Structural and functional characterization of a recombinant leucine aminopeptidase

Article abstract of DOI:10.1016/j.molcatb.2014.12.013

The function of proteins, such as the catalytic enzyme activity, depends on the interaction of their active sites with their specific substrates and the environment conditions that affect the stability of those sites. This study presents a structure-to-fu

Full text of DOI:10.1016/j.molcatb.2014.12.013

Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Structural and functional characterization of a recombinant leucine
aminopeptidase
a
a
Ana V. Hernández-Moreno , Francisco C. Perdomo-Abúndez ,
a
b
c
Victor Pérez-Medina Martínez , Gabriel Luna-Bárcenas , Francisco Villase n˜ or-Ortega ,
Néstor O. Pérez , Carlos A. López-Morales , Luis F. Flores-Ortiz , Emilio Medina-Rivero
Unidad de Investigación y Desarrollo, PROBIOMED S.A. de C.V., Tenancingo, Estado de México, Mexico
Grupo de investigación de polímeros y biopolímeros, Cinvestav Querétaro, QRO, Mexico
a
a
a
a,∗
a
b
c
Departamento de Ingeniería Bioquímica, Instituto Tecnológico de Celaya, GTO, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 11 September 2014
Received in revised form
The function of proteins, such as the catalytic enzyme activity, depends on the interaction of their active
sites with their specific substrates and the environment conditions that affect the stability of those sites.
This study presents a structure-to-function characterization of the folding process of a recombinant
2
6 December 2014
6
×-His tag leucine aminopeptidase (rLAP) based on a platform of analytical techniques. The results
Accepted 27 December 2014
demonstrated an increase up to 31 U/mg in the activity of the enzyme after folding as revealed by
circular dichroism, intrinsic fluorescence, differential scanning calorimetry, and free thiol analysis. Collec-
tively, these techniques revealed a larger number of covalent and non-covalent bonds within the protein
seen as an increase in the chemical and thermal stability, while exhibited a lower level of non-bonded
cysteines after the protein was folded. Mass spectrometry analysis showed the maintenance of the dis-
tribution of the enzyme isoforms related to N-terminal histidine residues after folding, which confirmed
that the enzymatic activity of rLAP depends on its three-dimensional structure rather than N-terminal
self-processing activity. In summary, the studied attributes allow a better understanding of the structure-
to-function relationship of rLAP, that permit a more proficient manufacturing of the enzyme that would
improve the bioprocesses in which is employed.
Available online 3 January 2015
Keywords:
Recombinant leucine aminopeptidase
Folding process
Spectroscopic and thermodynamic
techniques
Thermal and chemical stability
©
2015 Elsevier B.V. All rights reserved.
1
. Introduction
and used under specific environments [9–11]. Additionally, iden-
tifying the principles of protein folding allows understanding the
forces that drive and/or preclude this phenomenon [12,13].
The gaining of knowledge about the relationship between struc-
ture, function, and the environment conditions; depends on the
study of variables that affect folding, such as: protein concentration,
pH, temperature, and ionic strength [14–16]. So, adjustments on
the conditions of the protein environment can be planned toward
the improvement of its manufacturing bioprocesses. On this regard,
enzymes are ideal model proteins, because they act on specific sub-
strates that can be employed to measure their biological activity,
and thus investigate the impact of environmental variables on their
functionality.
Recombinant leucine aminopeptidase (rLAP) is an enzyme com-
monly used by the biopharmaceutical industry for the remotion
of N-terminal methionine [17–21]. The mature 6×-His tag rLAP
under study is a non-glycosilated monomeric protein of 304 amino
acids, with an apparent averaged mass of 32.8 kDa (by reducing
SDS-PAGE) [17]. This enzyme exhibits a characteristic secondary
structure, as reported by circular dichroism [22]; furthermore, as
DNA technology has been widely used for the production of
recombinant proteins such as therapeutic proteins and enzymes.
Currently, the production of recombinant proteins in prokaryotic
hostess is an advantageous process because allows yielding higher
amounts of product [1,2]; nevertheless, the expression of recombi-
nant proteins in Escherichia coli, and the purification of insoluble
inclusion bodies represent major challenges in determining the
correct structure of the recombinant proteins that ensure their
biological activity [3–6].
The study of the physical attributes of proteins, and their effects
on their biological function, is an important asset in the establish-
ment of the processes and conditions to obtain the appropriate
protein structure [7,8]; in addition, is also important to determine
the optimal conditions in which these proteins will be preserved
∗ Corresponding author. Tel.: +52 55 11662280; fax: +52 55 53527651.
E-mail address: emilio.medina@probiomed.com.mx (E. Medina-Rivero).
http://dx.doi.org/10.1016/j.molcatb.2014.12.013
1
381-1177/© 2015 Elsevier B.V. All rights reserved.

