lead therefore to a direct removal of PAHs. In this
study, in order to remove PAHs of waste water, such as in the
waste water of oil refineries, Hb/GOx–NPS (6 g/L) was added to
an aqueous solution containing 11 PAH (300 nM each) without
any peroxides. Then 150 mM of glucose was added and let react
for 2 h. The biomaterial is then removed by centrifugation and
the solution is analyzed. All PAHs have reacted with Hb/GOx–NPS
(Fig. 9). Some PAHs are more reactive than others as observed
for anthracene and acenaphtene with 100 and 86% disappearance,
respectively, and only 18 and 10% for phenanthrene and naphtha-
lene. In total, more than 63% of PAH have disappeared from the
water solution by this process. It is to notice that a blank test done
without glucose addition shows that high molecular weight PAHs
are strongly adsorbed in NPS materials, and that low molecular
weight PAH, like anthracene and acenaphtene, are weakly adsorbed
but strongly oxidized by the nanobioreactor. Hb/GOx–NPS oxi-
dizes 80% of anthracene and 75% of acenaphtene in 2 h without
adding any peroxides, whereas more than 20 h are needed for con-

100
P. Laveille et al. / Catalysis Today 157 (2010) 94–100
and in the same time protect the embedded enzymes. Lipase-SMS
biocatalysts exhibit a better activity for ester hydrolysis compared
to commercial immobilized biocatalysts. Hb/GOx–NPS nanobiore-
actors were able to oxidize strong pollutants in water such as
polycyclic aromatic hydrocarbon by simply adding glucose to the
solution, without any use of peroxides. These new nanobioreactors
bring the possibility to develop biotechnology for water depollu-
tion.
Acknowledgements
The authors are grateful for the financial support from the TOTAL
France S.A company, and, thank Yann Ambacher and Gerard Bau-
douin from ENSCM for their participation in the experiment design
plan.
References
[1] K.J.J. Balkus, J.F. Diaz, J. Mol. Catal. B: Enzym. 2 (1996) 115.
[2] M.T. Reetz, R. Wenkel, D. Avnir, Synthesis 6 (2000) 781.
[3] J. Livage, T. Coradin, C. Roux, J. Phys.: Condens. Matter 13 (2001) 673.
[4] Y. Wang, F. Caruso, Chem. Mater. 17 (2005) 953.
[5] P. Villeneuve, J.M. Muderhwa, J. Graille, M.J. Haas, J. Mol. Catal. B: Enzym. 9
(2000) 113.
[6] M. Hartmann, Chem. Mater. 17 (2005) 4577.
[7] H.H.P. Yiu, P.A. Wright, J. Mater. Chem. 15 (2005) 3690.
[8] A.S. Roger, Adv. Synth. Catal. 349 (2007) 1289.
[9] M. Mureseanu, A. Galarneau, G. Renard, F. Fajula, Langmuir 21 (2005) 4648.
[10] A. Galarneau, M. Mureseanu, S. Atger, G. Renard, F. Fajula, New J. Chem. 30
(2006) 562.
Fig. 9. (Grey bars) percentage of PAH removal from water using Hb/GOx–NPS6
nanobioreactor with glucose addition (without peroxide). (White bars) percentage
of adsorbed PAH, the difference between grey and white bars are the percentage of
oxidized PAH. List of PAHs: naphthalene (NAP), acenaphtene (ACE), phenanthrene
(PHE), anthracene (ANT), fluoranthene (FLT), pyrene (PYR), benzo(b)fluoranthene
(BbF), benzo(k)fluoranthene (BkF), benzo(a)pyrene (BaP), benzo(g,h,i)perylene
(BgP), indeno(1,2,3,cd)pyrene (IND).
[11] Y.S. Chaudhary, S.K. Manna, S. Mazumdar, D. Khushalani, Micropor. Mesopor.
Mater. 109 (2008) 535.
[12] L.H. Zhou, Y.H. Tao, J. Hu, X. Han, H.L. Liu, Y. Hu, J. Porous Mater. 15 (2008) 653.
