Employment of immobilised lipase from Candida rugosa for the bioremediation of waters polluted by dimethylphthalate, as a model of endocrine disruptors

Article abstract of DOI:10.1016/j.molcatb.2009.09.016

A planar bioreactor, equipped with a polypropylene membrane on which a lipase was immobilised, has been employed in a bioremediation process involving water polluted by dimethylphthalate (DMP), a model for a class of endocrine disruptors. The dependence of enzyme activity on pH, temperature and DMP concentration has been characterised under isothermal conditions, whereas the kinetics parameters have been studied under non-isothermal conditions. The following sequence was found for the values of lipase affinity, K_m, towards the DMP: K_m^free < K_{m, non-isoth}^imm < K_{m, isoth}^imm. A comparison of the results obtained under isothermal and non-isothermal conditions indicated that there was an advantage in using non-isothermal bioreactors in the environmental field. These advantages in particular resulted in: (i) an increase in the enzyme activity proportional to the applied transmembrane temperature difference and (ii) a reduction in the bioremediation times and, consequently, the process costs. The advantages in using bioremediation processes in the place of classical membrane processes, such as ultrafiltration or reverse osmosis, are also discussed.

Full text of DOI:10.1016/j.molcatb.2009.09.016

Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Employment of immobilised lipase from Candida rugosa for the bioremediation
of waters polluted by dimethylphthalate, as a model of endocrine disruptors
Luigi Mitaa,b, Vincenzo Sica , Marco Guida , Carla Nicolucci , Tiziana Grimaldi , Lucia Caputo ,
c
d
a
a
e
a,c
e
e
f
Mariangela Bianco , Sergio Rossi , Umberto Bencivenga , Mohamed S. Mohy Eldin ,
Maria Antonietta Tufano , Damiano G. Mita
b
a,b,e,∗
a,b
, Nadia Diano
a
National Institute of Biosystems and Biostructures (INBB), Viale Medaglie d’Oro, 305, 00136 Rome, Italy
b
c
d
e
f
Department of Experimental Medicine, Second University of Naples, Via S. M. di Costantinopoli, 16, 80138 Naples, Italy
Department of General Pathology, Second University of Naples, Via L. De Crecchio, 7, 80138 Naples, Italy
Department of Biological Sciences, University of Naples “Federico II”, Via Mezzocannone, 8, 80134 Naples, Italy
Institute of Genetics and Biophysics “ABT”, Via P. Castellino, 111, 80131 Naples, Italy
Institute of Advanced Technology and New Materials (IATNM), Mubarak City for Scientific Research and Technology Applications (MuCSAT),
New Boarg El-Arab City 21934, Alexandria, Egypt
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 24 April 2009
Received in revised form
A planar bioreactor, equipped with a polypropylene membrane on which a lipase was immobilised,
has been employed in a bioremediation process involving water polluted by dimethylphthalate (DMP), a
model for a class of endocrine disruptors. The dependence of enzyme activity on pH, temperature and DMP
concentration has been characterised under isothermal conditions, whereas the kinetics parameters have
been studied under non-isothermal conditions. The following sequence was found for the values of lipase
1
6 September 2009
Accepted 24 September 2009
Available online 12 October 2009
free
imm
imm
affinity, Km, towards the DMP: Km < K
< K
. A comparison of the results obtained under
m,non-isoth
m,isoth
isothermal and non-isothermal conditions indicated that there was an advantage in using non-isothermal
bioreactors in the environmental field. These advantages in particular resulted in: (i) an increase in the
enzyme activity proportional to the applied transmembrane temperature difference and (ii) a reduction
in the bioremediation times and, consequently, the process costs. The advantages in using bioremediation
processes in the place of classical membrane processes, such as ultrafiltration or reverse osmosis, are also
discussed.
Keywords:
Bioremediation
Phthalates
Endocrine disruptors
Non-isothermal bioreactors
Lipase
©
2009 Elsevier B.V. All rights reserved.
1
. Introduction
receptors and consequently the neutralisation of their activity; and
iii) interference with the synthesis, transport, metabolism and
(
Endocrine disrupting chemicals (EDCs) are a heterogeneous
secretion of natural hormones, altering their physiological concen-
trations and therefore their corresponding endocrine functions.
There are two main sources of evidence favouring these
hypotheses. The first is the extensive experimental evidence show-
ing that many xenobiotic compounds have intrinsic hormonal
activity. Laboratory animal studies indicate that a wide variety
of hormone-dependent physiological effects are associated with
exposure to these environmental pollutants. Furthermore, in vitro
molecular studies have demonstrated that these pollutants not only
have affinity for hormone receptors but also are able to induce, via
the receptor, agonistic or antagonistic effects. The second piece of
evidence derives from incidents of human and wildlife acute expo-
sure to hormonally active xenobiotics, which result in impaired
reproductive capability in the exposed individuals and in their off-
spring. Cancer has also been implicated.
group of chemicals that interfere with the endocrine system
and cause adverse alterations in reproductive and development
processes while also causing metabolic diseases. Depending on
their action on female or male hormones, they are defined as
anti-estrogens or anti-androgens. Endocrine disruptors include
polychlorinated biphenyls (PCBs), dioxins, some pesticides,
alkylphenols, phthalates, polybrominated flame retardants, phy-
toestrogens, and some heavy metals. The main mechanisms
through which these substances interfere with the endocrine
system are: (i) simulation of the activities of physiological hor-
mones, thereby participating in the same reactions and causing the
same effects; (ii) inactivation, with competitive action, of hormone
Because of the broad spectrum of diseases correlated with expo-
sure to EDCs, affecting not only the quality of wild and human life
but also that of their offspring, many research activities have tried
∗ Corresponding author at: Institute of Genetics and Biophysic “ABT” of CNR, Via
P. Castellino, 111, 80131 Naples, Italy. Tel.: +39 081 6132608; fax: +39 081 6132608.
E-mail address: mita@igb.cnr.it (D.G. Mita).