4
0
A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
in the wild type leucine aminopeptidase (LAPwt) [19,23], rLAP has
only one disulfide bond in its mature form which makes easier to
study its folding.
rLAP). For the mass spectroscopy analysis the enzyme was incu-
bated at 15 C under the main folding conditions, and was analyzed
every 24 h.
◦
In a previous study, we reported an increase in the enzymatic
activity of rLAP after an incubation period that was conceived as
the best for folding [22]. Herein, we focus on the structural changes
of rLAP and the repercussion on its functional attributes headed to
the improvement of bioprocesses.
2.4. Circular dichroism (CD)
The analysis was conducted at 15 ␮M of enzyme in the far UV
CD region (190–270 nm) at 25 C using a 0.1 cm path length cell
◦
To achieve this objective, a set of sensitive techniques capa-
ble to detect conformational changes in proteins was employed,
such as: circular dichroism (CD), intrinsic fluorescence (IF), differ-
ential scanning calorimetry (DSC), and mass spectrometry (MS),
along with colorimetric free thiol analysis, and enzymatic activ-
ity assays that helped to track the changes of rLAP subjected to
folding and/or induced denaturation. Collectively, these techniques
provided a quick and reliable platform to assess structural and
functional changes in the protein associated to predetermined vari-
ations on its environment.
in a Jasco J-815 spectropolarimeter (Jasco Corporation Ltd.; Tokyo,
Japan). Three scans per test were accumulated at 1 nm bandwidth.
2.5. Intrinsic fluorescence (IF)
The analysis was conducted at 3 ␮M of enzyme in 3 mm quartz
cells. Fluorescence spectrum was monitored from 290 to 400 nm
◦
at 25 C, using an excitation wavelength of 283 nm in a Fluorolog-
3
spectrofluorometer (Horiba Ltd.; Tokyo, Japan). The wavelength
of maximal emission (ꢀmax) was defined as the point of maximal
intensity from the fluorescence spectra.
2
. Materials and methods
2.6. Thermal analysis by IF
2.1. Materials
◦
IF was monitored in a range from 5 to 90 C, at a scanning rate
of 1 C min using the same instrument and settings mentioned
above.
Wild type leucine aminopeptidase from Vibrio proteolyti-
◦
−1
cus, Leucine-p-nitronilide (L-pNA), zinc chloride (ZnCl ), dl-
2
dithiotreitol, tricine, acetonitrile (LC–MS grade), formic acid
(
LC–MS grade), and water (LC–MS grade) were obtained from
2
.7. Fluorescence lifetime using time correlated single photon
Sigma–Aldrich (St. Louis, MO); citric acid, sodium citrate,
mono-dibasic sodium phosphate, sodium carbonate, and sodium
bicarbonate from J. T. Baker (Center Valley, PA); Urea from
Millipore Corp. (Billerica, MA); DyLight 488 Maleimide and Slyde-
A-Lyzer 10 kDa MWCO Cassettes from Thermo Fisher Scientific Inc.
counting (TCSPC)
Fluorescence lifetime analysis by the Time-Correlated Sin-
◦
gle Photon Counting method (TCSPC) was performed at 25 C
using 332 nm as emission wavelength, 2 nm band-pass, and a
2
ried out with the DAS6 software (Horiba Ltd.; Tokyo, Japan) using
two exponential decay components. The intensity decays were cal-
culated as the sum of exponentials using the Eq. (1) [25].
(
Waltham, MA); and Recombinant leucine aminopeptidase (rLAP)
80 nm ± 10 Nanoled as excitation source. Data analysis was car-
in-house produced.
2
.2. Activity assay
ꢁ
ꢂ
ꢀ
−
t
Enzymatic activity was determined by hydrolysis of L-pNA.
I(t) =
˛ exp
(1)
i
ꢁ
The assays were conducted by triplicate for each sample of rLAP
i
i
−
8
in a final volume of 480 ␮L; 0.5 mM L-pNA and 3 × 10 M of
purified enzyme were incubated in 10 mM tricine, 1 mM ZnCl2
buffer at pH 8.0. A negative control without enzyme was used to
account for non-enzymatic hydrolysis of the substrate. The reac-
where ꢂ␣ is normalized to the unit, ˛ is the pre-exponential
factor, and ꢁ is the lifetime. The lifetime average (ꢁ) was calculated
using the Eq. (2) [26].
i
i
i
◦
ꢃ
tions were incubated at 25 C for 5 min, and then 20 ␮L of 20 mM
ꢁi2
i=1˛i
dl-dithiotreitol were added to quench the reaction. Absorbance
ꢀꢁꢁ = ꢃ
(2)
−
1
−1
˛iꢁi
from p-nitroaniline at 405 nm (ε = 10,800 M cm ) [24], and from
i=1
−1
−1
rLAP (ε = 38,700 M cm ) was measured in a Beckman Coulter
2
.8. Chemical denaturation observed by IF and TCSPC
DU 640 spectrophotometer (Beckman Coulter Inc.; Brea, CA). One
enzyme unit (U) was defined as the amount of enzyme that released
◦
Solutions of 6 ␮M rLAP were prepared at different concentra-
1
␮mol of p-nitroaniline at 25 C in 1 min.
tions of urea (0–8 M) in 10 mM tricine, 1 mM ZnCl buffer at pH 8.0.
