Coaxial electrospun fibermat of poly(AM/DAAM)/ADH and PCL: Versatile platform for functioning active enzymes

Article abstract of DOI:10.1246/BCSJ.20200131

In this study, we prepared and characterized enzyme (αchymotrypsin or lactase)-encapsulating core-shell fibermats by electrospinning. The hydrophilic copolymer of acrylamide (AM) and diacetone acrylamide (DAAM), poly(AM/DAAM), was used as the base materia

Full text of DOI:10.1246/BCSJ.20200131

Advance Publication Cover Page
Coaxial Electrospun Fibermat of Poly(AM/DAAM)/ADH and PCL: Versatile
Platform for Functioning Active Enzymes
Yuji Tanikawa, Yuya Ido, Ren Ando, Akiko Obata, Kenji Nagata,
Toshihiro Kasuga, and Toshihisa Mizuno*
Advance Publication on the web June 11, 2020
doi:10.1246/bcsj.20200131
© 2020 The Chemical Society of Japan
Advance Publication is a service for online publication of manuscripts prior to releasing fully edited, printed versions. Entire
manuscripts and a portion of the graphical abstract can be released on the web as soon as the submission is accepted. Note that the
Chemical Society of Japan bears no responsibility for issues resulting from the use of information taken from unedited, Advance
Publication manuscripts.

Coaxial Electrospun Fibermat of Poly(AM/DAAM)/ADH and PCL: Versatile
Platform for Functioning Active Enzymes
Yuji Tanikawa,1 Yuya Ido,1 Ren Ando,1 Akiko Obata,1 Kenji Nagata,1 Toshihiro Kasuga,1 and Toshihisa Mizuno*1, 2
1Department of Life Science and Applied Chemistry, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso-cho, Showa-
ku, Nagoya, Aichi 466-8555, Japan
2Department of Nanopharmaceutical Sciences, Graduate School of Engineering, Nagoya Institute of Technology, Gokiso-cho Showa-ku,
Nagoya, Aichi 466-8555, Japan
E-mail: toshitcm@nitech.ac.jp
Toshihisa Mizuno
Toshihisa Mizuno received PH.D. Degree from Kyushu University in 2000. He is currently an Associate Professor in the
Department of Life and Applied Chemistry, Graduate School of Engineering, Nagoya Institute of Technology. His
research interests include design of biomacromolecules and materials, having biological functions.
Abstract
In this study, we prepared and characterized enzyme (α-
enough mechanical strength) of nanofibrous scaffold structure in
aqueous solvents, choice of hydrophobic polymers for the base
materials of fibermats seems a reasonable strategy. However,
mixing with enzyme molecules in organic solvents for preparing
the precursor solution for electrospinning and/or further direct
contact of enzyme molecules with hydrophobic polymers in the
obtained fibermat nanofibers would cause serious damages in
enzymatic functions. Therefore, several sophisticated
preparation methods, compatible with maintenance of
nanofibrous scaffold structure in aqueous solvents and decrease
in damages to enzymatic functions, have been studied so far.
One of the methods able to prepare the water-insoluble
fibermats, encapsulating proteins with less damages, is the
emulsion electrospinning method.5 In this method, hydrophobic
polymers such as poly(lactic acid) (PLA) are used as the base
materials and an emulsion solution, formed by the hydrophobic
polymer in organic solvent and protein in aqueous buffer, was
used for electrospinning. To reduce the protein structural
damage and activity loss, Ca alginate,5 methyl cellulose,8 or
gelatin2 is supplementary added to the aqueous phases of
precursor emulsion solution. The second one is the post-
crosslinking method, in which post crosslinking treatments were
applied to the beforehand-prepared water-soluble protein-
encapsulated fibermats.9 Several post-treatments have been
studied to obtain water-insoluble fibermats including chemical
crosslinking using bifunctional crosslinkers such as
glutaraldehyde,10 photoirradiation11 or thermal treatments12 in
the presence of corresponding crosslinkers. However, to reduce
damage to the encapsulated proteins, careful choice of reaction
conditions is necessary.
chymotrypsin or lactase)-encapsulating core-shell fibermats by
electrospinning. The hydrophilic copolymer of acrylamide (AM)
and diacetone acrylamide (DAAM), poly(AM/DAAM), was
used as the base material to obtain the core unit of nanofibers.
