Soluble and nanoporous silica gel-entrapped C. freundii methionine γ-lyase

Article abstract of DOI:10.1166/jnn.2018.14333

Methionine γ-lyase is a pyridoxal 5'-phosphate dependent tetramer that catalyzes the α,γ-elimination of methionine in ammonia, methanethiol and γ-ketobutyrate. MGL catalytic power has been exploited as a therapeutic strategy to reduce the viability of cancer cells or bacteria. In order to obtain a stable enzyme to be delivered at the site of action, MGL can be encapsulated in a variety of matrices. As a reference encapsulation strategy we have prepared MGL nanoporous wet silica gels. Immobilized MGL gels were characterized with regards to activity, stability, absorption, circular dichroism and fluorescence properties and compared with soluble MGL. We found that MGL gels exhibit (i) spectroscopic properties very similar to MGL in solution, (ii) a higher stability with respect to the soluble enzyme and (iii) catalytic activity six-fold lower than in solution. These findings prove that MGL encapsulation is a suitable strategy for therapeutic applications.

Full text of DOI:10.1166/jnn.2018.14333

Article
Journal of
Nanoscience and Nanotechnology
Vol. 18, 2210–2219, 2018
Copyright © 2018 American Scientific Publishers
All rights reserved
Printed in the United States of America
www.aspbs.com/jnn
Soluble and Nanoporous Silica Gel-Entrapped
C. freundii Methionine ꢀ-Lyase
1
1
2
2ꢀ3
Elena A. Morozova , Vitalia V. Kulikova , Serena Faggiano , Samanta Raboni ,
3
3
1
1
Edi Gabellieri , Patrizia Cioni , Natalia V. Anufrieva , Svetlana V. Revtovich ,
Tatyana Demidkina , and Andrea Mozzarelli
Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Moscow, 119991, Russia
1ꢀ∗
2ꢀ3ꢀ4ꢀ∗
1
2
Department of Food and Drug, University of Parma, Parma, 43124, Italy
Institute of Biophysics, National Research Council, Pisa, 56124, Italy
National Institute of Biostructures and Biosystems, Rome, 00136, Italy
3
4
ꢀ
Methionine ꢀ-lyase is
a
pyridoxal 5 -phosphate dependent tetramer that catalyzes the
ꢁ
,ꢀ-elimination of methionine in ammonia, methanethiol and ꢁ-ketobutyrate. MGL catalytic power
has been exploited as a therapeutic strategy to reduce the viability of cancer cells or bacteria.
In order to obtain a stable enzyme to be delivered at the site of action, MGL can be encapsulated
in a variety of matrices. As a reference encapsulation strategy we have prepared MGL nanoporous
wet silica gels. Immobilized MGL gels were characterized with regards to activity, stability, absorp-
tion, circular dichroism and fluorescence properties and compared with soluble MGL. We found that
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
MGL gels exhibit (i) spect rCo so cp o yp r ii cg hp tr :o Ap emr t ei er si c va en r yS sc i ime ni l at i rf i ct o PMu Gb Ll isi hn e sr os lution, (ii) a higher stability
with respect to the soluble enzyme a nDd e(l ii i vi )e cr ae tda l by tyi cI na gc tei vn i tt ya six-fold lower than in solution. These
findings prove that MGL encapsulation is a suitable strategy for therapeutic applications.
Keywords: Methionine, Protein Drug, Protein Encapsulation, Pyridoxal Phosphate, Enzyme
Activity, Fluorescence.
1
. INTRODUCTION
at ∼330 nm, attributed to the enolimine tautomer of the
3
Methionine ꢀ-lyase (MGL) (EC 4.4.1.11) is a homote-
Schiff base. MGL external aldimines with some com-
trameric PLP-dependent enzyme that catalyzes the
petitive inhibitors absorb at ∼430 nm whereas quinonoid
ꢁ
,ꢀ-elimination of methionine in ammonia, methanethiol
and ꢁ-ketobutyrate. In addition, MGL catalyzes the
-replacement of L-methionine and its derivatives. As a
intermediates formed along the catalytic processes cat-
4
alyzed by PLP-dependent enzymes absorb at ∼500 nm.
ꢀ
The three-dimensional structures of MGL from Citrobac-
ter freundii as well as from other bacteria, including
secondary reaction, MGL catalyzes the ꢁ,ꢂ-elimination of
cysteine and S-substituted cysteines and the ꢂ-replacement
5
Clostridium sporogenes (PDB code 5DX5) and Pseu-
domonas putida (PDB code 2O7C), and from protozoa
such as Trichomonas vaginalis and Entamoeba histolytica
have been solved. MGL belongs to the fold type I of the
PLP-enzyme structural family. Its polypeptide chain with
molecular mass of around 43 kDa is composed of two
domains, a small domain and a large domain, with PLP
located between the two domains within the enzyme active
site. MGL is a dimer of dimer with each active site con-
taining residues of the interacting subunit. Tyr113 in the
active site is highly conserved through MGLs from sev-
1
ꢃ2
reaction of these amino acids. MGL has been isolated
in bacteria and protozoa. The PLP coenzyme is bound
as a Schiff base to the active site lysine residue forming
the internal aldimine (Scheme 1). In the presence of the
substrate methionine or other substrate analogues, MGL
forms the external aldimine that in a multi-step process
is converted to products. Citrobacter freundii MGL inter-
nal aldimine, in which PLP is bound to Lys210, exhibits
an absorption band at ∼420 nm, attributed to the ketoe-
namine tautomer of the Schiff base, and a weak band
∗
6–8
Authors to whom correspondence should be addressed.
eral species. Cys115 replacement changes the substrate
2210
J. Nanosci. Nanotechnol. 2018, Vol. 18, No. 3
1533-4880/2018/18/2210/010
doi:10.1166/jnn.2018.14333

Morozova et al.
Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
Scheme 1. Catalytic mechanism of MGL ꢁ,ꢀ-elimination reaction.
8
9
selectivity, as demonstrated in P. putida, T. vaginalis,
with consequent risk of protein denaturation during parti-
cle formation. The main drawback of using natural matri-
ces like alginate and chitosan is the possible leakage of
1
0
11
C. freundii, and Clostridium sporogenes MGLs.
MGL catalytic power has been exploited towards the
development of drugs to reduce I Pt h :e 5 .v 5i a3 b. 1i l 1i t 8y . 1o 3f 0e Oi t hn e: rSat, e1n 6t r Ja pu pn e 2d 0p 1r o8 t e1 i 6n : 5d u1 e: 4 t0o the fast biodegradability of these
1
2
Copyright: American S mc i ae tne tr iif ai cl s .P ublishers
cancer cells or bacteria. Regarding the application in
Delivered by Ingenta
A stable protein encapsulation strategy, used for the
characterization of protein function and dynamics, has
been achieved using organosilicate precursors of differ-
ent polarities, such as tetramethylortosilicate (TMOS) or
tetraethylortosilicate (TEOS). These compounds polymer-
ize forming a silicate network that individually entraps
proteins and is characterized by nanopores that allow
substrate and product in and out diffusion, while immo-
cancer, the rationale is that tumor cells require an abun-
1
3ꢃ14
dant supply of methionine for proliferation.
MGL
delivered in the proximity of cancer cells can lead to selec-
tive methionine degradation, thus to methionine starva-
tion, a condition that can also be partially obtained by a
1
5ꢃ16
rigid diet with methionine-depleted proteins.
A sim-
ilar approach is applied for drugs against acute lym-
phoblastic leukemia, which contain E. coli L-asparaginase,
2
4ꢃ25
bilizing the protein within the gel matrix.
Silica
1
7
able to deprive leukemic cells of circulating asparagine.
nanoparticles, given their high porosity and biocompat-
ibility, have attracted growing attention for biomedical
MGL can be also exploited against bacteria, since it cat-
alyzes the degradation of substrate analogs, such as alliin
and S-substituted L-cysteine sulfoxides to form allicin or
S-alkylthiosulfinates, that have been demonstrated to be
potent antibacterial agents. The concomitant delivery of
MGL and substrate analogs to the site of infection might
2
6ꢃ27
applications.
In 2011, a silica nanoparticle drug deliv-
ery platform for targeted molecular imaging of cancer
2
8
was approved by the US Food and Drug Administration.
A significant property of silica gels is that they are
optically transparent and several spectroscopies, includ-
ing UV-visible spectrophotometry, spectrofluorimetry and
1
8
represent a novel antibiotic strategy. For biopharmaceu-
tical applications against either cancer or bacterial infec-
tions, MGL needs to be properly formulated in order to
be protected from protease action and to increase stability,
thus increasing bioavailability. Towards this goal, protein
circular dichroism, can be used for the characterization
29–32
of the entrapped proteins.
Moreover, the resulting
nanoporous silica gels have been demonstrated to be a
powerful tool to stabilize proteins while preserving and
even enhancing their biological activities. This is achieved
by selecting distinct protein conformations. For example,
it was found that lipase gels exhibited about 1000 fold
higher activity than the soluble enzyme because the apolar
gel matrix stabilizes an open conformation of the enzyme
1
9–21
encapsulation represents a powerful strategy.
Proteins
are encapsulated in micro- or nanoparticles formed by
a variety of different polymeric matrices, such as poly-
lactic (PLA) and poly(lactic-co-glycolic) acid (PLGA),
2
2ꢃ23
alginates or chitosan.
Synthetic polymers, such as
3
3
PLA and PLGA, require organic solvents for dissolution,
more prone to catalysis. We previously encapsulated
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018
2211

Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
the PLP-dependent enzymes tryptophan synthase,34 tyro-
Morozova et al.
decrease of NADH absorption at 340 nm (ꢆꢇ =
3
5
−1
−1
ꢁ
sine phenol-lyase and tryptophan indole-lyase (TRPase),
6220 M cm ꢈ at 30 C. The reaction mixture contained
obtaining enzyme gels endowed with catalytic activity,
although about six-fold lower than the soluble enzyme.
Interestingly, encapsulation led to a shift of the tautomeric
equilibrium, stabilizing the enolimine species.
In the present work we have characterized and compared
soluble and silica gel-entrapped MGL from C. freundii
exploiting absorbance, fluorescence, circular dichroism,
chemical stability and activity.
100 mM potassium phosphate, pH 7.2, 0.1 mM pyridoxal
5 -phosphate, 5 mM DTT, 0.2 mM NADH, 70 U HO-
ꢀ
HxoDH, and 30 mM L-methionine. Protein activity in sil-
ica gels was measured exploiting the same reaction with
prior sonication of the gel matrix to avoid limitation to
3
5
substrate diffusion, as described previously. In brief, the
micro-fragments of the gels obtained by sonication were
about 4 ꢄm, as checked with a Zeiss microscope. The
fragment critical thickness over which rates are diffusion
4
1
controlled can be determined using the equation:
2
2
. MATERIALS AND METHODS
.1. Protein Preparation
MGL from C. freundii was expressed in E. coli
ꢀ
1/2
d = ꢉꢊK +ꢉS ꢋ∗D /k ꢉEꢋꢋ
c
M
0
cat
3
6
BL21(DE3) and purified as described previously. The
where kcat and K are the catalytic parameters obtained
M
in solution for C. freundii MGL, E is the enzyme concen-
ꢁ
protein was stored at −80 C in buffer 100 mM potassium
tration expressed in mM (0.07 mM), S is the substrate
phosphate, 1 mM dithiothreitol (DTT), pH 8.0.