2
◦
Urea-induced unfolding was analyzed by IF and TCSPC at 25 C as
2.3. Folding conditions
described in the methodology of those techniques.
3
␮M rLAP (in the folding state in which was purified from the
◦
inclusion bodies, referred as unfolded) was incubated at 37 C in
0 mM phosphate, 10 ␮M ZnCl buffer at pH 6.0 for 72 h (main fold-
2.9. Chemical denaturation evaluated by enzymatic activity
1
2
ing conditions). The samples were collected every 12 h to track the
enzyme activity. Three different combinations of rLAP and ZnCl2
concentrations were also evaluated to observe their effect on the
folding process: (1) 30 ␮M and 20 ␮M, (2) 152 ␮M and 50 ␮M,
and (3) 305 ␮M and 100 ␮M respectively. After 72 h of folding, the
The urea-induced unfolding of 6 ␮M rLAP was measured by its
specific enzymatic activity at different urea concentrations (from
◦
0 to 8 M) in 10 mM tricine, 1 mM ZnCl2 buffer at pH 8.0 at 25 C.
Reversibility of the chemical denaturation was evaluated by sub-
jecting the enzyme to decreasing urea dilutions from 8 M to 0.4 M.
protein was dialyzed in Slyde-A-Lyzer 10 kDa MWCO cassettes, in
◦
1
0 mM tricine, 1 mM ZnCl2 buffer pH 8.0, at 4 C, for 24 h; the pro-
2.10. Differential scanning calorimetry (DSC)
tein obtained from this step was referred as rLAPfd (folded rLAP).
Enzyme no subjected to folding was dialyzed under the same con-
ditions prior to be analyzed and was referred as rLAPuf (unfolded
Transition temperatures (Tm) and enthalpies (ꢃHcal) were
measured at 46 ␮M of rLAP using a nano-DSC system from TA

A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
41
.3 mL min 1 at 70 C and a linear gradient from 30 to 65% phase B
0.10% formic acid in acetonitrile) were used for LC analyses. Pos-
itive mode of ESI ionization source, 3.0 kV capillary voltage, 40 V
−
◦
0
(
◦
◦
cone voltage, 350 C desolvation temperature, 120 C source tem-
perature, and acquisition range from 300 to 3500 m/z were used for
MS analyses. MS data was analyzed using MaxEnt1 algorithm. The
molecular masses of rLAP isoforms were calculated based on the
amino acid sequence of the mature form of LAPwt [29].
3. Results
3.1. Functional and structural properties of rLAP
In a previous study we reported an increase in the enzymatic
◦
activity from 1 to 18 U/mg when 3 ␮M rLAP were incubated at 37 C
for 72 h in a buffer of 10 mM phosphate, 10 ␮M ZnCl2 at pH 8.0
[
22]. In this study we conducted exploratory tests by employing
different pH values and buffers to assess the folding process of rLAP
through its activity toward the optimization of the process. When
tricine buffer at pH 8.0, or citrate buffer at pH 4.0, 5.0 or 6.0 were
employed, the enzymatic activity remained below 2.5 U/mg. A sub-
tle increase to 3 U/mg was observed using carbonate buffer at pH 8.5
and 9.0; however, when the pH increased to 10.0 the activity was
Fig. 1. Enzymatic activity of three in house-produced batches (LO11005 squares;
LO12001 circles, and LO10004 triangles) of rLAP over the time of folding, under
main folding conditions.
◦
totally lost after 72 h at 37 C. An increase to 15 U/mg was observed
Instruments (New Castle, DE) equipped with a 300 ␮L platinum
after 72 h of incubation in phosphate buffer at pH 7.0 and 8.0, which
remained similar to our original assessment, performed using a dif-
ferent batch of rLAPuf [22], however, when pH decreased down to
cell. Each sample was evaluated in triplicate at a scanning rate of
◦
−1
.
1
C min
6
.0, the specific activity of rLAP augmented to 31 U/mg after 72 h
2
.11. Free thiol analysis
of incubation. The enzymatic activity of rLAP folded in phosphate
buffer at pH 6.0, along with the aforementioned results at pH 8.0,
suggests that the folding process was reproducible among batches
(Fig. 1).
Free thiols were derivatized using DyLight 488 Maleimide to a
final concentration of 12.5 ␮M in a buffer of 100 mM phosphate
at pH 8.0 [27]. The analysis was carried out in an ACQUITY UPLC
System from Waters (Milford, MA), using a SEC column BEH 200,
CD spectra revealed structural changes among: rLAPfd, rLAPuf
,
and LAPwt (Fig. 2). Although, the minimum negative ellipticity
([Â]mre) of these enzymes was centered at 208 nm (Fig. 2A), the CD
spectrum intensity of rLAPfd was 0.4 times higher than the intensity
of rLAPuf, and 0.4 times lower compared to the intensity of LAPwt.