1
381-1177/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2009.09.016

1
34
L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
to address these problems in the last years. The main research
fields included: (a) bioremediation of polluted soils and waters;
DMP hydrolysis by lipase may involve both methyl groups get-
ting phthalic acid (PA) and two molecules of methanol, or may cause
the rupture of a single bond thus producing monomethylphthalate
(MMP) and methanol. To ascertain the relevant mechanism for our
enzyme, we have carried out experiments using free lipase from C.
rugosa and MMP. It was found that our lipase did not catalyse this
substrate, at least in any detectable amount after 4 h of incubation.
Incidentally it is important to stress the circumstance that the MMP
does no exhibit the same toxicological properties of DMP, as found
by us with the MTT test, a rapid and sensitive method for screening
the assessment of cytotoxicity of materials.
Polypropylene (PP) membranes were purchased from GE
Osmonics (GE Labstore-Osmonics, Minnetonka, MN) and used as
supports for enzyme immobilisation. Their thickness was 150 ␮m
and the nominal pore diameter was 0.22 ␮m.
All the chemicals, including the enzyme, were purchased from
Sigma (Sigma Italia, Milan, Italy) and used without further purifi-
cation.
(
b) biodetermination using biosensors; (c) study of the molecu-
lar mechanisms by which these harmful substances interact with
the cellular machinery; and (d) establishment of epidemiological
correlations between exposure and the diseases.
Recently we have studied the bioremediation of waters polluted
by phenol compounds. In particular, we have studied the endocrine
disruptor Bisphenol A [1], its biodetermination in aqueous solution
by means of tyrosinase-based carbon paste electrodes [2], and the
changes in the proliferation and viability indices of MCF-7 cancer
cells exposed to BPA, untreated or enzyme treated [3,4].
In this paper, we have focused our attention to phthalates.
Phthalates are used in many consumer products including toys,
baby products, lotions, cosmetics, personal care products, fra-
grances, air fresheners, medical tubing components and devices,
blood bags, PVC pipes and flooring, and pharmaceuticals [5,6]. They
are ubiquitous in the environment. Several studies have shown
that although phthalate exposure levels in humans are generally
low, close to the detection limit, a small percentage of people are
exposed to higher levels of phthalates. This information is based on
the level of phthalate metabolites identified in the urine of some
pregnant women and in human amniotic fluid [7–13]. In rats, at
certain levels of exposure, phthalates cause liver cancer, sponta-
neous abortions, and reproductive tract malformations in male and
female offspring [14–19]. The adverse reproductive effects seen
in the male offspring, described as the “Phthalate Syndrome”, are
currently the focus of regulatory agencies because this syndrome
occurs at lower dosage levels than other toxicities.
2.2. Apparatus
The apparatus employed is represented in Fig. 1. Fig. 1a shows
the modus operandi and Fig. 1b shows an exploded view. The biore-
actor consists of two metallic flanges, in each of which is bored
a shallow cylindrical cavity, 70 mm in diameter and 2.5 mm in
depth, constituting the working volume that is filled with the aque-
ous buffer solutions containing DMP. The catalytic membrane is
clamped between the two flanges so as to separate and, at the same
time, connect the solutions filling the half-cells. Solutions are circu-
lated in each half-cell, by means of two peristaltic pumps, through
hydraulic circuits starting and ending in a common glass tube. Using
independent thermostats, the two half-cells are maintained at pre-
determined temperatures. Thermocouples, placed 1.5 mm away
from the membrane surfaces, measure the temperatures of the
solutions at that point in each half-cell. This setup allows the cal-
culation of the temperature profile across the catalytic membrane,
as reported in Section 3.3.1.
Taking account of these concerns, we decided to study the
bioremediation of water polluted by phthalates. While numer-
ous studies have demonstrated that microorganisms play a major
role in the degradation of phthalates in the environment [20–26],
only a few papers [27,28] reported the use of purified enzymes in
these biodegradation processes. This paper describes the degrada-
tion of dimethylphthalate by means of lipase from Candida rugosa.
The lipase has been immobilised on a planar polypropylene mem-
brane in a bioreactor operating under isothermal or non-isothermal
conditions. The advantages in using a bioremediation process in
place of classical membrane processes, such as reverse osmosis
or ultrafiltration, are discussed. Moreover the advantages of using
non-isothermal bioreactors, compared to isothermal bioreactors,
are analysed in terms of: (i) the percentage increase in catalytic
2.3. Methods
2.3.1. Catalytic membrane preparation
The catalytic membranes were prepared in two successive
steps: (a) activation by means of a plasmo-chemical reactor and
(b) enzyme immobilisation by means of a diazotation process.