During electrospinning, poly(AM/DAAM) was crosslinked with
the bifunctional crosslinker adipic acid dihydrazide (ADH) in
the presence of enzyme molecules. The cores were wrapped with
hydrophobic poly(ε-caprolactone) (PCL) layers as shell unit.
Different from the fibermats of only poly(AM/DAAM)/ADH,
the core-shell fibermat of poly(AM/DAAM)/ADH and PCL
exhibited enough mechanical strength and stability of the
stacked nanofibrous structure in
a neutral-pH buffer.
Furthermore, when the PCL-shell thickness was controlled to be
less than 150 nm, the encapsulated enzymes exhibited an
apparent activity of >70–80% for low-molecular weight
substrates in an immersion buffer. These results indicate that the
core-shell fibermats of poly(AM/DAAM)/ADH and PCL (or
other hydrophobic polymer) could be used as effective enzyme-
immobilizing platforms.
Keywords: Fibermat, Co-axial electrospinning,
Protein-encapsulation
1. Introduction
The fabrication of electrospun fibermats, composed of
nanofibers that can encapsulate functional biomacromolecules
such as peptides, proteins, or nucleic acids, without any loss of
the original biological functions, has attracted considerable
research attention.1 This is because biomacromolecule -
encapsulating fibermats are considered favorable materials for
designing cell-incubation scaffolds that allow control of cell
proliferation and differentiation,2,3 protein-drug delivery
systems with controlled-release properties,4-7 and enzyme-
immobilizing supports.4 Aiming at application to enzyme-
immobilizing supports, the enzyme-encapsulating fibermats
need retain the nanofibrous scaffold structure even after being
immersed in aqueous buffer for a certain period, since it is
indispensable to exhibit efficient enzymatic functions even in
nanofibers. While, to increase stability (water-insolubility and
As the other new method to prepare enzyme-encapsulating
fibermats, we recently reported the in-situ crosslinking during
electrospinning (SCES) method. In this method, post-
crosslinkable polymers and the corresponding bifunctional
crosslinker were used.13-15 Because the spinning step of
nanofibers via electrospinning includes solvent volatilization
process, we expected effective in situ crosslinking between post-
crosslinkable polymers and the corresponding bifunctional
crosslinker via dehydration condensation reaction during
electrospinning. As result, by using the precursor aqueous
solution
of
poly(γ-glutamate)
(γ-PGA)
and
3-
1

glycidyloxypropyltrimethoxysilane (GPTMS), or the copolymer
of acrylamide (AM) and diacetone acrylamide (DAAM)
poly(AM/DAAM) and adipic acid dihydrazide (ADH), we
successfully obtained water-insoluble protein-encapsulated
fibermats from the precursor aqueous solution of water-soluble
polymers and proteins. Since we could prepare the protein-
encapsulating fibermats only use of aqueous solvents, we
successfully encapsulated the green fluorescent protein (GFP)
and α-chymotrypsin (α-Chy) without denaturation.15
Furthermore, because of the subtle porous structure of the
polymeric network formed by γ-PGA/GPTMS and
poly(AM/DAAM)/ADH, high-molecular weight (MW) enzyme
proteins could be hold in nanofibers without leakage but low-
MW substrate molecules could permeate the nanofibers. As
results, the enzymatic activities of the encapsulated α-Chy were
>85% of that of α-Chy (same amount) dissolved in a buffer.