0
ꢀ
concentration expressed in mM (30 mM), and D is the
diffusion coefficient of the substrate inside the gel. D can
ꢀ
2
.2. Encapsulation of MGL in TMOS and
TEOS Silica Gels
be calculated from the equation:
MGL was encapsulated in silica gels following the proto-
ꢀ
2
D /D = 1 −ꢊa /rꢈ
3
7–40
cols reported previously.
A solution containing tetram-
ethyl orthosilicate (TMOS), water and hydrochloric acid
was sonicated for 20 min at 4 C. An equal volume of
where a is the average molecular radius (4 Å for molecules
ꢁ
with molecular weights in the range 200–350 daltons),
1
0 mM potassium phosphate solution, pH 6.0, was added
−6
2
D (6×10 cm /sec) is the diffusion constant in water for
substances with this range of molecular weights, and r is
the average pore radius of the gel, 40–50 Å (a value of
to the sol and the mixture was deoxygenated for 40 min
ꢁ
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
at 4 C under flux of nitrogen. The sol was mixed in
1
Copyright: American Scientific Publishers
:1 volume ratio with 5.7 mg/ml MGL solution. Encap-
Delivered by Ingenta
5
0 Å is used). The calculated value of d is 150 ꢄm, a
c
sulation of MGL in TEOS was carried out following the
value significantly higher than the average gel thickness of
4
0
protocol reported by Brennan, with minor modifications.
In brief, 0.33 mL of 0.1 M HCl, 0.33 mL of H O and
about 4 ꢄm, indicating that diffusion of substrates within
the gel fragments does not limit the reaction rates.
The absence of protein leakage after sonication of the
gel was controlled by activity assays on the supernatant
after centrifugation of the gel microfragments. No activity
was detected, proving that sonication did not cause pro-
tein loss from the gel. The activity of MGL in solution
after application of the same sonication protocol did not
change. Since the addition of 5% PEG 1000 caused leak-
age of the protein in solution upon sonication, the activity
of immobilized MGL was calculated subtracting the activ-
ity of the supernatant to that of the microgel fragments
before centrifugation.
2
1
.5 mL of TEOS were mixed in ice until a homogeneous
solution was obtained. The resulting solution was stored
ꢁ
at −20 C. For gel preparation, 50 ꢄL of TEOS solution
were mixed with an equal volume of MGL in 100 mM
potassium phosphate, pH 7.2, and layered on quartz plates
(
2
8ꢅ4×15ꢅ3 mm). Usually protein concentration in gel was
.2 mg/mL. After gelification with TMOS and TEOS,
quartz plates were placed in 100 mM potassium phosphate,
ꢁ
pH 7.2, and stored at 4 C. The absence of protein leak-
age from gels was checked by activity assays on the gel
storage buffer. No protein was found to be released within
7
days after gel preparation.
TMOS silica gels were also prepared in the presence of
The kinetic parameters of MGL ꢁ,ꢀ-elimination in solu-
tion and in TMOS silica gels were determined using
methionine concentrations in the range between 0.05 mM
and 7 mM. The enzyme concentration was 0.15 ꢄM for
soluble MGL and 1.8 ꢄM for encapsulated MGL.
PEG 1000, PEG 8000 and sorbitol, known to be potential
protein stabilizers. These additives were mixed to the MGL
solution prior to the addition of the sol at 0.5 and 5% w/v
for PEGs and 0.25 M for sorbitol.
2
.4. Absorbance and Fluorescence Measurements
2
.3. Enzyme Activity
Absorption spectra were carried out by a double beam
Jasco V-550 UV/Vis spectrophotometer. Fluorescence
measurements were performed using a Fluoromax-4 spec-
trofluorometer (Horiba Jobin Yvon, Milan, Italy). Exci-
tation wavelength of 295 nm, 330 nm and 415 nm
MGL activity was determined using L-methionine as a
substrate by measuring the rate of ꢁ-ketobutyric acid
production in the coupled reaction with D-isocaproate
2
-hydroxy dehydrogenase (HO-HxoDH) monitoring the
2212
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018

Morozova et al.
Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
were used, with an excitation and emission slit width of
nm. All the spectra were corrected for instrumental
band at 330 nm, assigned to the enolimine tautomer of
the Schiff base of PLP, and a more prominent band cen-
tered at 420 nm, assigned to the ketoenamine form of PLP
(Fig. 1(A)). As a comparison the spectrum of the model
PLP-valine Schiff base is reported. In the apo form bands
associated to PLP absorbance disappeared as expected
5
response. To measure fluorescence emission from encapsu-
lated MGL, quartz plate with the protein-gel was placed in
a 10 ×10 mm fluorescence cuvette and held with its nor-
ꢁ
mal at 45 from the direction of the exciting radiation. The
ꢁ
front surface emission was collected at 90 from the exci-
(
Fig. 2(A)).
tation. Given both the measurement setup and the increase
in light scattering with respect to the solution, the changes
in shape of fluorescence spectrum were considered more
reliable than changes in fluorescence intensity. All mea-
surements were carried out in 100 mM potassium phos-
phate, pH 7.2, at 9 ꢄM MGL (monomer concentration) in
solution and 50–70 ꢄM (monomer concentration) in gel
The fluorescence emission spectra of MGL, obtained
upon excitation at 295, 330 and 415 nm, are reported in
Figures 1(B–D), respectively. The fluorescence spectrum
obtained by excitation at 295 nm exhibited an emission
maximum centered at 333 nm (Fig. 1(B)). This emis-
sion, that originates from the single Trp residue of MGL
per subunit, Trp252, suggests that this residue is located
in a partially apolar environment. Indeed, the inspection
of the three-dimensional structure of C. freundii MGL
shows that Trp252 is located at the interface between
monomers, in the center of the tetramer (Fig. 3(A)), sur-
rounded by apolar residues (Fig. 3(B)). In the apo form
of MGL the fluorescence intensity upon excitation at
(
estimated thickness about 1 mm).