◦
activity per 1 C of temperature difference across the catalytic
membrane and (ii) the consequent reduction of bioremediation
times.
2.3.1.1. Membrane activation. Polypropylene is a non-polar mate-
rial that lacks reactive groups for enzyme immobilisation.
Consequently, functional groups have been created on the
polypropylene membrane by means of a plasma reactor. Plasma
was powered by a mixture of acrylic acid (Sigma–Aldrich, 99%)
and He according to the ratio of 3:20 sccm (standard cubic cen-
timetres per minute). The experimental conditions (power = 80 W,
pressure = 400 mTorr, time = 10 min) gave rise to a very stable coat-
ing on the membrane, showing the following abundance of reactive
groups: COOH < C O < COH < CC.
2
. Experimental
2.1. Chemicals
To carry out our experiments dimethylphthalate (DMP) was
chosen as a model for the phthalate class. DMP is an oily liquid
that is slightly sweet. The chemical formula is C10H10O4 and the
molecular weight is 194.19 g/mole.
For the enzyme, we used lipase (EC 3.1.1.3) from C. rugosa
1170 U/mg). Lipases, like esterases, catalyse the hydrolysis and
2.3.1.2. Enzyme immobilisation. The lipase was immobilised on the
activated membrane through a diazotation process involving the
phenolic groups of tyrosine residues. This procedure was chosen
because the tyrosine residues are far from the catalytic site. To
generate aminoaryl derivatives on the plasma activated PP mem-
branes, the membranes were treated for 90 min with a 2% (w/v)
p-phenylenediamine (PDA) aqueous solution of 0.1 M sodium car-
bonate buffer, pH 9.0. Later, the membranes were washed with
double distilled water. The obtained aminoaryl derivatives were
(
transesterification of ester groups. However, while esterases act
on water soluble substrates, lipases catalyse reactions of insolu-
ble water substrates. The presence of a water/lipid mixture is an
essential prerequisite for an efficient catalysis reaction. The active
site of lipase from C. rugosa consists of three amino acid residues
Ser209, His449 and Glu341) that form a “catalytic triad” (as that of
serine proteases), which is responsible for the nucleophilic attack
that promotes ester bond cleavage.
(
◦
treated for 40 min at 0 C with an aqueous solution containing 4%

L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
135
Fig. 1. The bioreactor: (a) modus operandi of the bioreactor and (b) exploded view of the bioreactor.

1
36
L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
(
w/v) NaNO2 and 2 M HCl, in a ratio of 1:5. At the end of this treat-
ment the membranes were washed at room temperature in a buffer
◦
solution (0.1 M phosphate, pH 7.0), and then treated for 16 h at 4 C
with 30 mL of the same buffer solution containing 20 mg/mL of
extract from the container. At the end of this last step the mem-
branes were further washed with 0.1 M phosphate buffer, pH 7.0 to
remove material not bound.
The amount of immobilised lipase was calculated by subtract-
ing the amount of protein recovered in the solution at the end of
the immobilisation process (and in the washing solutions) from the
amount of protein initially used for the immobilisation. The protein
concentration was measured using the method of Lowry et al. [29].
Under the experimental conditions reported above, the amount
of immobilised protein on PP membranes was 3.26 ± 0.2 mg, i.e.,
0
.041 ± 0.002 mg cm−2
.
2.3.2. Catalytic activity determination
Lipase catalytic activity was determined by monitoring, over
time by means of a HPLC (LC-20 AT, Shimadzu, Kyoto), the dis-
appearance of DMP or the appearance of MMP. In particular, at
regular time intervals, 200 ␮L of solution were collected from the
reaction volume and processed by chromatographic analysis. Each
experiment lasted 3 h.
Fig. 2. Concentration of MMP (ꢀ), DMP (᭹) and DMP + MMP (ꢀ) as a function of
time.
◦
When not used, the membranes were stored at 4 C in 0.1 M
phosphate buffer, pH 7.0.
HPLC analyses were carried out with a Discovery HS C18 column
Membranes were stable for about two months. Membrane
stability was assayed by measuring (every day) the membrane
activity under standard conditions, i.e., 5 mM in 0.1 M phosphate
(
15 cm × 4.6 mm, 5 ␮m; Supelco, Bellefonte, USA). The samples
were eluted at a flow rate of 1.0 mL/min with 30% acetonitrile in
water (0–2 min), followed by a linear gradient from 30% to 50%
acetonitrile (2–4 min), and finally with 50% acetonitrile (4–10 min).
At the end of the chromatogram, the mobile phase returned to
its initial composition in 2 min and the column was equilibrated
for 10 min. The wavelength used for detection was 275 nm. The
chromatograms were processed with the software LC Solution (Shi-
madzu, Kyoto). The retention time for DMP occurred at 5.5 min,
while that for MMP occurred at 1.3 min. A decrease in the peak of
DMP and an increase in the peak for MMP were observed during
the enzyme reaction.
◦
buffer, pH 7.0 and T = 25 C. When the membrane activity exhib-
ited a value that was 8% lower than that of the initial value (i.e.,
a value twice the experimental error) the membrane was dis-
charged.
3. Results and discussion
The experimental results are grouped into three sections. The
first, addressed to highlight the activity changes induced by the
immobilisation process, will concern the study (under isothermal
conditions) of the dependence of lipase activity, soluble or insolu-
ble, on pH and temperature, and the determination of the kinetics
parameters. The second section will be focused on the effects of
non-isothermal conditions on the activity of immobilised lipase.