These findings showed that the protein-encapsulating fibermats
fabricated using the SCES method could be potentially used for
new type of methods to prepare the enzyme-encapsulating
fibermats. However, especially in case of poly(AM/DAAM)
/ADH fibermats, the fibermats showed low mechanical strength
and fragile, insufficient for practical use as enzyme-
immobilizing substrates. Further, because the covalent bonding
between the ketone groups in poly(AM/DAAM) and the
hydrazide groups in ADH is reversible, gradual fusion of
adjacent nanofibers in the obtained fibermats was observed after
long-term immersion in the buffer, which caused a reduction in
the total surface area of the nanofibers against immersing
solvents. Consequently, the activity of the encapsulated enzyme
decreased over time.
In this study, to overcome the two aforementioned
drawbacks of poly(AM/DAAM)/ADH fibermat, we modified it
to the core-shell type fibermats,16 wherein the protein-
encapsulating poly(AM/DAAM)/ADH nanofibers were
wrapped with thin hydrophobic layers of poly(ε-caprolactone)
(PCL). Enhancement in stability of hydrophilic nanofibers by
wrapping with a hydrophobic polymer, yielding core-shell
fibermats was reported, so it should be one of the rational
solutions.17 Application of core-shell fibermats to a protein (such
as BSA, growth factor proteins, etc.)-loadable/releasable
substrate was also reported.18 However, enzyme encapsulation
and detail evaluation of their biological activities was less
reported. In this study, therefore we prepare the enzyme-
use in reversible addition–fragmentation chain transfer (RAFT)
polymerization.19 α-Chy, Lac, PCL (number-average MW (Mn)
= 80,000), and trypsin-chymotrypsin inhibitor (TCI) from
Glycine max (soybean)20 were purchased from Sigma-Aldrich
(USA). Poly(AM/DAAM) featuring AM and DAAM at a molar
ratio of 8:2 was synthesized as described; in our previous paper,
poly(AM/DAAM) is described in the place of PAM-co-
PDAAM.15 The Mn and polydispersity index (PDI) of
poly(AM/DAAM) were determined using gel-permeation
chromatography (GPC).
Construction of Core-Shell Fibermats with/without
Encapsulated Proteins or Fluorescent Molecules. ADH (0.5
molar equiv. with respect to the DAAM unit in
poly(AM/DAAM)) was added to
a
solution of
poly(AM/DAAM) (0.5 g in 2.5 mL of 100 mM phosphate buffer,
pH 8) and transferred into a Luer lock syringe (5 mL). When
encapsulating proteins, such as GFP, α-Chy, or Lac, 5 wt% of
the protein of interest (with respect to the mass of the polymer)
was added to the solution. In the case of fluorescent molecules
such as fluorescein and rhodamine B, 0.4 mmol was added to the
solution. Concomitantly, 8 wt% PCL solution in TFE was
prepared and transferred into another Luer lock syringe (5 mL).
The two syringes were then placed in different syringe pumps
fitted with a linear actuator (KDS-100, KD Scientific, USA) and
connected to a coaxial spinneret (MECC Co. Ltd, Japan) with 27
G
needle (Terumo Corp., for the core solution of
poly(AM/DAAM)/ADH) through PTFE tube. The
a
aforementioned solutions were electrospun (SD-02, MECC Co.
Ltd, Japan) at a linear extrusion velocity of 0.1 mL/h for the
poly(AM/DAAM)/ADH solution and 0.4 mL/h for the PCL
solution under high voltage (25 kV). The core component of the
obtained fibermat was poly(AM/DAAM)ADH and the shell
component was PCL. The distance between the tip of the
spinneret and the grounded collector (aluminum plate, 150 × 200
mm) was set to 150 mm.
Scanning Electron Microscopy (SEM). Samples were coated
with amorphous osmium through plasma chemical vapor
deposition by using a JEE-420T vacuum evaporator (JEOL,
Japan). Sample morphologies were examined using field-
emission SEM (JSM-6301F, JEOL, Japan). The mean diameters
of the fibermat nanofibers and their standard deviations were
evaluated from the SEM images of 30 nanofibers, using the
software, Image J.
encapsulating core-shell fibermats using
a
co-axial
Transmission Electron Microscopy (TEM). Samples for TEM
observation were prepared by directly collecting the spun core-
shell nanofibers from the coaxial spinneret onto a TEM grid
(Formvar Carbon Film on Copper 100 mesh (50), Okenshoji Co.,
Ltd, Japan). To obtain adequate image contrast between core and
shell portions, sodium phosphotungstate was mixed into the core
precursor solution of poly(AM/DAAM) and ADH at a final
concentration of 0.001% (w/v). TEM images were acquired
using a JEM-z2500 instrument (JEOL, Japan), under an
acceleration voltage of 200 kV.