2
.5. Circular Dichroism
Spectra were acquired on a Jasco J-715 spectropolarime-
ter in the 190–260 nm range. Spectra were recorded for
a solution containing 3 ꢄM (monomer concentration),
1
00 mM potassium phosphate, pH 7.2, in a quartz cuvette
2
95 nm increased significantly with respect to that of
with optical pathway of 1 mm. A 1 mm thin film of TMOS
gels containing 3 ꢄM MGL was laid on a quartz slide and
soaked in 100 mM potassium phosphate, pH 7.2, within a
quartz cuvette with optical pathway of 5 mm.
the holo form (Fig. 2(B)) indicating either a quenching
of the Trp emission by the bound PLP or a conforma-
tional change that reduces non radiative emission pro-
cesses. The former case seems unlikely by the absence
of any emission band at higher wavelengths (see also
below).
The fluorescence spectrum upon excitation at 330 nm,
the wavelength where the enolimine tautomer absorbs,
exhibited a band at 388 nm and a shoulder around 500 nm
2
.6. MGL Stability
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
Chemical denaturation of both soluble and encapsulated
MGL was carried out in 100 mM potassium phosphate,
pH 7.2, at 25 C, at increasing guanidinium hydrochlo-
Copyright: American Scientific Publishers
Delivered by Ingenta
ꢁ
ride (GdnHCl) concentration. Samples were incubated for
(
4
Fig. 1(C)). The excitation spectrum of MGL monitored at
00 nm (data not shown) indicates that this emission orig-
6
hours before spectroscopic measurements. Fluorescence
spectra were unchanged after a further 6 hours incubation.
At each denaturant concentration, the extent of spectrum
shift was determined from the displacement of the center
inates from species absorbing at either 330 nm or 280 nm.
The former species is the enolimine tautomer. The lat-
ter species might be due to either a weak energy transfer
between Trp252 and PLP, a process previously observed
4
2ꢃ43
of mass of the fluorescence spectrum, ꢌc,
defined as
follows:
4
4–46
−1
in other PLP-dependent enzymes,
or a direct PLP
ꢌ = ꢍꢊꢎ ×F ꢈ/ꢍF
(1)
c
ꢎ
ꢎ
absorption band centered at 290 nm. The distance between
Trp252 and PLP is about 22 Å, and the planes of PLP and
Trp rings are almost perpendicular (PDB 2rfv). Whereas
the distance between Trp and PLP is within the possibil-
where F is the fluorescence emission at wavelength ꢎ.
ꢎ
The reversibility of the unfolding reaction was assessed
by measuring the degree of recovery of the center of mass
of the fluorescence spectrum upon removal of denaturant.
The protein, both in solution or encapsulated in TEOS and
TMOS, was incubated in the presence of 6 M GdnHCl.
The recovery was conducted transferring the quartz plates
with encapsulated protein or diluting 30-fold the solution
of denatured protein in 100 mM potassium phosphate,
47
ity for an energy transfer, the relative geometry of Trp
and PLP well accounts for its absence. Therefore, because
the absorption of PLP-valine Schiff base showed a band at
4
8
4
10 nm and one at 280 nm (Fig. 1(A), ), it is likely that
the Schiff base absorbance at 280 nm is responsible of the
emission at 400 nm observed for MGL. The emission at
wavelengths higher than 480 nm observed upon excitation
at 330 nm is likely due to an excited state proton trans-
fer converting the enolimine to ketoenamine with emis-
sions at 500 nm that is typical of the direct emission of
the ketoenamine tautomer. This emission band was previ-
ously observed in other PLP dependent enzymes, including
0
2
.1 mM PLP, 1 mM DTT, pH 7.2. Samples were kept at
5 C and the refolding reaction was followed until the
ꢁ
signal reached a constant value.
3
. RESULTS AND DISCUSSION
3
.1. Absorbance and Fluorescence Spectra
4
8
49
The absorption spectrum of MGL from C. freundii is char-
O-acetylserine sulfhydrylase and tryptophan synthase.
acterized by the typical protein band at 280 nm, a small
When MGL was excited at 415 nm, the wavelength where
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018
2213

Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
(A)
(B)
Morozova et al.
(
C)
(D)
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
Copyright: American Scientific Publishers
Delivered by Ingenta
Figure 1. (A) Absorption spectrum of MGL in 100 mM potassium phosphate, 1 mM DTT, pH 8.0 (solid line), and PLP-valine (dashed line).
B–D) Fluorescence emission spectrum of MGL in 100 mM potassium phosphate, 1 mM DTT, pH 7.2, upon excitation at (B) 295 nm, (C) 330 nm
(
and (D) 415 nm.
the ketoenamine tautomer of PLP absorbs, an intense emis-
sion band centered at 494 nm was observed (Fig. 1(D)).
The excitation spectrum of MGL recorded at 500 nm indi-
cates that the emission is originated from absorption bands
at 420, 330 and 290 nm (data not shown). Thus, this
confirms the excited state proton transfer of enolimine to
ketoenamine as well as the involvement of a PLP band
centered at about 290 nm.
(
A)
(B)
Figure 2. Comparison between the absorption spectra (A) and the fluorescence emission spectra (B) of holo- (solid line) and apo-MGL (dashed
line) in 100 mM potassium phosphate, 1 mM DTT, pH 7.2. Fluorescence spectra (ꢎex = 295 nm) are normalized to the respective protein absorbance
at 295 nm.
2214
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018

Morozova et al.
A)
Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
(
(B)
Figure 3. Crystal structure of MGL tetramer (PDB 2RFV). A. Schematic model of a tetramer. Catalytic dimers are shown in green/blue and red/cyan
colors. PLP-binding sites are shown as yellow balls. Trp252 is shown as ball-and-stick representation. B. Stereoview of the hydrophobic center
of tetramer.