The third section will deal with the evaluation of some quantita-
tive parameters that are significant for industrial processes. These
parameters are: (a) the actual temperature difference across the
catalytic membrane and (b) the coefficients ˛, P.A.I. and ꢀr that will
be defined in the following.
Interestingly, after the first 10 min of the experiment, subse-
quent time points indicate that the sum of the moles of substrate
and reaction product give a constant value equal to the initial value
of DMP, as the stoichiometry of the reaction is 1:1 (Scheme 1). This
sum is lower by about 10% with respect to the initial value of the
substrate concentration. This difference is attributed to an initial
substrate adsorption on either the membrane or the tubing of the
hydraulic circuit. This lost percentage of DMP was constant in each
experiment performed, regardless of the initial DMP concentration
values. It must be noted that the bioreactor with the catalytic mem-
brane was washed after each run with the 0.1 M phosphate buffer,
pH 7.0.
All experiments were repeated five times, and the experimental
points in the figures represent the average value. The experimental
error never exceeded 4%.
The enzyme activity, expressed as ␮moles min−1, is given by the
slope of the lines that best fit the experimental points showing the
decrease in DMP moles or the increase in MMP moles. No signif-
icant differences were found in the two calculations. Just to give
an example for the followed methodology in Fig. 2 we report the
case of an experiment carried out with a 5 mM DMP initial solution.
The DMP concentration is converted in ␮moles by multiplying the
actual concentration for the treated solution volume.
3.1. Isothermal characterisation of soluble and insoluble lipase
In Fig. 3a, the relative activities of soluble or insoluble lipase are
reported as a function of pH. In the case of soluble lipase we carried
out the experiments holding the enzyme concentration (2 mg/mL),
◦
the temperature (25 C) and the concentration of substrate (5 mM)
Scheme 1.

L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
137
As seen in Fig. 3a, both enzyme forms have an optimum pH value
at 7.0. The only difference appears in the range in which the rela-
tive catalytic activity is higher than the 90% value defined by us
as the “optimum pH range.” This range is from 5.6 to 8.0 for the
immobilised enzyme, while the free enzyme has a range from 5.8
to 8.3.
The data in Fig. 3b show the dependence of the relative activity
of the free or immobilised lipase as a function of temperature. The
experimental conditions, either for the free lipase or for the insol-
uble lipase, were the same as those used for the pH dependence
studies, with the exception that this time the pH was kept constant
at 7.0, whereas the temperature was varied in the range from 25
◦
to 70 C. Fig. 3b shows that the optimum temperature for soluble
◦
lipase occurs at about 45 C, whereas the optimum temperature for
◦
immobilised lipase is shifted to about 55 C. Moreover, the immo-
bilised enzyme presents an optimum temperature range between
◦
3
5 and 60 C, which is more extensive than that of the free coun-
◦
terpart, which spans between 40 and 55 C. Also in this case the
“optimum temperature range” is the range in which the relative
activity is higher than 90%. It follows that the immobilisation pro-
cedure imparts more thermal stability to the enzyme, protecting it
from the environmental conditions.
In Fig. 3c, the activities of the soluble and insoluble lipase are
reported as a function of DMP concentration. The experimental con-
◦
ditions (pH 7.0 and temperature 25 C) were the same for the free
and immobilised lipase, with only the DMP concentration being
varied. The choice of the DMP range was determined by the limit
of DMP solubility in water and by the HPLC detection limit. Both
enzyme forms exhibit a Michaelis–Menten behaviour. From the
data in Fig. 3c, through a Lineweaver–Burk plot, one obtains the
kinetics parameters Km and Vmax. For the free lipase, the Km value
−
1
is 3.5 mM and the Vmax value is 0.063 ␮moles min
per mg of
enzyme. For the immobilised lipase, the values were 5.5 mM for
−
1
the Km and 0.031 ␮moles min per mg of immobilised enzyme for
the Vmax, respectively.
When an enzyme is immobilised, there is a change in the kinetic
parameters. This is due to the changes in protein conformation
induced by the enzyme–support interactions and by the method
of immobilisation, i.e., whether the amino acid residues involved
in the immobilisation process are near or far from the catalytic site.
Other causes are the differential distribution of substrate in the
microenvironment around the catalytic site and in the bulk solu-
tion (partitioning effect), and the restrictions to substrate diffusion
introduced by the chemicals involved in the immobilisation pro-
cess. Specifically, if we compare the value of Km obtained for the
immobilised enzyme (5.5 mM) with that of the free lipase (3.5 mM),
it is clear that the increase in Km value appears as a loss of affinity
for the substrate. Moreover, a comparison of the Vmax values shows
that the immobilised enzyme loses about 50% of its activity in the
free form.
