Design and optimization of an enzymatic membrane reactor for tetracycline degradation

Article abstract of DOI:10.1016/j.cattod.2014.02.051

The tetracycline, antibiotic considered as a recalcitrant pollutant, was successfully depleted from model aqueous solutions by immobilized laccase from Trametes versicolor in an enzymatic membrane reactor. The results obtained show that tetracycline is depleted from water solutions at room temperature and without adding any extra chemicals. The degradation of tetracycline in aqueous solutions at 20 mg L^-1 during 24 h, with equivalent amounts of free or immobilized biocatalyst, allowed reaching a tetracycline degradation yield of 56% with an enzymatic membrane whereas it was only of 30% with free laccase. This result highlights the good reactivity and stability of the immobilized enzyme for the degradation of tetracycline. Moreover, the enzymatic membrane reactor was able to reach a constant degradation rate of 0.34 mg of tetracycline per hour during 10 days.

Full text of DOI:10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Design and optimization of an enzymatic membrane reactor for
tetracycline degradation
M. de Cazesa, M.-P. Bellevillea, E. Petita, M. Llorcab, S. Rodríguez-Mozazb, J. de Gunzburgc
, D. Barcelób,d, J. Sanchez-Marcanoa,∗
a Université de Montpellier 2, Institut Européen des Membranes, UMR 5635 (CNRS-ENSCM-UM2), CC 047, 2 Place Eugène Bataillon, 34095 Montpellier
Cedex 5, France
b Catalan Institute for Water Research (ICRA), Scientific and Technological Park of the University of Girona, H2O building, C/Emili Grahit 101, E-17003
Girona, Spain
c Da Volterra, Le Dorian, building B1, 172 rue de Charonne, 75011 Paris, France
d Water and Soil Quality Research Group, Dept. of Environmental Chemistry, IDAEA-CSIC, Jordi Girona 18-26, 08034 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 July 2013
Received in revised form 23 October 2013
Accepted 10 February 2014
Available online xxx
The tetracycline, antibiotic considered as a recalcitrant pollutant, was successfully depleted from model
aqueous solutions by immobilized laccase from Trametes versicolor in an enzymatic membrane reactor.
The results obtained show that tetracycline is depleted from water solutions at room temperature and
without adding any extra chemicals. The degradation of tetracycline in aqueous solutions at 20 mg L−1
during 24 h, with equivalent amounts of free or immobilized biocatalyst, allowed reaching a tetracycline
degradation yield of 56% with an enzymatic membrane whereas it was only of 30% with free laccase.
This result highlights the good reactivity and stability of the immobilized enzyme for the degradation of
tetracycline. Moreover, the enzymatic membrane reactor was able to reach a constant degradation rate
of 0.34 mg of tetracycline per hour during 10 days.
Keywords:
Enzymatic membrane reactor
Active membrane
Laccase
© 2014 Elsevier B.V. All rights reserved.
Tetracycline
Wastewater treatment
Micropollutant degradation
1. Introduction
antibiotic resistance by bacteria and then be an important source
of public health problems in the future [7].
With the background of an aging population and increasing
urbanization, pharmaceutical products (PPs) and endocrine dis-
rupting chemicals (EDCs) have been continuously released in the
environment for a long time without being considered as priority
pollutants to target.
As conventional wastewater treatment technologies are not effi-
cient enough to completely remove pharmaceuticals from water,
such products are currently found in water effluents from sewage
facilities, as well as in surface water, in groundwater, adsorbed on
sediments and even in drinking water [1–3]. Furthermore, ecotox-
icity studies have demonstrated that PPs could affect the growth,
reproduction and behavior of birds, fishes, invertebrates, plants
and bacteria [4–6]. In particular, the presence of low concentra-
tions of antibiotics in wastewaters could cause the development of
Indeed, lately important research efforts have been done in
order to find a system to eliminate the PPs before rejecting the
effluents to the environment. Among the different processes tested
(physical adsorption, chemical or biological reactions) for the
depletion of certain groups of pollutants [8–12], the use of bio-
catalysts such as laccases, glycosylases, proteases and lipases have
been found to be particularly efficient [13]. In particular laccases
are able to oxidize a wide range of pollutants at room temperature
within a large range of pH using as oxidant the oxygen dissolved
in water. Consequently, some reports have noticed the potential of
laccase-catalyzed reactions for the removal of a large spectrum of
pollutants [14–19].