3
.2. Activity of TMOS and TEOS MGL Gels
the encapsulation process was explored in the attempt to
5
1–53
enhance the activity of immobilized MGL.
The addi-
MGL was encapsulated in silica gel as a reference entrap-
ment method in order to evaluate the ability of the enzyme
to retain its structure and activity upon immobilization.
Two precursors, either TMOS or TEOS, were used to
assess the potential detrimental effect of the release dur-
ing gel preparation of residual amounts of methanol and
ethanol, respectively, on enzyme activity and structure. It
is worth noting that the resulting silicate matrix is formed
by the same network of silicate bonds when the hydrolysis
tion to the sol of MGL in the presence of PEG 8000 caused
immediate protein precipitation at either 0.5 or 5% final
PEG concentration. The activity of encapsulated MGL
with 0.25 M sorbitol decreased of 10% (specific activity
of 1.61 U/mg compared to 1.80 U/mg of MGL encapsu-
lated without additives), indicating that sorbitol does not
produce a significant improvement. The activity remained
stable with 5% PEG 1000, but under this condition a par-
tial loss of the protein in the solution occurred upon soni-
is complete, as it usually occurs. IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
cation, possibly due to the formation of larger pores. The
MGL activity was measured on a g Ce lo mp yi cr ri go -h st u: sAp me n es ir oi cn an Scientific Publishers
encapsulation of MGL in the presence of 0.5% PEG 1000
Delivered by Ingenta
prepared upon sonication of gels obtained from either
produced an increase in the activity of 25%, indicating that
this concentration of low MW PEG represents a beneficial
additive for the encapsulation of MGL in silica gels.
TMOS or TEOS. MGL gels exhibited six-fold lower spe-
cific activity than in solution, indicating that encapsulation
reduces the catalytic efficiency of MGL but does not abol-
ish it. The kinetic parameters for MGL ꢁ,ꢀ-elimination in
solution and in TMOS silica gels are reported in Table I.
The catalytic efficiency of the encapsulated enzyme is
3
.3. Spectroscopic Properties of TMOS and
TEOS MGL Gels
The absorption spectra of MGL in solution and encap-
sulated in silica gels from either TMOS (Fig. 4(A)) or
TEOS (data not shown) are comparable. This finding sug-
gests that the equilibrium distribution between enolimine
and ketoenamine tautomers does not change in the encap-
sulated enzyme. The fluorescence emission bands of MGL
either TMOS-derived or TEOS-derived gels are very simi-
lar to those observed in solution both for the excitation at
6
.9-fold lower compared to the soluble enzyme. A ten- to
three-fold decrease in activity was also observed for tryp-
tophan synthase, tyrosine phenol-lyase and tryptophan
indole-lyase in nanoporous silica gels.
During the gel polymerization, the gel matrix creates a
cage around the protein molecule.5 The presence of small
molecules as additives during gel formation can modu-
late the pore size and influence the interaction between
the protein and the gel matrix. The effect of osmolytes
as sorbitol and of two different PEG molecules during
3
4
3
5
0
2
95 nm (Fig. 4(B)), indicating that Trp252 environment is
unaltered upon gelification, and for the excitation at 415 nm
Fig. 4(D)). On the other hand, the emission band obtained
3
2
(
upon excitation at 330 nm appeared to be shifted to shorter
wavelengths than that of the soluble enzyme (Fig. 4(C)).
Circular dichroism (CD) spectra were recorded for MGL
in solution (Fig. 5). The spectrum is typical of a protein
containing both alpha and beta secondary structure ele-
ments. The analysis of the soluble enzyme carried out with
K2D (Dichroweb) calculated a value of 52% alpha, 24%
beta and 24% random, in relatively good agreement with
the percentage of secondary structure elements derived
Table I. Catalytic parameters of soluble and encapsulated C. freundii
MGL.
KM (mM) kcat (s 1
−
ꢈ
kcat/KM (mM
−1 −1
s ꢈ
C. freundii
MGL in solution
C. freundii
0.81± 0.1 7.90± 0.3
1.30± 0.2 1.85± 0.1
9.75
1.42
MGL in TMOS silica gel
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018
2215

Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
(A) (B)
Morozova et al.
(
C)
(D)
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
Copyright: American Scientific Publishers
Delivered by Ingenta
Figure 4. A. Absorption spectrum of MGL encapsulated in TMOS silica gels soaked in 100 mM potassium phosphate, pH 7.2 (solid line). The
absorption spectrum of soluble MGL is shown for comparison (dashed line). B–D. Fluorescence emission spectra of MGL encapsulated in TMOS
silica gels soaked in 100 mM potassium phosphate, pH 7.2, upon excitation at (B) 295 nm, (C) 330 nm and (D) 415 nm.
from the 3D structure of MGL (2RFV) that are 40% alpha,
5% beta and 45% random. Discrepancies might be due
both to limitations of fitting software and small differ-
ences between solution and crystal structure. When CD
spectra of TMOS MGL gels were recorded and analyzed
(Fig. 5), it was found that the content of secondary struc-
tures for soluble and encapsulated MGL was the same,
indicating that encapsulation in silica matrices did not alter
the enzyme secondary structure.
1
3
.4. MGL Stability in Solution and in Silica Gels
Encapsulation of proteins in nanoporous silica gels is
reported to affect stability with respect to thermal and
3
2
chemical denaturation, or long-term storage. The sta-
bilizing effect of encapsulation of the protein has been
5
4–57
previously observed,
been observed.
although conflicting results have
5
8ꢃ59
The stability of MGL either in solu-
tion or entrapped in TEOS silica gels was investigated
monitoring absorbance (Figs. 6(A–B)) and fluorescence
(
Figs. 7(A–D)) changes as a function of GdnHCl con-
centration. Figure 6(A) shows that, for soluble MGL, the
presence of 6 M GdnHCl caused the shift of the PLP
absorption band from 420 nm to 390 nm, the wave-
length typical of free PLP, suggesting that the coenzyme
was released from the protein. For MGL encapsulated
in gels (Fig. 6(B)), the addition of GdnHCl caused the
Figure 5. CD spectra of MGL in 100 mM potassium phosphate, pH
7.2 (solid line) and upon encapsulation in TMOS silica gels soaked in
the same buffer (dashed line).