3.2. Effects of non-isothermal conditions on the activity of
immobilised lipase
We now examine the behaviour of our catalytic membranes
in the presence of temperature gradients. The presence of two
independent thermostats, connected to the bioreactor, can keep
the two half-cells at the same or at different temperatures. The
subscripts “w” and “c” indicate the values of the warm and cold half-
cells, respectively. When Tw = Tc, the temperature difference ꢁT is
equal to zero and the experiments are carried out under isother-
mal conditions (ꢁT = 0). Alternatively, when Tw > Tc, ꢁT = Tw − Tc
is different from zero and the experiments are carried out under
non-isothermal conditions (ꢁT =/ 0). It must be remembered that
the temperatures are read at the positions of the thermocouples,
which are 1.5 mm away from the membrane surface. So, for exam-
Fig. 3. Dependence of lipase activity on pH, temperature and DMP concentration:
Relative initial enzyme activity as a function of pH (a) and temperature (b). Initial
enzyme activity as a function of DMP concentration (c). Symbols: (ꢀ) free enzyme;
(ꢁ) immobilised enzyme.
all constant while varying the pH in a range from 4.0 to 9.0. The
volume of substrate solution was 10 mL.
For the immobilised lipase, the experimental conditions were
the same as that for the free enzyme, with the exception of the
enzyme amount (3.26 ± 0.35 mg of enzyme) and the reaction vol-
ume (30 mL).

1
38
L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
additional thermodiffusive fluxes add to the diffusive ones [30].
This means that under non-isothermal conditions the immobilised
enzymes “see” in the microenvironment around the catalytic site a
substrate concentration that is higher than in the bulk solution. This
is a new kind of “partitioning effect” that is related to the presence
of the temperature gradient.
3.3. Evaluation of some quantitative parameters that are
significant for processes of industrial interest
3.3.1. Temperature profile in the bioreactor
To quantify the advantages of using non-isothermal bioreactors,
it is necessary to clarify that the catalytic membrane actually is not
subjected to the macroscopic temperature difference ꢁT detected
by the thermocouples, but instead is subjected to a lower temper-
ature difference ꢁT*. Indeed, we have demonstrated [30] that the
solution motion in the bioreactor is laminar and that, utilising the
principle of heat continuity, it is possible to write the following
*
relationship: ꢁT = kꢁT, where “k” is a numerical constant depend-
Fig. 4. Effect of ꢁT on the enzyme kinetics: initial enzyme activity as a function
ing on the thermal conductivities of both the solutions filling the
bioreactor and the membrane. In the present case, k is equal to
of DMP concentration. Symbols: (᭹) ꢁT = 0 C; (ꢁ) ꢁT = 10 ◦C; (ꢀ) ꢁT = 20 C; (ꢀ)
◦
◦
◦
ꢁ
T = 30 C.
0
.11. To obtain the k value we used the thermal conductivity of
water [31], for the solution filling the bioreactor, and the value
taken from Touloukian et al. [32] for the polypropylene membrane.
We are conscious of the fact that our calculation is approximate.
To give an example of the real situation, in Fig. 5 the temperature
profile in the bioreactor is depicted for the experimental condi-
◦
◦
ple, when Tw = 40 C and Tc = 10 C, the macroscopic experimental
◦
◦
conditions are identified as ꢁT = 30 C and Taverage = 25 C. To assess
the effects of the temperature gradients on the activity of the
immobilised lipase and, consequently, to assess the convenience
in using the technology of non-isothermal bioreactors, we have
carried out the experiments while maintaining a constant average
◦
◦
tions ꢁT = 30 C and Taverage = 25 C. From this figure it is possible to
◦
appreciate how a macroscopic ꢁT = 30 C across the bioreactor is
◦
◦
temperature (Taverage = 25 C), and by applying three temperature
differences: ꢁT = 10 or 20 or 30 C.
reduced to an effective ꢁT* = 3.2 C across the catalytic membrane.
◦
Fig. 4 shows the results obtained under non-isothermal con-
ditions by varying the DMP concentration from 1 to 15 mM.
For comparison, the data obtained under isothermal conditions
3.3.2. Evaluation of P.A.I. and ˛ coefficients
Let us consider again Fig. 4 where it is possible to observe that at
each DMP concentration the enzyme reaction rate increases with
an increase in the applied temperature gradient. To verify the type
of dependence, it is appropriate to plot, at each substrate concen-
tration, the enzyme reaction rate as a function of the applied ꢁT.
This has been done, as one example, in Fig. 6 for the case of 5 mM
DMP. From this figure, it is possible to observe that the lipase cat-
alytic activity linearly increases with the applied ꢁT. The best line
(
(
ꢁT = 0) have also been added. Data in this figure indicate that:
i) the dependence of the reaction rate on the substrate concentra-
tion shows a behaviour described by a Michaelis–Menten equation
either under isothermal or non-isothermal conditions; (ii) for each
DMP concentration the enzyme reaction rate under non-isothermal
conditions is higher than the corresponding reaction rate under
isothermal conditions; and (iii) at each substrate concentration,
the enzyme reaction rate increases with the increase in the applied
ꢁT. From the curves in Fig. 4, we calculated the kinetic param-
eters reported in Table 1. These values show that: (i) the Km
values obtained when working under non-isothermal conditions
are lower than those obtained in the corresponding isothermal
condition, thus demonstrating that the non-isothermal conditions
increase the apparent affinity of immobilised lipase for DMP; and
(
ii) under non-isothermal conditions the Vmax values for the immo-
bilised lipase increase with an increase in the macroscopic value
of ꢁT, compared to those obtained under isothermal conditions,
−1
−1
and approach the value (0.063 ␮moles min mg of enzyme)
obtained for the free enzyme, thus proving the effectiveness of
non-isothermal reactors.