To overcome the drawbacks related to enzymes cost, biocata-
lysts can be immobilized on a large variety of supports [16,18] as
well as in membranes and used in enzymatic membrane reactors
(EMRs) [20–22]. In EMRs the substrate solution flows through the
membrane to the biocatalyst as a result of transmembrane pres-
sure. Then the reaction takes place simultaneously with the mass
transfer process through the membrane and the product is recov-
ered in the permeate. Thus a precise control of the reaction with
∗
Corresponding author. Tel.: +33 467 149 149/+33 467 1419;
fax: +33 467 149 119.
E-mail address: sanchez@iemm.univ-montp2.fr (J. Sanchez-Marcano).
http://dx.doi.org/10.1016/j.cattod.2014.02.051
0920-5861/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
M. de Cazes et al. / Catalysis Today xxx (2014) xxx–xxx
2.2. Immobilization protocol
2
Active membranes were prepared according to a 3-step pro-
cedure developed by Belleville et al. [26] and adapted for laccase
immobilization by Chea et al. [14]. First, the wet ceramic supports
(␣-alumina tubular membranes from Pall-Exekia, France (15 cm of
length, 1 cm of external diameter and 0.7 cm of internal diameter,
0.2 ␮m of mean pore size) were coated with a gelatin layer by fil-
trating a gelatin solution at 1 g L−1. Then the bio-polymer layer was
activated by a glutaradehyde solution (4% (w/v) for 1 h and finally,
the laccase (10 g L−1 solution) was let to react with free aldhehyde
groups of glutaraldehyde during 2 h. All the solutions were pre-
pared in a 50 mM phosphate buffer pH 7 and after each step, the
excess solution was removed by rinsing 4 times the membrane with
the same phosphate buffer. The active membranes were then stored
in a desiccator with P2O5 until being used. Blank membranes were
prepared with the same method but without enzymes.
Fig. 1. Structure of tetracycline molecule.
minimized substrate and catalyst losses, faster reactions, higher
yields and cleaner products can be expected [23].
This work describes the study of the potential enzymatic degra-
dation of tetracycline (TC), a recalcitrant antibiotic present in some
wastewaters. The laccase from Trametes versicolor was chosen as
the biocatalyst because this biocatalyst has already demonstrated
its activity for the degradation of this antibiotic [24]. For this pur-
pose the enzyme was covalently linked onto a ceramic membrane
previously coated with a polymer layer (i.e. gelatin). A comparison
between free and immobilized enzymes was also carried out to
evaluate the performances of the EMR and the stability of grafted
enzymes. In addition different operating parameters like pH and
temperature were studied in order to determine the optimal oper-
ating conditions for the tetracycline depletion. To our knowledge
it is the first work reporting tetracycline degradation with immo-
bilized laccase on ceramic membranes.
2.3. Pilot unit
The pilot unit used for EMR runs is shown in Fig. 2. It was built
with stainless steel and PTFE in order to avoid adsorption problems.
The EMR can be operated with or without recirculating the reten-
tate (Fig. 2). In this exploratory work we carried out experiments in
batch configuration (the permeate valve was kept close except dur-
ing sample withdrawing and permeate flux measurements). The
temperature, the trans-membrane pressure and the flow rate can
be measured and controlled with sensors and a cryothermostat.
Since the feeding tank was open to atmospheric pressure, there
was no need to add extra oxygen in the water. It was indeed proved
that adding oxygen by sending pressurized air into water had no
significant impact on dissolved oxygen concentration level.
2. Experimental
2.1. Products
2.4. Enzymatic degradation
Commercial powder of laccase from T. versicolor (activity
≥10 U mg−1, ref. 51639), tetracycline (≥98.0%, ref. 87128) (Fig. 1),
gelatin, glutaraldehyde and ABTS (≥98.0%, ref. 11557) were pur-
chased from Sigma-Aldrich. Mono-channel ceramic membranes
(␣-alumina) were supplied by Pall-Exekia (pore diameter of 0.2 ␮m,
15 cm long, external/internal diameter of 1 and 0.7 cm, effective
area of 28.6 × 10−4 m2). Tetracycline solutions were prepared at
20 mg L−1 in osmosed water (pH = 6) or in different 50 mM cit-
rate/phosphate buffers (pH 3 to 7). The TC is usually encountered
in wastewaters (hospital and wastewaters treatment plants) in a
concentration range from 1 to 900 ␮g L−1 [25]. The model solutions
tested in this work were prepared with buffers or osmosed water
and pure TC at 20 mg L−1. This concentration is high compared to
the actual concentrations noted above. It was arbitrarily chosen in
order to have a good precision on depletion rates while allowing
the identification of the degradation products.