2216
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018

Morozova et al.
A)
Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
(
(B)
Figure 6. Denaturation of MGL in GdnHCl. Comparison between absorption spectrum of native (solid line) and unfolded in 6 M GdnHCl (dashed
line) state of MGL. A. 100 mM potassium phosphate, pH 7.2 and B. TEOS-derived silica gels.
disappearance of the absorption band at 420 nm with the
concomitant appearance of a weak intensity band of free
PLP at 390 nm. This occurs because free PLP escapes
from gel and equilibrates with the surrounding medium,
that has a volume about 40-fold larger compared to the gel
volume. The release of the coenzyme took place already
at low concentration of denaturant for both soluble and
encapsulated MGL (data not shown).
(
A)
(B)
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
Copyright: American Scientific Publishers
Delivered by Ingenta
(
C)
(D)
Figure 7. Changes of fluorescence emission upon denaturation. Comparison between fluorescence emission from native (solid line) and in 6 M
GdnHCl (dashed line) MGL, in solution and encapsulated in TEOS gels. A and C: MGL in solution; B and D: encapsulated MGL. A and B:
ꢎex = 295 nm; C and D: ꢎex = 330 nm.
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018
2217

Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
Morozova et al.
sticky hydrophobic regions of the protein. Similar aggrega-
tion of denaturation intermediates has been observed for a
6
1–63
wide variety of proteins.
Within the gel network, inter-
protein interactions are prevented and removal of denatu-
rant after complete unfolding in 6.0 M GdnHCl results in
a partial recovery of the initial fluorescence center of mass
(
data not shown) with no evidence of aggregation.
4
. CONCLUSIONS
Our results indicate that encapsulated MGL retains most
of the structural features of the soluble enzyme. The stabil-
ity of MGL towards chemical denaturation increases upon
immobilization. The reduction in activity is likely due to
some conformational constraint during the catalytic cycle.
This is also confirmed by the increase in activity measured
upon addition of PEG 1000 during the polymerization pro-
cess. However, the decrease in catalytic efficiency does not
hamper the possibility to exploit entrapped MGL for the
development of either an anti-cancer or antibiotic agent.
Encapsulation of MGL in silica gels or in different matri-
ces might represent an effective strategy to enhance the
stability of the protein towards proteases in the body, to
increase the bioavailability, and to reduce side effects due
to protein immunogenicity.
Figure 8. Equilibrium denaturation curves in 100 mM potassium phos-
phate, pH 7.2, at 25 C. Samples are: Soluble MGL (squares), and MGL
encapsulated in TEOS (circles) or in TMOS (stars). The line through
the points represents the best fit of the experimental data to Boltzmann
equation.
ꢁ
The spectral characteristics of the fluorescence emission
of encapsulated MGL, excited at 295, 330 and 415 nm,
at different concentrations of denaturant were similar to
that of the protein in solution. In the presence of 6 M
GdnHCl the maximum of fluorescence emission, upon
excitation at 295 nm, moved from 333 nm to 350 nm
Acknowledgments: This work was carried out with
(
Figs. 7(A–B)), the emission wavelength typical for the
IP: 5.53.118.130 On: Sat, 16 Jun 2018 16:51:40
the financial support of a bilateral project between the
Copyright: American Scientific Publishers
National Research Council of Italy (CNR) (code MD
unfolded form of a protein. Enzyme denaturation is accom-
panied by a two-fold increase in the fluorescen cD ee il inv t ee nr e- d by Ingenta
P01.008) and the Russian Foundation for Basic Research
RFBR) (code 15-54-78021). Authors are grateful to
sity, indicating PLP release. Upon direct excitation of
the coenzyme at 330 nm, the unfolded MGL showed an
increase in the emission intensity and a shift of the max-
imum to 396 nm (Figs. 7(C–D)), wavelength typical of
free PLP. Only little changes in the shape and intensity of
the fluorescence spectrum were observed upon excitation
at 415 nm (data not shown).
(
Mr. Gianluca Presciuttini and Mr. Alessandro Puntoni for
their valuable technical input. The Interdepartmental Cen-
ter for Measurements (CIM) of the University of Parma is
acknowledged for CD equipment.
The dependence of MGL fluorescence emission on
GdnHCl concentration was determined by measuring the
displacement of central of mass of protein emission, cal-
culated as described in Materials and Methods, at increas-
ing denaturant concentration (Fig. 8). The comparison of
the denaturation curves indicates that the apparent unfold-
ing midpoint, 2.2 ± 0.1 M in solution and 3.0 ± 0.1 M
in silica gel, is shifted toward higher concentration of
denaturant upon encapsulation (Fig. 8). It is likely that
during encapsulation, proteins are trapped within the sil-
ica cage in pores tailored to their size and the restriction
of protein mobility within this confined space results in
thermodynamic stabilization of native states. A straight-
forward analysis of the equilibrium curves obtained for
MGL and the determination of meaningful thermodynamic
parameters were prevented by the lack of complete recov-
ery. Control refolding experiments for MGL in solution
showed the appearance of large amount of aggregates. It is
feasible that the aggregation is due to the exposure of
References and Notes
1. H. Tanaka, N. Esaki, and K. Soda, Enzyme Microb. Technol. 7, 530
(1985).
2
3
. H. Tanaka, N. Esaki, and K. Soda, Biochemistry 16, 100 (1977).