Attenuated Total-Reflectance Fourier-Transform Infrared
(ATR-FTIR) Spectroscopy. ATR-FTIR spectra were acquired
using an FT-IR-4000 spectrometer (JASCO, Japan) equipped
with an ATR PRO450-S unit (JASCO, Japan). The FID spectra
scanned 200 times at ambient temperature were accumulated and
Fourier-transformed to obtain FTIR spectra at 4 cm-1 resolution.
Release Behavior of Encapsulated GFP from Core-Shell
Fibermats under Different pH Conditions. Stability of core-
shell fibermats upon immersion in buffers of different pH was
evaluated based on the release behavior of the encapsulated GFP
from the core unit of the fibermats. GFP-loaded fibermats (2 mg)
were placed in a 10-mL glass vial and subsequently immersed in
1.5 mL of 100 mM acetate buffer (pH 3–5), 100 mM phosphate
electrospinning technique and examined not only whether the
wrapping of poly(AM/DAAM)/ADH nanofibers with PCL
could effectively overcome the aforementioned drawbacks but
also the relationship between the shell thickness and the apparent
enzymatic reactivity of the encapsulating enzyme molecules.
2. Experimental
Materials. Unless stated otherwise, all chemicals and reagents
were commercially obtained and used without further
purification. Tris(hydroxymethyl)aminomethane (Tris), p-
nitrophenyl
acetate
(p-NAc),
p-nitrophenyl-β-D-
glucopyranoside (p-NGlu), acrylamide (AM), 2,2,2-
trifluoroethanol (TFE), 2,2'-azobis[N-(2-carboxyethyl) -2-
methylpropionamidine] tetrahydrate (VA-057), disodium
phosphate (Na2HPO4), sodium dihydrogen phosphate
(NaH2PO4), fluorescein, and rhodamine B were purchased from
Wako Pure Chemical Ind. Ltd. (Japan). 100 mM phosphate
buffer (pH 8) was prepared by mixing 100 mM Na2HPO4 and
100 mM NaH2PO4. Diacetone acrylamide (DAAM) and adipic
acid dihydrazide (ADH) were purchased from Tokyo Chemical
Industry Co., Ltd. (Japan). DAAM was recrystallized several
times from ethyl acetate to remove the inhibitor mequinol before
2

buffer (pH 6–8), or 100 mM carbonate buffer (pH 9–10) for
defined periods (0–300 min) at room temperature; the fibermats
were then removed from the buffer and the UV-vis absorbance
(A) of the remaining solution was measured (V-670, JASCO,
Japan). To quantify the amount of GFP released in pH 6–10
samples, A485, which is characteristic of GFP, was measured; for
the pH 3–5 samples, the GFP released in the supernatant was
quantified using the Bradford method.21
Tensile-Strength Measurement of Fibermats. For testing
tensile strength, fibermats were cut into pieces of a defined size
(20 mm long, 5 mm wide) and set between the sensor tips of the
testing machine (Autograph, AGS-G, Shimadzu, Japan)
equipped with a 50 N load cell according to Japanese Industrial
Standard (JIS L 1015). Mechanical testing was performed on the
following condition; span length 20 mm, crosshead speed 1 mm
min-1, n = 5.
containing 5.56 mM p-NGlu and allowed to react for 15 min on
50 °C. After terminating the reaction by adding 4 mL of 100 mM
carbonate buffer (pH 8) to the supernatant, A404 was measured to
determine the amount of p-nitrophenol produced (as described
above for α-Chy samples). As controls, 100 mM phosphate
buffer (pH 8) containing α-Chy in amounts corresponding to
those in 2.0, 4.0, or 8.0 mg of α-Chy-encapsulating fibermats or
25 mM Mcllvaine buffer (pH 4) containing Lac in amounts
corresponding to those in 2.0, 4.0, or 8.0 mg of Lac-
encapsulating fibermats were used to determine enzymatic
activities by using p-NAc and p-NGlu, respectively.