In order to determine the optimal operating conditions and
to compare the activity of free and immobilized enzymes, sev-
eral experiments were carried out with 100 mL of tetracycline
solutions (20 mg L−1) prepared either in 50 mM citrate-phosphate
buffer (pH 3 to 7) or in osmosed water (pH 6) at 25 ◦C for 24 h in
a stirred tank bioreactor. In the case of free enzyme experiments
an amount of commercial powder was added to the tetracycline
solution, whereas for immobilized enzymes we used small pieces
of a crushed enzymatic membrane (EM). The amount of crushed
EM used for the reaction corresponds to the same concentration
of enzymes used for free enzymes experiments. Before degrada-
tion experiments the small portions of the crushed membrane were
washed with osmosed water in order to eliminate potential free lac-
cases that might not have been rinsed properly during the grafting
step. Tetracycline auto-degradation and adsorption were estimated
Fig. 2. Enzymatic membrane bioreactor and pilot unit.
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
M. de Cazes et al. / Catalysis Today xxx (2014) xxx–xxx
3
with blanks experiments carried out under identical conditions:
one containing only a solution of tetracycline (blank free) and the
other with a portion of a blank membrane crushed in small pieces
(blank immobilized).
initial mass of enzymes in the grafting solution and the mass of
immobilized enzymes (Eq. (3)).
ꢂ
ꢃ
Agrafted = Ainitial
=
Aleft + ˙Arinsing + ˙Aconditioning
(2)
(3)
Tetracycline degradation in the EMR was carried out in a batch
configuration where the retentate was continuously recycled and
the permeate exit was closed. Before each run, membranes were
first hydrated with osmosed water in the pilot unit in order
to eliminate potential free laccases that might not have been
rinsed properly during the grafting step. Reactions were run with
20 mg L−1 of TC in osmosed water with a total volume of 2 L. The
experiment was carried out during 10 days but the substrate solu-
tion was periodically replaced by a fresh one. The residual TC
percentage contained in each sample was calculated using the fol-
lowing equation (Eq. (1)):
Agrafted
ꢀimmobilization
=
× 100
Ainitial
where Agrafted corresponds to the activity of enzymes immobi-
lized on the membrane, Ainitial is the activity of enzymes in an
aliquot of the initial solution left at room temperature and ana-
lyzed with other solutions after the immobilization process, Aleft
is the enzymatic activity of the solution used for grafting after
enzymes immobilization, Arinsing is the activity of enzymes in the
rinsing solutions and Aconditioning the enzymatic activity in the rins-
ing solution used for the membrane hydration before using the
enzymatic membrane for degradation. ꢀimmobilization corresponds
to the immobilization yield of laccase on the ceramic membrane
(%).
ꢀ
ꢁ
CTC,sample
CTC,feed
Residual tetracycline % =
( )
× 100
(1)
where CTC,feed corresponds to the TC concentration in the initial
solution (mg L−1) and CTC,sample is the TC concentration measured
in the retentate (mg L−1).
The results obtained and presented in this study are the mean
values of several experiments realized under similar conditions.
After being used the ceramic supports were washed using the
standard cleaning procedure and their permeability was measured
to determine if they were reusable [14].
The characterization of transformation products (TPs) generated
during the treatment was performed by liquid chromatogra-
phy coupled to a high resolution Orbitrap mass spectrometer
(LC–HRMS). Samples were prepared and eluted as described by
Gros et al. [27]. Chromatographic separation was achieved by using
an Aquity HSS T3 column (50 mm × 2.1 mm i.d., 1.7 ␮m particle
size) in a TurboflowTM system coupled to an AccelaTM UHPLC
system (Thermo Scientific). The MS analysis was performed with
an electrospray ionization (ESI) interface coupled to a linear ion
trap-orbitrap mass spectrometer (LTQ Orbitrap VelosTM, Ther-
mos Scientific) under positive and negative ionization conditions
as described in Llorca et al. (under preparation). The mass accu-
racy accepted for the experiments was always within 5 ppm. The
MS acquisition was performed in full scan mode [100–1000 Da]
at resolving power of 60,000 FWHM. In parallel, data-dependent
MS/MS acquisition was carried out according to mass-to-charge
(m/z) ion intensity higher than 100.