IF spectrum of rLAPfd confirmed structural changes during the
folding process, showing a decrease in the intensity of IF, and a blue
shift of the ꢀmax from 330 to 325 nm exhibiting a closer profile to
LAPwt (Fig. 2B). On the other hand, when rLAPfd was chemically
denatured with urea revealed a red shift of its ꢀmax to 345 nm
−
1
4
.1 × 300 mm, using a flow rate of 1.7 ␮M at a 0.4 mL min
for
protein-fluorophore separation and quantification [28].
2.12. Mass spectrometry (MS)
Liquid chromatography electrospray ionization mass spectrom-
etry (LC–ESI-MS, hereafter referred as MS) was performed in an
ACQUITY UPLC System coupled to a SYNAPT G2 HDMS mass spec-
trometer from Waters (Milford, MA). A BEH300 C18 (1.7 ␮m)
reverse phase column, UV detection at 214 nm, a flow rate of
(Fig. 2B) and an increment in the intensity of its IF exhibiting a
closer profile to rLAPuf
.
Fig. 2. (A) Circular dichroism spectra: LAPwt (A-a), rLAPfd (A-b), and rLAPuf (A-c). (B) Intrinsic fluorescence spectra: LAPwt (B-a), rLAPfd (B-b), rLAPuf (B-c), and rLAPuf denatured
in 8 M urea (B-d).

4
2
A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
◦
◦
Fig. 3. (A) Thermal unfolding of rLAP analyzed by DSC; rLAPuf (short-dot line) exhibits only one Tm at 54 C, where as rLAPfd (solid line) exhibits three Tm at 54, 85, and 88 C.
(
◦ ◦ ◦
B) Thermal denaturation of rLAP analyzed by IF; rLAPuf (closed circles) exhibits two inflection points (16 C and 50 C), whereas rLAPfd exhibited only one around 85 C.
Table 1
Thermodynamic parameters of rLAP by DSC.
Sample
Tm
Tm
( C)
Tm
( C)
ꢃH1
(kCal mol
ꢃH2
(kCal mol
ꢃH3
(kCal mol
ꢃHcal
1
2
3
◦
◦
◦
−1
−1
−1
−1
(kCal mol )
(
C)
)
)
)
rLAPuf
rLAPfd
54 ± 0.1
54 ± 0.3
NP
85 ± 0.4
NP
88 ± 0.1
55 ± 2
4 ± 1
NP
65 ± 3
NP
31 ± 2
55 ± 2
100 ± 5
NP: not present.
3
.2. Thermal stability of rLAP
addition, when ꢀmax of rLAPuf was analyzed as the first derivative in
function of the temperature (dꢀmax/dt) a minimum and a maximum
◦
◦
Thermal stability evaluated through DSC showed that rLAPfd
inflection points (at 16 C and 50 C, respectively) were observed.
◦
◦
exhibited three different Tm (54, 85, and 88 C) while rLAP only
exhibited one Tm at 54 C (Fig. 3A and Table 1). The transition
at 54 C found in both, folded and unfolded enzyme, exhibited
enthalpies that differed in about 13 times, where the higher
A thermal transition was observed around 85 C for rLAPfd.
uf
◦
◦
3.3. Urea-induced unfolding of rLAP by lifetime and intrinsic
fluorescence
enthalpy was observed in rLAPuf (Table 1). The total enthalpy
(
ꢃHcal) differed in almost two times between rLAPfd and rLAPuf
Thermal denaturation of rLAP was analyzed by IF, and showed
.
The values of ꢁ and ꢀmax (Fig. 4) augmented as the concentration
of urea increased, although ꢁ showed a marked inflection across the
denaturation gradient in both, folded or unfolded enzymes. Fluo-
rescence lifetime values and ꢄ2 (as the measure of fitting) of rLAP
and LAPwt are shown in Table 2. Another interesting result is that
a different behavior between rLAPuf and rLAPfd (Fig. 3B). For
rLAPuf, ꢀmax behavior was defined by temperature regions, where
the displacement below 35 C shifted toward blue fluorescence,
◦
whereas above this temperature shifted toward red fluorescence. In
the ꢁ value of rLAPfd is the closest to the ꢁ value of LAPwt.
Fig. 4. Chemical denaturation of rLAP in function of the urea concentration assessed by ꢁ (solid squares) and ꢀmax (open circles). (A) The range of ꢁ between the initial and
the final stress levels for rLAPuf was 2.20–2.90 ns, with a red shift of ꢀmax from 331 to 345 nm. (B) The range of ꢁ for rLAPfd was from 1.97 to 2.15 ns and its ꢀmax was from
3
30 nm to 333 nm.

A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
43
Table 2
Fluorescence lifetime values of rLAP.