The above behaviour is explained by considering that, in the
presence of a temperature gradient, the immobilised enzymes
in the unit of time “encounter” more substrate molecules, since
Table 1
Kinetic parameters of immobilised lipase.
◦
◦
−1
−1
Tav ( C)
ꢁT ( C)
Km (mM)
Vmax (␮moles min mg of enzyme)
2
5
0
5.5
5.1
4.7
4.2
0.031
0.033
0.035
0.037
1
2
3
0
0
0
Fig. 5. Actual temperature profile into the bioreactor: temperature profile in a
◦
non-isothermal bioreactor under the following experimental conditions: ꢁT = 30 C,
◦
Tav = 25 C.

L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
139
membrane is enough to obtain increases of over 20% in the removal
of DMP concentration.
3.3.3. Evaluation of the reduction in bioremediation times under
non-isothermal conditions
From the industrial point of view, the results discussed above
indicate a substantial reduction in the processing times to bioreme-
diate water polluted by DMP and thus a reduction in the process’s
costs. In fact, it is possible to correlate the parameters ˛*or P.A.I.,
which are functions of ꢁT*, with the reduction in the bioremedia-
tion time, ꢀr, defined as
ꢀiso − ꢀ
non-iso
ꢀr(%) =
× 100
ꢀ
iso
where ꢀiso and ꢀnon-iso are the time required to obtain the same
percentage of DMP biodegradation under isothermal and non-
isothermal conditions, respectively. To correlate ꢀr with the applied
ꢁT* it is necessary to calculate the time required to obtain the same
amount of DMP removal under isothermal and non-isothermal con-
ditions. This calculation can be done graphically or analytically.
As one example of the graphical calculation let us see Fig. 8,
where we have reported, as a function of the time of enzyme treat-
Fig. 6. Dependence of the enzyme activity on the applied ꢁT: enzyme activity as a
function of the temperature gradient ꢁT. DMP concentration = 5 mM.
◦
ment, the DMP decrease in the case of Taverage = 25 C, with ꢁT = 0
interpolating the experimental points is described by an equation
of the type:
ꢀ
ꢁ
˛
yꢁT =/ 0(Tav) = yꢁT=0(Tav) 1 + 1 ꢁT
00
where yꢁ
(Tav) and y
(Tav) are the catalytic activity values
T =/ 0
ꢁT=0
measured under non-isothermal (ꢁT =/ 0) or isothermal (ꢁT = 0)
conditions, respectively, at a fixed value of Taverage.
From the above equation one obtains
yꢁ
(Tav) − y
ꢁ
(Tav) 100
P.A.I.
T =/ 0
ꢁT=0
(Tav)
T=0
˛
=
=
ꢁT ꢁT
y
◦
−1
where the ˛ coefficient (% C ) represents the percentage activ-
ity increase (P.A.I.), when a macroscopic temperature difference
◦
ꢁ
T = 1 C is read at the thermocouple positions.
Following the procedure described above, the values of P.A.I.,
calculated for each substrate concentration of Fig. 4, have been
reported in Fig. 7a as a function of DMP concentration. Fig. 7a
shows that the P.A.I. values decrease with the increase in sub-
strate concentration. This is explained by considering that when
the immobilised enzyme works at substrate concentrations close
to saturation, the addition of further substrate fluxes driven by the
temperature gradients is less effective in increasing the enzyme
activity. Alternately, when the enzymes work at low concentra-
tions, far from the saturation value, any additional substrate flux
effectively increases the activity.
◦
−1
Another interesting parameter for the process is ˛* (% C ),
the percentage activity increase of the enzyme reaction rate for
◦ ◦
T* = 1 C, i.e., when an actual temperature difference of 1 C is
ꢁ
applied across the catalytic membrane.
The expression for ˛* is:
P.A.I.
T
∗
˛
=
ꢁ
∗
Fig. 7b shows, as a function of DMP concentration, the ˛* val-
ues obtained from the results in Fig. 7a. Because this process is
similar to a normalisation process, overlapping curves are present.
It is interesting to observe that at low concentrations of DMP
(
1 mM) the ˛* value amounts to 24%, while at high DMP concen-
trations (10–15 mM) the ˛* value amounts to 6%. From the applied
point of view, these results indicate that at concentrations of DMP
lower than those used by us, such as those actually existing in the
Fig. 7. Significant parameters of industrial interest: (a) percentage activity increase
as a function of DMP concentration. (b) ␣* coefficient as a function of DMP concen-
tration. Symbols: (ꢁ) ꢁT = 10 C; (ꢀ) ꢁT = 20 C; (ꢀ) ꢁT = 30 C.
◦
◦
◦
◦
environment, a temperature difference of 1 C across the catalytic

1
40
L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
Fig. 8. Methodology for the calculation of the reduction times: DMP concentration
◦
◦
as a function of the enzyme reaction time. Symbols: (᭹) ꢁT = 0 C; (ꢁ) ꢁT = 10 C;
◦
◦
(
ꢀ) ꢁT = 20 C; (ꢀ) ꢁT = 30 C.