2.5. Determination of the apparent kinetics parameters
The apparent kinetics parameters of Michaelis–Menten equa-
tion (Km and Vmax) of free and immobilized laccase were estimated
from initial reaction rates determined in batch experiments using
either TC or ABTS as substrate. All the reactions were carried out at
25 ◦C, with a concentration equivalent to 35 U mL−1 of free enzymes
or to 20 U mL−1 of immobilized enzymes. TC solutions were pre-
pared in osmosed water (pH 6) varying the concentration from 22.5
to 225 ␮M whereas ABTS solutions were prepared at 10, 15, 25, 50
and 100 ␮M in a 50 mM citrate-phosphate buffer pH 4.
3. Results and discussion
3.1. Characterization of the active membrane
2.6. Analytical methods
SEM micrographs (Fig. 3) show that the gelatin is only deposited
on the ceramic membrane surface, the gelatin layer thickness is
0.6 ␮m. The average immobilization yield determined as explained
above (Section 2.5) was equal to 13%. According to this value, a 3 cm
long piece of active membrane crushed in small parts into 100 mL of
water corresponds to a concentration of 0.01 g L−1 of Sigma product
containing laccase from T. versicolor.
The kinetics parameters of Michealis–Mentel equation (Km and
Vmax) of both free and immobilized enzymes were estimated for
two substrates (ABTS and TC). The calculated values obtained
thanks to Lineweaver–Burk equation are reported in Table 1.
It is known that the enzymatic activity is better with high Vmax
and low Km values. For both substrates, the Km value is 5 about times
lower with grafted enzymes. Therefore the bounded enzymes have
a higher apparent affinity with the substrate and in principle have
to be more efficient to degrade ABTS and TC than free enzymes.
The literature confirms these results because it was proved that
grafting enhanced the enzyme stability thanks to the bonding and
the accessibility of the active site by the substrate [28,29]. How-
ever regarding the Vmax values, although the concentration of free
enzyme was higher (35 U mL−1 and 20 U mL−1 for free and immo-
bilized enzyme, respectively) the Vmax value observed for grafted
enzymes was always higher. The rate of reaction would be thus
higher with immobilized laccase compared to the free laccase.
The structure of ceramic support and active membrane was
observed using scanning electron microscopy (SEM) (Hitachi
S4500).
Tetracycline concentration was determined by high-
performance liquid chromatography coupled to triple-quadrupole
mass spectrometry (HPLC–MS). Samples were injected through
an Onyx® C18 column (25 mm × 4.6 mm) with a Waters e2695
Separations Module, and the 410 m/z fragment was detected with
a Micromass Quattro micro API device at 120 ◦C. The isocratic
mobile phase used was made of HPLC grade water, 99% formic
acid, and acetonitrile (74.9:0.1:25) at a flow rate of 2 mL min−1. The
amount of TC was quantified with calibration standards prepared
by dilution of the TC stock solution with 20% of ethanol in osmosed
water.
The activity of free and immobilized laccase was determined at
different pH with ABTS as substrate as previously reported [14]. One
activity unit (U) corresponds to the quantity of enzymes needed to
oxidize 1 ␮mol of ABTS per minute.
The immobilization yield was determined from the difference
in enzyme activity (U) of the laccase solution before and after the
immobilization step, taking into account the enzyme activity mea-
sured in the solution after the immobilization and the totality of
washing solutions (Eq. (2)). It was calculated with the values of the
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
4
M. de Cazes et al. / Catalysis Today xxx (2014) xxx–xxx
Fig. 3. Membrane characterization by scanning electron microscopy: ceramic support (a) and active membrane (b).
Table 1
tetracycline was removed from the solution after 24 h of reac-
tion. This last result is very interesting since the use of osmosed
water enables to match with targeted conditions because the pH of
wastewater effluents is supposed to be close to 7. In consequence,
all the further experiments were carried out with TC solutions pre-
pared in osmosed water.
Concerning the impact of enzyme concentration, it can be
observed in Table 2 that 78% of the tetracycline was degraded at
a laccase concentration of 0.1 g L−1 while only 30% was degraded
when the laccase concentration was ten times lower. The use of a
relatively high concentration of free laccase seems very efficient for
tetracycline degradation. However as it was explained above, the
use of free enzymes generates high process costs, especially con-
cerning enzymes loss. Therefore it is preferable to graft enzymes
on a carrier, even if the quantities are lower. The costs of process
will thereby be lower.