. E. A. Morozova, N. P. Bazhulina, N. V. Anufrieva, D. V. Mamaeva,
Y. V. Tkachev, S. A. Streltsov, V. P. Timofeev, N. G. Faleev, and
T. V. Demidkina, Biochemistry (Mosc) 75, 1272 (2010).
4. L. Davis and D. Metzler, The Enzymes 3rd (edn.), edited by P.D.
Boyer, Academic Press, New York (1972), Vol. 7, p. 33.
5. A. Nikulin, S. Revtovich, E. Morozova, N. Nevskaya, S. Nikonov,
M. Garber, and T. Demidkina, Acta Crystallogr D Biol. Crystallogr
64, 211 (2008).
6
. M. Fukumoto, D. Kudou, S. Murano, T. Shiba, D. Sato, T. Tamura,
S. Harada, and K. Inagaki, Biosci. Biotechnol. Biochem. 76, 1275
(2012).
6
0
7. E. A. Morozova, S. V. Revtovich, N. V. Anufrieva, V. V. Kulikova,
A. D. Nikulin, and T. V. Demidkina, Acta Crystallogr D Biol. Crys-
tallogr 70, 3034 (2014).
8
. D. Kudou, S. Misaki, M. Yamashita, T. Tamura, N. Esaki, and
K. Inagaki, Biosci. Biotechnol. Biochem. 72, 1722 (2008).
. A. E. McKie, T. Edlind, J. Walker, J. C. Mottram, and G. H. Coombs,
J. Biol. Chem. 273, 5549 (1998).
9
2218
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018

Morozova et al.
Soluble and Nanoporous Silica Gel-Entrapped C. freundii Methionine ꢀ-Lyase
1
0. E. A. Morozova, V. V. Kulikova, A. N. Rodionov, S. V. Revtovich,
N. V. Anufrieva, and T. V. Demidkina, Biochimie 128–129, 92
2016).
36. I. V. Manukhov, D. V. Mamaeva, E. A. Morozova, S. M. Rastorguev,
N. G. Faleev, T. V. Demidkina, and G. B. Zavilgelsky, Biochemistry
(Mosc) 71, 361 (2006).
(
1
1. V. V. Kulikova, N. V. Anufrieva, S. V. Revtovich, A. S. Chernov,
G. B. Telegin, E. A. Morozova, and T. V. Demidkina, IUBMB Life
37. S. Bettati and A. Mozzarelli, J. Biol. Chem. 272, 32050 (1997).
38. S. Abbruzzetti, S. Faggiano, S. Bruno, F. Spyrakis, A. Mozzarelli,
S. Dewilde, L. Moens, and C. Viappiani, Proc. Natl. Acad. Sci. USA
106, 18984 (2009).
39. L. Ronda, S. Faggiano, S. Bettati, N. Hellmann, H. Decker,
T. Weidenbach, and A. Mozzarelli, Gene 398, 202 (2007).
40. J. D. Brennan, J. S. Hartman, E. I. Ilnicki, and M. Rakic, Chemistry
of Materials 11, 1853 (1999).
68, 830 (2016).
1
1
1
2. D. Sato and T. Nozaki, IUBMB Life 61, 1019 (2009).
3. Y. Tan, M. Xu, and R. M. Hoffman, Anticancer Res. 30, 1041 (2010).
4. A. Amadasi, M. Bertoldi, R. Contestabile, S. Bettati, B. Cellini,
M. L. di Salvo, C. Borri-Voltattorni, F. Bossa, and A. Mozzarelli,
Curr. Med. Chem. 14, 1291 (2007).
1
1
5. P. Cavuoto and M. F. Fenech, Cancer Treat Rev. 38, 726 (2012).
6. T. Takakura, A. Takimoto, Y. Notsu, H. Yoshida, T. Ito, F. Nagatome,
M. Ohno, Y. Kobayashi, T. Yoshioka, K. Inagaki, S. Yagi, R. M.
Hoffman, and N. Esaki, Cancer Research 66, 2807 (2006).
7. T. Batool, E. A. Makky, M. Jalal, and M. M. Yusoff, Applied Bio-
chemistry and Biotechnology 178, 900 (2016).
41. S. A. Ahmed, C. C. Hyde, G. Thomas, and E. W. Miles, Biochem-
istry 26, 5492 (1987).
42. G. Mei, A. Di Venere, F. M. Campeggi, G. Gilardi, N. Rosato, F. De
Matteis, and A. Finazzi-Agro, Eur. J. Biochem. 265, 619 (1999).
43. E. Gabellieri, E. Balestreri, A. Galli, and P. Cioni, Biophysical Jour-
nal 95, 771 (2008).
44. B. Campanini, S. Raboni, S. Vaccari, L. Zhang, P. F. Cook, T. L.
Hazlett, A. Mozzarelli, and S. Bettati, J. Biol. Chem. 278, 37511
(2003).
45. G. B. Strambini, P. Cioni, A. Peracchi, and A. Mozzarelli, Biochem-
istry 31, 7535 (1992).
1
1
8. N. V. Anufrieva, E. A. Morozova, V. V. Kulikova, N. P. Bazhulina,
I. V. Manukhov, D. I. Degtev, E. Y. Gnuchikh, A. N. Rodionov, G. B.
Zavilgelsky, and T. V. Demidkina, Acta Naturae 7, 128 (2015).
9. V. R. Sinha and A. Trehan, J. Control Release 90, 261 (2003).
0. R. Solaro, F. Chiellini, and A. Battisti, Materials 3, 1928 (2010).
1. R. Vaishya, V. Khurana, S. Patel, and A. K. Mitra, Expert Opinion
on Drug Delivery 12, 415 (2015).
1
2
2
46. M. Marchetti, S. Bruno, B. Campanini, S. Bettati, A. Peracchi, and
A. Mozzarelli, Amino Acids 47, 163 (2015).
2
2. S. Mohammadi-Samani and B. Taghipour, Pharm. Dev. Technol.
47. J. R. Lakowicz, Principles of Fluorescence Spectroscopy 3rd edn.,
Plenum Press, New York (2006).