◦
or 10 or 20 or 30 C. The initial DMP concentration was 5 mM. To
obtain the same biodegradation of DMP, for example, a 50% reduc-
tion, 161 min are needed for the isothermal condition, while 145 or
◦
◦
◦
1
34 or 125 min are required for ꢁT = 10 C, ꢁT = 20 C, ꢁT = 30 C,
respectively. It follows that a value of ꢀr = 9.9% is obtained with
◦
◦
a ꢁT = 10 C, a ꢀr = 16.8% with a ꢁT = 20 C, and a ꢀr = 25% with a
ꢁ
ꢁ
◦
T = 30 C. The ꢀr values increase with an increase in the applied
T and therefore with the P.A.I.
The analytical approach is based on the considera-
tion that the same DMP degradation is obtained when
RRC,ꢁT=0ꢀiso = RR ꢀnon-iso, where RR stands for the reaction
C,ꢁT =/ 0
rate. By recalling that
ꢀ
ꢁ
˛
RR
ꢀiso = RRC,ꢁT=0 1 + 1 ꢁT ꢀnon-iso
C,ꢁT=0
00
one obtains after a series of mathematical steps
Fig. 9. Values of the bioremediation reduction times: percentage reduction of the
bioremediation times (ꢀr) as a function of percentage activity increase (a) and as a
function of DMP concentration (b).
ꢀ
ꢁ
ꢀ
ꢁ
∗
∗
˛
ꢁT
˛ ꢁT
∗
ꢀ
r (%) =
100 =
100
∗
˛
ꢁT + 100
P.A.I.
00 + P.A.I.
˛ ꢁT + 100
ꢀ
ꢁ
=
100
centage decreases of DMP concentration after 180 min of enzyme
1
treatment, calculated as R = (C
− C
)/C
.
t=0
t=180
t=0
C
t=0
and C
are the DMP concentrations at the beginning of
t=180
In Fig. 9a, the ꢀr values obtained at different DMP concentra-
the experiment and after 180 min of lipase treatment, respectively.
There are other considerations with respect to the usefulness of
employing a bioremediation process in place of a classical mem-
brane remediation process. For water treatment problems in small
ecosystems, the traditional membrane-based processes are not
useful because they alter the local life conditions. Ultrafiltration or
reverse osmosis, for example, can remove phthalates but, because
the filtrate is pure water, their intake in the ecosystem alters the
concentrations of salts and bioelements necessary for life. As a con-
sequence, the ecosystem is destroyed. In contrast, the selective
◦
tions with ꢁT = 30 C have been reported as a function of the P.A.I.
calculated for each DMP concentration. As expected, the reduction
in bioremediation time is an increasing function of the percentage
increase in the enzyme activity (P.A.I.) and, consequently, of the
temperature difference applied across the membrane. Highlighted
in black is the case for a DMP concentration equal to 5 mM, for
which the P.A.I. is 33.4% and the ꢀr is 25%.
Because the P.A.I. is related to the substrate concentration, we
have reported the reduction in biodegradation time as a function of
the DMP concentration in Fig. 9b. Again, highlighted in black is the
result relative to the DMP concentration of 5 mM. As is evident from
Fig. 9b, the reduction in the biodegradation time decreases with an
increase in the DMP concentration. Also, this result is interesting
for practical applications, because the concentrations used by us
are higher than those actually found in polluted water, owing to
DMP’s small solubility in water.
Table 2
The percentage decreases of DMP concentration after 180 min of enzyme treatment
under different experimental conditions.
Tav (◦C)
ꢁT (◦C)
[DMP] =
1 mM
[DMP] =
3 mM
[DMP] =
5 mM
[DMP] =
8 mM
From the above results, it follows that the decrease in DMP con-
centration is a linear function of the applied temperature difference
and is inversely proportional to the initial DMP concentration. To
quantise this observation, in Table 2 we have reported the per-
25
0
10
0
0
68.5%
84%
100%
100%
63%
58.1%
66.1%
72.9%
79.3%
43.4%
47.7%
51.9%
55.1%
72.5%
82.7%
89.5%
2
3

L. Mita et al. / Journal of Molecular Catalysis B: Enzymatic 62 (2010) 133–141
141
removal of phthalates by means of enzyme treatment (bioreme-
diation) appears suitable, because the treatment is effective only
towards the harmful molecule target, leaving unaltered the con-
centrations of other components.
[7] N. Pant, M. Shukla, D.K. Patel, Y. Shukla, N. Mathur, Y.K. Gupta, D.K. Saxena,
Toxicol. Appl. Pharm. 231 (2008) 112–116.
[
8] J.J. Adibi, R.M. Whyatt, P.L. Williams, A.M. Calafat, D. Camann, R. Herrick, H.
Nelson, H.K. Bhat, F.P. Perera, M.J. Silva, R. Hauser, Environ. Health Perspect.
116 (2008) 467–473.
[
9] G. Latini, C. De Felice, G. Presta, A. Del Vecchio, I. Paris, F. Ruggieri, P. Mazzeo,
Biol. Neonate 83 (2003) 22–24.
4
. Conclusion
[
[
[
10] G. Latini, A. Del Vecchio, M. Massaro, A. Verrotti, C. De Felice, Curr. Med. Chem.
3 (2006) 2527–2534.
11] M. Matsumoto, M. Hirata-Koizumi, M. Ema, Regul. Toxicol. Pharm. 50 (2008)
7–49.
1
The experimental results reported above have shown the possi-
3
bility of using the enzyme lipase from C. rugosa in the pathway for
the biodegradation of phthalates to bioremediate water polluted
by these compounds. The use of non-isothermal bioreactors proved
the utility of this technology in solving some of the pollution prob-
lems affecting human life and wildlife. Moreover, our studies may
increase the limited knowledge regarding the direct exploitation of
purified enzymes in the hydrolysis of phthalates.