Kinetics constants for the degradation of ABTS and TC with free and immobilized
laccase.
Experiment
Substrate
ABTS
Vmax (␮mol min−1
)
Km (␮mol L−1
)
Free enzymes
Grafted enzymes
Free enzymes
0.13
0.58
0.60
0.67
127
28
282
57
Tetracycline
Grafted enzymes
3.2. Impact of the pH and enzyme concentration on tetracycline
degradation
As far as enzymatic activity toward a substrate is a complex
reaction which depends on many parameters like specific enzyme-
substrate interactions at a specific pH, the free laccase activity
toward tetracycline was studied at different pH conditions. Lac-
cases usually present their best degradation activity at low pHs
[14,18]. In order to determine the optimal degradation conditions,
we studied ABTS and TC oxidation with free laccase (0.1 g L−1) in a
range of pH from 3 to 7. Preliminary experiments with ABTS con-
firm the literature results. In fact the maximum of laccase activity
(6.4 U mg−1) was reached at pH 3 when experiments were carried
out with this substrate (results not shown). The results for TC deple-
tion at 24 h of reaction are presented in Table 2. We can observe that
with TC the best results were obtained when the pH of the solu-
tion was around neutrality (i.e. pH 6–7). This result differs from
experiments with ABTS and with the previous works of Plagemann
et al. [18] and Chea et al. [14] who reported that the same lac-
case from T. versicolor was more active at pH 4 for the degradation
of a dye and phenolic compounds, respectively. This result con-
firms that optimal pH value strongly depends on enzyme-substrate
interactions. Morevover, it is important to notice that when TC solu-
tions were prepared in osmosed water as solvent (pH = 6) 78% of
3.3. Comparison between free and immobilized laccase
Preliminary experiments were carried out in order to determine
the optimal pH for TC degradation with immobilized enzymes. For
these experiments carried out in a stirred tank reactor with crushed
EM, we worked under the same conditions and pHs described above
in Section 3.2 but at a lower enzyme concentration (0.01 g L−1). In
such case the optimum of degradation was reached at pH 6 (49%
of residual TC) whereas for pH 3 and 7 the residual TC was, respec-
tively, of 80% and 71%, these results show that the optimal pH for
both free and immobilized enzymes for TC degradation is quite
similar.
In order to compare the TC degradation by free and immobilized
enzymes in similar conditions (pH 6 and 25 ◦C) we used equiva-
lent quantities of free or immobilized enzymes with 100 mL of a
20 mg L−1 tetracycline solution (see Section 2.4). The results pre-
sented in Table 3 show that after 24 h of reaction, a 0.01 g L−1
enzymatic solution enables the elimination of 30% of the tetra-
cycline whereas the degradation yield can reach 56% with the
equivalent quantity of grafted enzymes. The difference of tetra-
cycline degradation obtained for the blanks shows that some
tetracycline could be auto degraded and/or adsorbed on the mem-
brane.
Table 2
Degradation of a 20 mg L−1 tetracycline solution by free laccase after 24 h of reaction
at different pH or enzyme concentrations.
Solvent
Residual
tetracycline (%)
pH valuea
3
4
5
6
7
6
0.01
0.05
0.1
Citrate-phosphate buffer
Citrate-phosphate buffer
Citrate-phosphate buffer
Citrate-phosphate buffer
Citrate-phosphate buffer
Osmosed water
Osmosed water
Osmosed water
Osmosed water
88
67
59
36
28
22
70
45
22
Table 3
TC degradation yield after 24 h of reaction with free and immobilized laccasea.
Experiment
Degradation yield (%)
Enzyme concentrationb
Free enzymes
Immobilized enzymes
Blank (free)
30 ( 7)
56 ( 8)
5 ( 5)
9 ( 5)
(g sigma product L−1
)
Blank (immobilized)
a
Enzyme concentration 0.1 g L−1
pH 6.
.
b
a
Enzyme concentration: 0.01 g L−1 in osmosed water at 25 ◦C.
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
M. de Cazes et al. / Catalysis Today xxx (2014) xxx–xxx
5
Fig. 4. Chromatograms corresponding to sample at 0 h and after the treatment (A), MS spectra (B) and MS/MS spectra (C) corresponding to OTC TP generated during enzymatic
treatment.