Protein
ꢁ
ꢄ2
(
ns)
LAPwt
rLAPuf
rLAPuf in 8 M urea
rLAPfd
rLAPfd in 8 M urea
1.61
2.23
2.91
1.96
2.17
1.20
1.20
1.26
1.16
1.24
Data fitted using two exponential decays.
3.4. Reversibility of the urea-induced unfolding process
The enzymatic activity evaluated at several dilutions of urea,
◦
from 8 M to 0.4 M, at 25 C, showed that the enzymatic activity
increased as the urea concentration was diluted (Fig. 5).
3.5. Folding promoted the formation of disulfide bond of rLAP
Fig. 6. Percentage of established (white area) and non-established disulfide bonds
A 0.54 mol of free cysteines/mol of protein ratio was observed for
(
shaded area) for rLAPuf and rLAPfd.
rLAPuf, whereas for rLAPfd the ratio was 0.34 mol of cysteines/mol
of protein. In regard to an implicit amount of rLAP molecules
that buried free cysteines as disulfide bonds (S S) (without any
consideration of thiol oxidation, improper S S bonding, or other
3.7. Effect of enzyme and zinc concentration on folding through
the activity of rLAP
degradation), an increase from 47% (for rLAPuf) to 66% (for rLAPfd
was observed (Fig. 6).
)
To evaluate the feasibility of scaling up the folding process,
different concentrations of rLAP and ZnCl2 were tested from an
enzymatic activity perspective: (1) 30 ␮M and 20 ␮M, (2) 152 ␮M
and 50 ␮M, and (3) 305 ␮M and 100 ␮M, respectively. No more than
3.6. Mass spectrometry analysis
7
U/mg of enzymatic activity was observed when the concentration
Due to the increase in the enzymatic activity after folding, we
of enzyme and ZnCl2 were augmented (Fig. 8).
considered the possibility that the enzyme was capable to self-
catalyze its 6×-His tag as suggested by other researchers [18]. To
explore this hypothesis we evaluated the enzyme by mass spec-
4
. Discussion
◦
trometry during 15 days of incubation at 15 C. As shown in Fig. 7A
4.1. Structure and activity of rLAP after folding
and B, the enzyme exhibits the same distribution of mass isoforms
at the beginning and at the end of the folding process (Fig. 7C); the
mass of each isoform is summarized in Table 3.
The enzyme activity was also determined during this incuba-
tion period (Fig. 7D); an increase in the activity was observed from
The biological activity of a protein is strongly linked to its con-
formation; thus, is expected that structural changes may have a
positive or negative impact on its functionality [10,20,30,31]. It
have been reported that N-and C-terminal propeptides are essential
for the folding of rLAP when it was expressed in E. coli [32,33]; how-
ever, we were capable of expressing a mature protein fused to the
2
.1 ± 0.1, and 8.5 ± 0.1 U/mg, at the beginning and the end of the
folding period respectively.
6
×-His tag in N-terminal, and determine the optimal folding con-
ditions by testing different buffer solutions and pH values, where
the higher activity was 31 U/mg when the protein was incubated
in phosphate buffer at pH 6.0 (Fig. 1).
The increase of the enzymatic activity at pH 6.0 is potentially
associated to the couple of two zinc atoms with the two histidines
along with other residues (mainly acidic) located in the catalytic
site through coordinate covalent bonds. At pH 6.0 the theoretical
relative amount of cationic zinc species is about 90% (as denoted by
the ion distribution as function of pH in the zinc-gluconic acid sys-
tem) [34], whereas the deprotonated imidazole groups of the his-
tidines (which have a pKa of 6.02) would be around 50%, resulting
Table 3
Isoforms molecular masses of rLAP obtained by LC–ESI-MS.
Symbol
Isoforms of rLAP
Molecular mass (Da)
A
B
C
D
E
F
G
H
Met – 6 His tag – rLAP
6 His tag – rLAP
5 His tag – rLAP
4 His tag – rLAP
3 His tag – rLAP
2 His tag – rLAP
1 His tag – rLAP
rLAP
32,828
32,665
32,528
32,391
32,253
32,116
31,995
31,880
Fig. 5. Evaluation of the reversibility of the urea-induced denaturation of rLAP. The
open squares curve shows the denaturation process of the enzyme as the concen-
tration of urea was increased, while the open circles curve shows the reversibility
of the denaturation process as the urea concentration was diluted.

4
4
A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
Fig. 7. Spectrograms obtained from the MS analyses of rLAP. Deconvoluted mass spectra at: (A) 0, and (B) 15 days of incubation; (C) relative abundance of rLAP isoforms,
each group of bars contains 0–15 days of incubation from left to right; and (D) enzyme activity monitored throughout the whole folding period.
favorable for their binding [35]. This electrostatic interaction also
explains the absence of enzymatic activity when the citrate buffer
system is used for folding, even at pH 4.0, because cationic species
of zinc (above the 50% of relative abundance) are theoretically only
present at pH lower than 3.7 under the Zn-citric acid system [36],
which is no longer useful, due to the imidazole substituent pKa.