20, 385 (2015).
2
2
3. W. R. Gombotz and D. K. Pettit, Bioconjug Chem. 6, 332 (1995).
4. S. Bettati, B. Pioselli, B. Campanini, C. Viappiani, and
A. Mozzarelli, Encyclopedia of Nanoscience and Nanotechnology,
edited by H. S. Nalwa, American Scientific Publishers, Los Angeles
48. S. Benci, S. Vaccari, A. Mozzarelli, and P. F. Cook, Biochimica
Et Biophysica Acta-Protein Structure and Molecular Enzymology
1429, 317 (1999).
49. S. Vaccari, S. Benci, A. Peracchi, and A. Mozzarelli, Biophysical
Chemistry 61, 9 (1996).
(2004), Vol. 9, p. 81.
25. S. Bruno, L. Ronda, S. Abbruzzetti, C. Viappiani, S. Bettati,
50. S. Abbruzzetti, C. Viappiani, S. Bruno, S. Bettati, M. Bonaccio, and
S. Kumar Maji, and A. Mozzarelli, EI Pn c :y c5l o. 5p e3d .i 1a 1o 8f .N1 a3n 0o s cOi e nn c: eSat, 16 AJ u. Mn o 2z z0 a 1r e8l li ,1 J6.:N5a1n:o4sc0i. Nanotechnol. 1, 407 (2001).
and Nanotechnology, edited by H. S. Nal Cw ao ,p Ay mr i egr ihc at :n AS mc ie en tr i ific can S 5c 1i .e nJ .t iDf i. cB Pr e nu nba ln i ,s Dh .e Br es njamin, E. DiBattista, and M. D. Gulcev, Chem.
Publishers, Los Angeles (2011), Vol. 21, p. 481.
6. T. Andreani, A. M. Silva, and E. B. Souto, Biotechnol. Appl.
Biochem. 62, 754 (2015).
7. X. Y. Li, M. Y. Wu, L. M. Pan, and J. L. Shi, International Journal
of Nanomedicine 11, 93 (2016).
8. M. S. Bradbury, E. Phillips, P. H. Montero, S. M. Cheal, H. Stam-
buk, J. C. Durack, C. T. Sofocleous, R. J. Meester, U. Wiesner, and
S. Patel, Integr. Biol. (Camb) 5, 74 (2013).
Mater. 15, 737 (2003).
Delivered by Ingenta
2
2
2
52. D. K. Eggers and J. S. Valentine, J. Mol. Biol. 314, 911 (2001).
53. C. M. Soares, O. A. dos Santos, H. F. de Castro, F. F. de Moraes,
and G. M. Zanin, Appl. Biochem. Biotechnol. 113-116, 307 (2004).
54. D. Bolis, A. S. Politou, G. Kelly, A. Pastore, and P. A. Temussi,
Journal of Molecular Biology 336, 203 (2004).
55. N. Tokuriki, M. Kinjo, S. Negi, M. Hoshino, Y. Goto, I. Urabe, and
T. Yomo, Protein Science 13, 125 (2004).
2
3
3
9. L. Ronda, S. Bruno, S. Faggiano, S. Bettati, and A. Mozzarelli,
56. M. Bottini, A. Di Venere, L. Tautz, A. Desideri, P. Lugli,
L. Avigliano, and N. Rosato, Journal of Sol–Gel Science and Tech-
nology 30, 205 (2004).
57. K. Flora and J. D. Brennan, Analytical Chemistry 70, 4505 (1998).
58. B. Campanini, S. Bologna, F. Cannone, G. Chirico, A. Mozzarelli,
and S. Bettati, Protein Science 14, 1125 (2005).
Methods Enzymol. 437, 311 (2008).
0. S. Abbruzzetti, S. Bruno, S. Faggiano, L. Ronda, E. Grandi,
A. Mozzarelli, and C. Viappiani, Methods Enzymol. 437, 329 (2008).
1. G. Chirico, F. Cannone, S. Beretta, A. Diaspro, B. Campanini,
S. Bettati, R. Ruotolo, and A. Mozzarelli, Protein Sci. 11, 1152
(
2002).
59. L. Fuentes, J. Oyola, M. Fernandez, and E. Quinones, Biophysical
Journal 87, 1873 (2004).
60. D. Avnir, T. Coradin, O. Lev, and J. Livage, Journal of Materials
Chemistry 16, 1013 (2006).
3
2. L. Ronda, S. Bruno, B. Campanini, A. Mozzarelli, S. Abbruzzetti,
C. Viappiani, A. Cupane, M. Levantino, and S. Bettati, Current
Organic Chemistry 19, 1653 (2015).
3
3
3
3. M. T. Reetz, A. Zonta, and J. Simpelkamp, Biotechnology and Bio-
engineering 49, 527 (1996).
4. B. Pioselli, S. Bettati, and A. Mozzarelli, FEBS Lett. 579, 2197
61. R. Rudolph, Modern Methods in Protein and Nucleic Acid Research,
edited by H. Tschesche, Walter de Gruyter, Berlin (1990), p. 149.
62. P. Dominici, P. S. Moore, and C. B. Voltattorni, Biochemical Journal
295, 493 (1993).
63. S. Seshadri, K. A. Oberg, and V. N. Uversky, Current Protein and
Peptide Science 10, 456 (2009).
(2005).
5. B. Pioselli, S. Bettati, T. V. Demidkina, L. N. Zakomirdina, R. S.
Phillips, and A. Mozzarelli, Protein Sci. 13, 913 (2004).
Received: 13 December 2016. Accepted: 30 January 2017.
J. Nanosci. Nanotechnol. 18, 2210–2219, 2018
2219