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
6
M. de Cazes et al. / Catalysis Today xxx (2014) xxx–xxx
Table 4
Literature results for tetracycline (TC) degradation with free enzymes.
Reference
Enzyme
Suda et al. [19]
Wen et al. [30]
Wen et al. [31]
Laccase (Trametes versicolor)
Lignin peroxidase
Manganese peroxidase
(Phanerochaete chrysosporium)
(Phanerochaete chrysosporium)
10−4 mol L−1 (≈6.4 g L−1
)
40 U L−1
pH 4.2
50
40 U L−1
pH 4.8
pH 4.5
0.2
TC (mg L−1
)
50
Extra-chemicals
Degradation yield
1-Hydroxybenzotriazole
99% in 1 h
H2O2
95% in 5 min
Mn(II) and H2O2
73% in 9 h
The degradation enhancement obtained with grafted enzymes
could be the result of the stability improvement reached with the
immobilization. Even though, concerning the beneficial effect of the
immobilization the results reported in the literature are sometimes
contradictory in terms of enzymes activity and stability. In some
cases, there are no conformational changes and the active sites are
still accessible thanks to the length of the covalent grafting bond.
The grafted biocatalyst can either keep the same activity as the free
enzyme by maintaining a good process efficiency [23] or its activity
can be improved compared to free enzymes as it has been shown
by other researchers [28,29]. In this work we can notice that the
hydrophilic character of the gelatin layer and its nature (protein)
could maintains a favorable environment around the enzymes and
probably plays a role on the activity enhancement observed.
If we compare our results with those reported in the litera-
ture (Table 4) we can notice that tetracycline degradation with free
enzymes is very efficient under some particular conditions. How-
ever in such previous works the enzyme concentrations used were
much higher than the concentrations used in this study. In addition
in all of these processes the TC degradation needs the use of sup-
plementary chemicals and additives which is a major drawback in
terms of cost mainly if we consider the treatment of wastewaters.
by-products of tetracycline degradation by ozonation [32] or by
sludge bacteria [33].
3.6. Degradation in the enzymatic membrane reactor
The enzymatic degradation in the membrane reactor was tested
in order to assess the preservation of enzymatic activity after sev-
eral cycles of newly prepared TC solution degradation. As explained
above, the reaction was run at 25 ◦C with 20 mg L−1 of tetracycline
in osmosed water with a total volume of 2 L in batch configuration.
This configuration was used in this work because we took the pre-
liminary assumption that the reaction takes place at the surface of
the membrane. As it was seen in Fig. 3, the gelatin only coated the
membrane surface as a thin layer but not the inner pores. Then it
can be reasonably concluded that enzymes were grafted only on
the surface of the gelatin layer. During the runs duration (10 days,
more than 200 h) the substrate solution was regularly replaced by a
fresh one. Samples were taken from permeate and retentate and the
tetracycline concentration was measured as previously. The per-
centage of adsorption within the whole pilot unit was measured;
it was less than 1%. The results are shown in Fig. 5.
We can notice that the reaction rate is almost constant and equal
to 0.34 ( 0.01) mg of tetracycline degraded per hour, even after 4
cycles with feed of fresh tetracycline solution. As it can be seen, the
enzymatic membrane is still active after 200 h of reaction. From
this result we can notice that probably the gelatin layer ensures a
high resistance to environmental conditions of the enzymes and
prevents them from being stressed by the flow. Thus the active site
of the biocatalyst can still react with the substrate without leach-
ing or conformational changes. It is important to notice that the
degradation rate reached in the membrane reactor is much higher
than the degradation rate in the stirred tank (Section 3.3) which
3.4. Effect of the temperature
The effect of the temperature on the TC degradation for free and
grafted laccase is shown in Table 5. Three temperatures were stud-
ied 15, 25 and 35 ◦C, higher temperatures were not considered as
relevant because the global goal of this study is to degrade TC in
wastewaters. We can notice that for all of the temperatures stud-
ied the activity of grafted laccase is always higher than the free
laccase activity. Moreover, for both enzyme forms the TC degrada-
tion increases from 15 ◦C to 25 ◦C and then remains stable up to
35 ◦C.