Under the phosphate buffer conditions, the zinc speciation
behavior should be more similar to the observed in the gluconic acid
system rather than in the citric acid system, based on the phosphate
second pKa around 6.0. This also could explain the reason why the
enzymatic activity remains similar to the activity of the unfolded
protein at pH below or above 6.0, despite that the enzyme was
subjected to folding (see results in Section 3.1).
The increase in the activity of rLAP was attributable to structural
changes, which was confirmed by the comparison of the con-
formation of rLAPuf, rLAPfd and the LAPwt, through spectroscopic
techniques (Fig. 2). The dichroic response indicated that there was
a gaining in structure in rLAPfd with respect to rLAPuf. The decrease
in intensity, and the blue shift of the ꢀmax observed by IF in rLAPfd
,
was an indicator of the most folded conformation of the enzyme as
well (Fig. 2B), as have been observed in proteins that contain buried
tryptophans [11,37,38].
Disulfide bonds also play a key role in promoting and maintain-
ing a more stable conformation of proteins [4,39], due to an increase
in their thermal stability. Is well known that LAPwt has a unique
disulfide bond that relies on the hydrophobic pocket adjacent to
the active site [23]; in our study the number of established bonds
was indirectly estimated by the quantification of free cysteines. The
results showed that the increase of the enzymatic activity could
be associated to an additional 20% in the formation of disulfide
bonds in rLAPfd (Fig. 6) and consequently to a gaining in struc-
ture for the active site. On the other hand, when the protein was
not folding-processed, the level of free cysteines was higher than
in the folding-processed protein; which is an indicator of a lower
level of established disulfide bonds in rLAPuf, which stands for a
less structured protein.
The slow progression of rLAP folding (Fig. 1) might be related
to the cis/trans isomerization of the peptide bond [40–42], which
in other proteins takes to the appropriate formation of the disul-
fide bond [40], that in rLAP are adjacent to the catalytic site.
The cis-peptide bond in LAPwt is located between the Asp-117
and the Asp-118 in its folded conformation [23], although rLAP
came through different expression and purification conditions than
LAPwt, it also contains this particular sequence, thus it is possible
Fig. 8. Curves of enzymatic activity of rLAP under different folding conditions: (1)
circles represent 3 ␮M rLAP, and 10 ␮M ZnCl2; (2) squares 30 ␮M rLAP, and 10 ␮M
ZnCl2; (3) triangles 152 ␮M rLAP, and 10 ␮M ZnCl2; (4) diamonds 305 ␮M rLAP, and
10 ␮M ZnCl2; (5) asterisks 30 ␮M rLAP, and 20 ␮M ZnCl2; (6) stars 152 ␮M rLAP, and
0 ␮M ZnCl2; and (7) cross 305 ␮M rLAP, and 100 ␮M ZnCl2.
5

A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
45
that its folding takes longer due to the slow establishment of a cor-
5. Conclusion
rect cis/trans configuration followed by the establishment of the
disulfide bond; as have been reported for cytrochrome c whose fast
or slow folding are determined by already present specific motifs
in its molten globule state [12].
The results obtained so far helped us to understand the factors
that impact on the folding process, and consequently to improve the
downstream process toward a more stable and functional enzyme
for industrial purposes. On this regard, the effect of the enzyme and
ZnCl2 concentrations during folding evaluated through the enzy-
matic activity showed that the increase in the concentration of
the enzyme reduces the folding yield, even if we increased ZnCl2
concentration (Fig. 8).
The collective information obtained from the spectroscopic
and thermodynamic techniques, along with the enzymatic assays
employed on this study, are valuable for describing the folding
processing of rLAP, and the way that pH favored the coupling of
zinc to the catalytic site. These findings allowed the establishment
of a characterization approach for a better understanding of the
folding process in bimetallic proteins. Hence, the conditions of fold-
ing for rLAP were optimized with respect to our previous study,
since we obtained a recombinant enzyme with a higher activity
resulting from a better structural stability associated to the fold-
ing treatment. In this sense, orthogonal techniques represent an
advantageous approach for distinguishing phenomena that occur
at different molecular levels during the folding process of a protein
that could be helpful for a better understanding of the processes
toward their improvement.
4.2. Increased thermal stability of rLAP
It is well known that thermal stability of a protein is related to
its level of covalent and non-covalent bonds that maintain an orga-
nized structure [43–45]. Herein, we observed a positive effect of
folding in the thermodynamic stability of rLAP that increased its
Tm, (Fig. 3, and Table 1) which is an indicator of a higher number of
hydrogen bonds and disulfide bonds, also consistent with the free
cysteines analysis (Fig. 6). The results obtained by DSC suggest that
Acknowledgements
The authors thank to Antonio Hernández-García, Maribel
Jardón-Castillo, and Mario E. Abad-Javier, for their technical
contribution in the cysteines quantification, fluorescence, and
calorimetric techniques at the Research and Development Unit of
PROBIOMED S.A. de C.V.