12] M.J. Silva, J.A. Reidy, A.R. Herbert, J.L. Preau Jr., L.L. Needham, A.M. Calafat, Bull.
Environ. Contam. Toxicol. 72 (2004) 1226–1231.
[13] M.S. Wolff, S.L. Teitelbaum, G. Windham, S.M. Pinney, J.A. Britton, C. Chelimo, J.
Godbold, F. Biro, L.H. Kushi, C.M. Pfeiffer, A.M. Calafat, Environ. Health Perspect.
115 (2007) 116–121.
[14] P.R. Dalsenter, G.M. Santana, S.W. Grande, A.J. Andrade, S.L. Araujo, Hum. Exp.
Toxicol. 25 (2006) 297–303.
[
15] B.J. Davis, R. Weaver, L.J. Gaines, J.J. Heindel, Toxicol. Appl. Pharm. 128 (1994)
24–228.
16] M. Ema, R. Kurosaka, H. Amano, Y. Ogawa, Reprod. Toxicol. 8 (1994) 231–236.
2
[
Acknowledgments
[17] L.E. Gray Jr., J. Laskey, J. Ostby, Toxicol. Sci. 93 (2006) 189–195.
[
18] L.E. Gray Jr., J. Ostby, J. Furr, M. Price, D.N. Veeramachaneni, L. Parks, Toxicol.
Sci. 58 (2000) 350–365.
This work was partially supported by the “Regione Campania”
[
19] R.L. Melnick, Int. J. Occup. Environ. Health 8 (2002) 284–296.
(
Law 5 – Annuality 2005) and by the ISPESL. Moreover, this activ-
[20] D.W. Liang, T. Zhang, H. Fang, J. He, Appl. Microbiol. Biotechnol. 80 (2008)
83–198.
1
ity was sponsored by the Italian Ministry for Foreign Affairs in the
frame of a collaboration project between Italy and Egypt.
[
[
[
21] S.-S. Hwang, H.T. Choi, H.-G. Song, J. Microbiol. Biotechnol. 18 (2008) 770–775.
22] W.L. Chao, C.M. Lin, I.I. Schiung, Y.L. Kuo, Chemosphere 63 (2006) 1377–1383.
23] R. Kleerebezem, L. Hulshoff, G. Lettinga, Appl. Environ. Microbiol. 65 (1999)
1
152–1160.
References
[
24] N. Scholz, R. Diefenbach, I. Rademacher, D. Linnemann, Bull. Environ. Contam.
Toxicol. 58 (1997) 527–534.
25] J. Wang, P. Liu, H. Shi, Y. Qian, Chemosphere 36 (1997) 527–534.
26] K. Ziogou, P. Kirk, J.N. Lester, Water Res. 23 (1989) 743–748.
27] J.Y. Ahn, Y.H. Kim, J. Min, J. Lee, Curr. Microbiol. 52 (2006) 340–344.
28] C.M.L. Carvalho, M.R. Aires-Barros, J.M.S. Cabral, Biotechnol. Bioeng. 66 (1999)
[
1] N. Diano, V. Grano, L. Fraconte, P. Caputo, A. Ricupito, A. Attanasio, M. Bianco,
U. Bencivenga, S. Rossi, I. Manco, L. Mita, G. Del Pozzo, D.G. Mita, Appl. Catal. B:
Environ. 69 (2007) 252–261.
[
[
[
[
[
[
[
2] D.G. Mita, A. Attanasio, F. Arduini, N. Diano, V. Grano, U. Bencivenga, S. Rossi,
A. Amine, D. Moscone, Biosens. Bioelectron. 23 (2007) 60–65.
3] A. Bolli, P. Galluzzo, P. Ascenzi, G. Del Pozzo, I. Manco, M.T. Vietri, L. Mita, L.
Altucci, D.G. Mita, M. Marino, IUBMB Life 60 (2008) 843–852.
4] A. Ricupito, G. Del Pozzo, N. Diano, V. Grano, M. Portaccio, M. Marino, A. Bolli,
P. Galluzzo, P. Bontempo, L. Mita, L. Altucci, D.G. Mita, Environ. Int. 35 (2009)
1
7–34.
29] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, J. Biol. Chem. 193 (1951)
65–275.
[
[
2
30] N. Diano, M.M. El-Masry, M. Portaccio, M. Santucci, A. De Maio, V. Grano, U.
Castagnolo, U. Bencivenga, F.S. Gaeta, D.G. Mita, J. Mol. Catal. B: Enzym. 11
2
1–26.
5] F. Alatriste-Mondragon, R. Iranpour, B.K. Ahring, Water Res. 37 (2003)
260–1269.
6] S.K. Marttinen, R.H. Kettunen, K.M. Sormunen, J.A. Rintala, Water Res. 37 (2003)
385–1393.
(
2000) 97–112.
[
[
[
[
31] D.R. Lide, Handbook of Chemistry and Physics, Press C Editor, Boka Raton, 1990.
32] Y.S. Touloukian, P.E. Liley, S.C. Saxena, in: Y.S. Touloukian (Ed.), Thermophysical
Properties of Matter, IFI, Plenum, New York, 1970.
1
1