3.5. Study of transformation products (TPs)
According to the chromatograms presented in Fig. 4, two
different compounds were detected as major TPs after the
enzymatic treatment, the major TP was oxytetracycline (OTC)
(m/z = 461.1550 corresponding to [M + H]+) followed by anhy-
drotetracycline (m/z = 427.1501 corresponding to [M + H]+). These
two products have been previously reported in the literature as
Table 5
Degradation of a 20 ppm TC solution by laccase at different temperatures after 24 h
of reactiona.
Residual tetracycline (%)
T (◦C)
Free enzymes
Grafted enzymes
15
25
35
89
65
64
87
50
47
Fig. 5. Tetracyclinedegradationby immobilizedlaccase in the enzymaticmembrane
reactor.
Enzyme concentration: 0.01 g L−1 in osmosed water.
a
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051

G Model
CATTOD-8952; No. of Pages7
ARTICLE IN PRESS
M. de Cazes et al. / Catalysis Today xxx (2014) xxx–xxx
7
was only of 0.047 mg of tetracycline degraded per hour. We can
advance some hypothesis about this conversion enhancement. First
the ratio substrate amount/biocatalyst is much higher in the mem-
brane reactor; in fact the initial substrate concentration did not
change (20 mg L−1) but the enzyme concentration is much lower
(i.e. equivalent to 0.002 g L−1 instead of 0.01 g L−1) because of the
higher volume of reaction (i.e. 2 L instead of 0.1 L). It is well known
that reaction rates are lower under substrate limiting conditions.
Second, it is possible that the reactor configuration allowed a better
contact between the biocatalyst and the substrate because we are
continuously sweeping the solution on the membrane surface and
unlike the stirred tank where the flow-lines are not necessarily hav-
ing an effect on a possible boundary layer around the membrane
pieces where the enzymes are grafted.
a collaboration between Da Volterra (France), C-Lecta (Germany),
ChiralVision (The Netherlands), Institut Européen des Membranes
(France), Catalan Institute for Water Research (Spain) and Goethe
University (Germany). This study has been co-financed by the Euro-
pean Union through the European Regional Development Fund
(FEDER). This work was partly supported by the Generalitat de
Catalunya (Consolidated Research Group: Water and Soil Quality
Unit 2009-SGR-965). Prof. Barcelo acknowledges King Saud Uni-
versity for his visiting professorship.
References
[1] T. Deblonde, C. Cossu-Leguille, P. Hartemann, Int. J. Hyg. Environ. Health 214
(2011) 442–448.
[2] T. Heberer, Toxicol. Lett. 131 (2002) 5–17.
[3] K. Kummerer, Chemosphere 75 (2009) 417–434.
[4] S. Flint, T. Markle, S. Thompson, E. Wallace, J. Environ. Manage. 104 (2012)
19–34.
4. Conclusions
[5] F. Liu, G.G. Ying, R. Tao, J.-L. Zhao, J.F. Yang, L.F. Zhao, Environ. Pollut. 157 (2009)
1636–1642.
Laccase from T. versicolor was immobilized on ceramic mem-
branes and successfully used for tetracycline degradation under
the targeted working conditions (osmosed water at 25 ◦C) without
the presence of any extra chemicals. The grafting process allowed
avoiding enzyme leaching by the covalent bond created between
the biocatalyst and the carrier. Among the parameters studied for
the tetracycline degradation the pH resulted to be one of the most
important. We demonstrated that on the contrary of other sub-
strates degradation by laccase from T. versicolor which are depleted
in acidic conditions, the optimum the tetracycline degradation was
reached at a pH of 6, this is a very important result because this pH
approaches wastewaters pH. Moreover, the enzymatic concentra-
tion has also an impact on the degradation rate but it is limited by
the immobilization yield in the enzymatic membrane reactor.
Covalently grafted enzymes proved to be more efficient for
tetracycline degradation than free enzymes. This stability enhance-
ment is partly due to the protecting effect of the gelatin. The
covalent bonds also prevent from conformational changes in the
protein structure, making probably the active site always accessi-
ble.
[6] B. Quinn, F. Gagne, C. Blaise, Sci. Total Environ. 389 (2008) 306–314.
[7] F. Baquero, J.L. Martinez, R. Canton, Curr. Opin. Biotechnol. 19 (2008) 260–265.
[8] P. Gao, Y.J. Ding, H. Li, I. Xagoraraki, Chemosphere 88 (2012) 17–24.