◦
the Tm found at 54 C in rLAP corresponds to a residual amount
fd
◦
of rLAPuf, whereas the transitions at 85 and 88 C in rLAP corre-
fd
spond to a more folded protein that increases its ration after folding
(
Fig. 3A, and Table 1). In addition, IF confirms a higher stability of
References
◦
rLAPfd up to 85 C when a red displacement of ꢀmax was observed
(
Fig. 3B) which is consistent with previous IF responses in other
[1] T. Dingermann, Biotechnol. J. 3 (2008) 90–97.
[
[
[
2] K. Graumann, A. Premstaller, Biotechnol. J. 1 (2006) 164–186.
3] A. Villaverde, M. Mar Carrió, Biotechnol. Lett. 25 (2003) 1385–1395.
4] R. Rudolph, H. Lilie, FASEB J. 10 (1996) 49–56.
protein studies [46–48]. For rLAP IF revealed that the folding pro-
cess occurs below 40 C, based on the blue shift of ꢀmax, whereas the
uf
◦
◦
◦
thermal denaturation takes place from 40 C until 60 C. Unlike to IF,
DSC has a lower sensitivity to detect gradual molecular transitions
of subtle exothermic enthalpies during the establishment of hydro-
gen bonds; however, it is capable to detect the single endothermic
transition that comprises the rupture of all the non-covalent inter-
actions.
[5] E. De Bernardez Clark, Curr. Opin. Biotechnol. 9 (1998) 157–163.
[
[
6] L. Grillberger, T.R. Kreil, S. Nasr, M. Reiter, Biotechnol. J. 4 (2009) 186–201.
7] E.M. Gorman, B.E. Padden, E.J. Munson, S.C. Gad, Stability: physical and
chemical, in: Pharmaceutical Sciences Encyclopedia, John Wiley & Sons, Inc.,
2010.
[
[
8] C. Tanford, M.L. Anson, J.T. Edsall, C.B. Anfinsen, M.R. Frederic, Protein denatur-
ation, in: Advances in Protein Chemistry, Academic Press, 1968, pp. 121–282.
9] K. Hejnaes, F. Matthiesen, L. Skriver, Protein stability in downstream processing,
in: Bioseparation and Bioprocessing, Wiley-VCH Verlag GmbH, 2008, pp. 31–65.
[
[
10] A. Fersht, Structure and Mechanism in Protein Science, 2nd ed., W. H. Freeman
and Company, New York, 1999.
11] J.R. Lakowicz, Fluorescence sensing, in: Principles of Fluorescence Spec-
troscopy, Springer US, 2006, pp. 623–673.
4
.3. Chemical stability of rLAP
Urea was employed as chaotropic agent to denature rLAPuf
[
[
12] T.R. Sosnick, L. Mayne, R. Hiller, S.W. Englander, Struct. Biol. 1 (1994) 149–156.
13] K.A. Dill, J.L. MacCallum, Science 338 (2012) 1042–1046.
and rLAPfd to compare their chemical stability at different stress
levels (0–8 Murea) (Fig. 4); rLAPuf exhibited a wider range of ꢁ
between the initial and the final levels of the denaturation curve
[14] A.G. Szabo, D.M. Rayner, J. Am. Chem. Soc. 102 (1980) 554–563.
[15] M.Y. Berezin, S. Achilefu, Chem. Rev. 110 (2010) 2641–2684.
[
16] J.R. Lakowicz, M.R. Eftink, Intrinsic fluorescence of proteins, in: Topics in Fluo-
rescence Spectroscopy, Springer US, 2002, pp. 1–15.
(
2.23–2.90 ns), with respect to rLAPfd (ꢁ from 1.97 to 2.17 ns). On
the other hand, the red displacement of ꢀmax after denaturation is
higher in rLAPuf (331–345 nm) than in rLAPfd (330–333 nm), which
was chemically more stable.
[
17] G. Pérez-Sánchez, L.I. Leal-Guadarrama, I. Trelles, N.O. Pérez, E. Medina-Rivero,
Process Biochem. 46 (2011) 1825–1830.
[
[
[
18] M. Hartley, B. Bennett, Protein Expr. Purif. 66 (2009) 91–101.
19] R.C. Holz, Coord. Chem. Rev. 232 (2001) 5–26.
As part of the extended characterization of rLAP, we observed
that urea-induced denaturation of rLAPuf was a reversible process,
since the protein recovered its enzymatic activity as the urea was
diluted (Fig. 5). These results demonstrated that once rLAP is pro-
duced as insoluble inclusion bodies in E. coli, it would be feasible to
increase its enzymatic activity through folding, even after stressful
solubilization procedures that are typically employed for harvest-
ing the protein from insoluble inclusion bodies.