[13] N. Nassif, C. Roux, T. Coradin, M.N. Rager, O.M.M. Bouvet, J. Livage, J. Mater.
Chem. 13 (2003) 203.
[14] N. Nassif, O. Bouvet, M.N. Rager, C. Roux, T. Coradin, J. Livage, Nat. Mater. 1
(2002) 42.
[15] N. Nassif, A. Coiffier, T. Coradin, C. Roux, J. Livage, O. Bouvet, J. Sol-Gel Sci.
Technol. 26 (2003) 1141.
[16] M.T. Reetz, Tetrahedron 58 (2002) 6595.
ventional heterogeneous catalysts like Ti–SiO2 using peroxides to
oxidize only anthracene, the more reactive PAH [29]. A compari-
son has been performed with a separate encapsulation of the two
enzymes Hb and GOx to evaluate the efficiency of the proximity of
the enzymes. A mixture of two NPS composed of 0.8 g/L GOx–NPS
(50 mg GOx/g TEOS) and 6 g/L Hb–NPS (300 mg Hb/g TEOS) has been
added to the water solution of the 11 PAHs. After addition of
glucose, similar results (not shown) have been obtained for the
(Hb–NPS + GOx–NPS) separated encapsulations compared to the
(Hb/GOx–NPS) simultaneous encapsulation, with a slightly higher
oxidation rate reaction for the biencapsulated Hb/GOx–NPS system.
NPS encapsulation is therefore suitable for mono- and polyenzy-
matic systems and can perform oxidation reaction from air, when
GOx is used to produce in situ H2O2.
[17] M.T. Reetz, P. Tielmann, W. Wiesenhofer, W. Konen, A. Zonta, Adv. Synth. Catal.
345 (2003) 717.
[18] R.E. Childs, W.G. Bardsley, Biochem. J. 145 (1975) 93.
[19] A. Galarneau, G. Renard, F. Fajula, patent, Fr. 04. 13119 (2006).
[20] A. Galarneau, G. Renard, M. Mureseanu, A. Tourrette, C. Biolley, M. Choi, R. Ryoo,
F. Di Renzo, F. Fajula, Micropor. Mesopor. Mater. 104 (2007) 103.
[21] A. Wendel, Kirk-Othmer Encyclopedia Chem. Technol. 15 (1995) 192.
[22] A.A. Gurtovenko, J. Anwar, J. Phys. Chem. B 113 (2009) 1983.
[23] M. Ortiz-Leon, L. Velasco, R. Vazquez-Duhalt, Biochem. Biophys. Res. Commun.
215 (1995) 968.
[24] E. Torres, R. Vazquez-Duhalt, Biochem. Biophys. Res. Commun. 273 (2000) 820.
[25] P. Laveille, A. Galarneau, J. Drone, G. Renard, C. Bailly, C. Bedel-Cloutour, S.
Pulvin, D. Thomas, Biotechnol. J. 4 (2009) 1460.
[26] C. Regalado, B.E. García-Almendárez, M.A. Duarte-Vázquez, Phytochem. Rev. 3
(2004) 243.
4. Conclusion
“Sponge mesoporous silica” (SMS) and “nanoporous silica cap-
sules” (NPS), synthesized using lecithin/dodecylamine/lactose as
templates in an ethanol/aqueous media, are suitable for enzyme
encapsulation. SMS and NPS encapsulation combines the sol–gel
method with a templating process using bilayers of phospholipids
to provide an organized network of phospholipids inside the silica
[27] P.A. England, C.F. Harford-Cross, J.A. Stevenson, D.A. Rouch, L.L. Wong, FEBS
Lett. 424 (1998) 271.
[28] A. Majcherczyk, C. Johannes, A. Huttermann, Enzyme Microb. Technol. 22
(1998) 335.
[29] R.S. Araujo, D.C.S. Azevedo, E. Rodriguez-Castellon, A. Jimenez-Lopez, C.L. Cav-
alcante Jr., J. Mol. Catal. A: Chem. 281 (2008) 154.