[9] V. Homem, L. Santos, J. Environ. Manage. 92 (2011) 2304–2347.
[10] R. Molinari, F. Pirillo, V. Loddo, L. Palmisano, Catal. Today 118 (2006) 205–213.
[11] C.P. Silva, M. Otero, V. Esteves, Environ. Pollut. 165 (2012) 38–58.
[12] I. Sirés, E. Brillas, Environ. Int. 40 (2012) 212–229.
[13] P. Demarche, C. Junghanns, R.R. Nair, S.N. Agathos, Biotechnol. Adv. 30 (2012)
933–953.
[14] V. Chea, D. Paolucci-Jeanjean, M.P. Belleville, J. Sanchez, Catal. Today 193 (2012)
49–56.
[15] S.S. Desai, C. Nityanand, J. Biotechnol. 3 (2011) 98–124.
[16] N. Duran, M.A. Rosa, A. D’Annibale, L. Gianfreda, Enzyme Microb. Technol. 31
(2002) 907–931.
[17] T. Kudanga, G.S. Nyanhongo, G.M. Guebitz, S. Burton, Enzyme Microb. Technol.
48 (2011) 195–208.
[18] R. Plagemann, L. Jonas, U. Kragl, Appl. Microbiol. Biotechnol. 90 (2011) 313–320.
[19] T. Suda, T. Hata, S. Kawai, H. Okamura, T. Nishida, Bioresour. Technol. 103 (2012)
498–501.
[20] P. Jochems, Y. Satyawali, S. Van Roy, W. Doyen, L. Diels, W. Dejonghe, Enzyme
Microb. Technol. 49 (2011) 580–588.
[21] K.P. Katuri, S.V. Mohan, S. Sridhar, B.R. Pati, P.N. Sarma, Water Res. 43 (2009)
3647–3658.
[22] R. Mazzei, L. Giorno, E. Piacentini, S. Mazzuca, E. Drioli, J. Membr. Sci. 339 (2009)
215–223.
[23] G.M. Rios, M.P. Belleville, D. Paolucci, J. Sanchez, J. Membr. Sci. 242 (2004)
189–196.
[24] J. De Gunzburg, C. Bensoussan, Methods for the Inactivation of Antibiotics,
France, 2012, European Patent Application Number EP20100305786.
[25] A. Pena, M. Paulo, L.J.G. Silva, M. Seifrtova, C.M. Lino, P. Solich, Anal. Bioanal.
Chem. 396 (2010) 2929–2936.
[26] M.P. Belleville, P. Lozano, J.L. Iborra, G.M. Rios, Sep. Purif. Technol. 25 (2001)
229–233.
[27] M. Gros, S. Rodríguez-Mozaz, D. Barceló, J. Chromatogr. A 1248 (2012) 104–121.
[28] M. Catapane, C. Nicolucci, C. Menale, L. Mita, S. Rossi, D.G. Mita, N. Diano, J.
Hazard. Mater. 248 (2013) 337–346.
The identification of the degradation products shows that
laccase allows the transformation of tetracycline mainly on oxy-
tetracycline and anhydrotetracycline products which have been
previously reported in the literature as by-products of tetracycline
degradation by ozonation or by sludge bacteria.
The results obtained show that active membranes can be used
for several cycles of tetracycline degradation and during 10 days
without enzymatic activity loss.
[29] C. Nicolucci, S. Rossi, C. Menale, T. Godjevargova, Y. Ivanov, M. Bianco, L. Mita,
U. Bencivenga, D.G. Mita, N. Diano, Biodegradation 22 (2011) 673–683.
[30] X. Wen, Y. Jia, J. Li, Chemosphere 75 (2009) 1003–1007.
[31] X. Wen, Y. Jia, J. Li, J. Hazard. Mater. 177 (2010) 924–928.
[32] M.H. Khan, H. Bae, J.Y. Jung, J. Hazard. Mater. 181 (2010) 659–665.
[33] B. Halling-Sorensen, G. Sengelov, J. Tjornelund, Arch. Environ. Contam. Toxicol.
42 (2002) 263–271.
Acknowledgments
The research leading to these results has received funding from
the European Union Seventh Framework Programme (FP7/2007-
2013) under ENDETECH grant agreement no. 282818. This work is
Please cite this article in press as: M. de Cazes, et al., Design and optimization of an enzymatic membrane reactor for tetracycline
degradation, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.051