20] A. Taylor, J. Fed. Am. Soc. Exp. Biol. 7 (1993) 290–298.
[21] M. Matsui, J.H. Fowler, L.L. Walling, Biol. Chem. 387 (2006) 1535–1544.
[
22] A.V. Hernández-Moreno, F. Villase n˜ or, E. Medina-Rivero, N.O. Pérez, L.F. Flores-
Ortiz, G. Saab-Rincón, G. Luna-Bárcenas, Int. J. Biol. Macromol. 64 (2014)
306–312.
[23] B. Chevrier, C. Schalk, H. D’Orchymont, J.-M. Rondeau, D. Moras, C. Tarnus,
Structure 2 (1994) 283–291.
[
24] K.P. Bzymek, S.I. Swierczek, B. Bennett, R.C. Holz, Inorg. Chem. 44 (2005)
574–8580.
8
[25] J.R. Lakowicz, Time-domain lifetime measurements, in: Principles of Fluores-
cence Spectroscopy, Springer US, 2006, pp. 97–155.
[
[
[
26] V. Kayser, N. Chennamsetty, V. Voynov, B. Helk, B. Trout, J. Fluoresc. 21 (2011)
275–288.
27] C. Zhang, C. Rodriguez, M. Circu, T. Aw, J. Feng, Anal. Bioanal. Chem. 401 (2011)
4
.4. Molecular mass distribution of rLAP during folding
2165–2175.
The relative abundance of all the molecular masses isoforms
28] G.T. Hermanson, Bioconjugate Techniques, 2nd ed., 2008, San Diego, CA, USA.
of rLAP was maintained across the folding process (Fig. 7C, and
Table 3). There was no evidence that the enzyme was capable of
self-catalyze; hence, the increase in the enzymatic activity (Fig. 7D)
is more likely to be associated to structural changes as have been
discussed throughout this study.
[29] C. Guenet, P. Lepage, B.A. Harris, J. Biol. Chem. 267 (1992) 8390–8395.
[
30] T.E. Creighton, Proteins. Structures and Molecular Properties, 2nd ed., W. H.
Freeman and Company, New York, 1993.
31] J.R. Lakowicz, S. D’ Auria, M. Rossi, I. Gryczynski, Fluorescence of Extreme ther-
mophilic proteins, in: Topics in Fluorescence Spectroscopy, Springer US, 2002,
pp. 285–306.
[

4
6
A.V. Hernández-Moreno et al. / Journal of Molecular Catalysis B: Enzymatic 113 (2015) 39–46
[
[
[
[
32] Z.-Z. Zhang, S. Nirasawa, Y. Nakajima, M. Yoshida, K. Hayashi, Biochemistry 350
2000) 671–676.
33] S. Nirasawa, Y. Nakajima, Z.-Z. Zhang, M. Yoshida, K. Hayashi, Biochemistry 341
1999) 25–31.
34] M. Dekker, Handbook of Metal–Ligand Interaction in Bilogic Fluids – Bioinor-
ganic Medicine, 1995, New York.
[40] F.X. Schmid, Ann. Rev. Biophys. Bimol. Struct. 22 (1993) 123–143.
[41] W.J. Wedemeyer, E. Welker, H.A. Scheraga, Biochemistry 41 (2002)
14637–14644.
[42] K. Lang, F.X. Schmid, G. Fischer, Nature 329 (1987) 268–270.
[43] P.L. Privalov, Stability of proteins, in: Advances in Protein Chemistry, Academic
Press, 1982, pp. 1–104.
(
(
35] L. Zhou, S. Li, Y. Su, X. Yi, A. Zheng, F. Deng, J. Phys. Chem. B 117 (2013)
[44] G. Privalov, V. Kavina, E. Freire, P.L. Privalov, Anal. Biochem. 232 (1995)
79–85.
8
954–8965.
[
[
[
[
36] R.B. Martin, Antimicrob. Agents Chemother. 32 (1988) 608–609.
37] M.R. Eftink, C.A. Ghiron, Biochemistry 15 (1976) 672–680.
38] J.T. Vivian, P.R. Callis, Biophys. J. 80 (2001) 2093–2109.
39] C. Tanford, J.T. Edsall, C.B. Anfinsen, M.R. Frederic, Protein denaturation. Part C:
Theoretical models for the mechanism of denaturation, in: Advances in Protein
Chemistry, Academic Press, 1970, pp. 1–95.
[45] S.P. Manly, K.S. Matthews, J.M. Sturtevant, Biochemistry 24 (1885)
3842–3846.
[46] A.I. Dragan, P.L. Privalov, J. Mol. Biol. 321 (2002) 891–908.
[47] T.Q. Faria, J.C. Lima, M. Bastos, A.L. Ma c¸ anita, H. Santos, J. Biol. Chem. 279 (2004)
48680–48691.
[48] M. Weichel, S. Bassarab, P. Garidel, Bioprocess Tech. 6 (2008) 42–45.