3. Results and Discussion
Preparation of Core-Shell-Type Fibermats by Using a
Coaxial Spinneret.
Contact-Angle Measurement of Fibermats. Fibermats were
stuck onto a coverslip (18 × 18 mm) and placed on the sample
plate of a contact-angle measuring instrument (SImage mini7,
Excimer Inc., Japan). A 5-μL droplet of water was placed on the
fibermats and the shapes of the hemispherical water droplet were
observed using video-image capture and the contact angles were
determined; 10 samples per group were tested to determine
average contact angles and standard deviations.
Fluorescence Spectra and Emission Properties of GFP -
Encapsulating Core-Shell Fibermats. First, GFP-
encapsulating core-shell fibermats were wetted with 20 mM
Tris-HCl (pH 7.8) and adhered to a coverslip (20 × 20 mm, Eagle
XG, Corning Inc., USA). The coverslip was then placed into the
Core-shell fibermats were constructed using two
separately prepared precursor solutions, poly(AM/DAAM)
/ADH in phosphate buffer (pH 8) and PCL in TFE, which were
spun through a coaxial spinneret (Figure 1(a)) under constant
voltage (details in Experimental Section). Several combinations
of polymer species can be used to construct core-shell
fibermats,24,25 but combination of poly(AM/DAAM)/ADH
and PCL was not reported yet. The chemical structures of
poly(AM/DAAM), ADH, and PCL are illustrated in Figure 1(b).
Poly(AM/DAAM)
was
synthesized
through
RAFT
polymerization26 of AM and DAAM by using a RAFT agent.
(a)
sample chamber of
a
FluoroMax-4 spectrofluorometer
(HORIBA Scientific, Japan), and the fluorescence emission
spectrum was acquired at a photoexcitation wavelength of 460
nm. To obtain fluorescence images of GFP-encapsulating
fibermats, the cover slips were wetted in 100 mM phosphate
buffer (pH 8) and then exposed to UV light (320 nm).
Leakage/Release of Encapsulated Fluorescent Molecules
upon Immersion in Buffer. Loaded fibermats (2 mg) were
placed in a 10-mL glass vial and immersed in 1.5 mL of 100 mM
phosphate buffer (pH 8) for defined periods (0–6 h) at room
temperature, after which the fibermats were removed from the
buffer and the A values of the remaining solutions were
measured (V-670, JASCO, Japan); specifically, A494
(characteristic of fluorescein) and A554 (characteristic of
rhodamine B) were measured to determine the degree of release
of the fluorophores during the assay period.
(b)
O
H
N
O
NH2
O
NH
O
NH2
n
m
H2N
N
H
O
poly(AM/DAAM)
ADH
O
Enzymatic Activities of Enzyme-Encapsulating Core-Shell
Fibermats. Core-shell fibermats encapsulating α-Chy or Lac of
2.0, 4.0, or 8.0 mg were immersed in 1.5 mL of 100 mM
phosphate buffer (pH 8) for the α-Chy-encapsulated fibermat or
in a 50 mM Mcilvaine buffer (pH 4)22 for the Lac-encapsulated
fibermat in 10 mL glass vial for 1 h. After removal from the
buffer solution, the α-Chy-encapsulating fibermats were briefly
washed thrice with 100 mM phosphate buffer and placed in a 10-
mL glass vial containing 3.0 mL of 100 mM phosphate buffer
(pH 8); to each of these samples, 20 μL of p-NAc in DMSO (60
mM) was added and allowed to react for 10 min. 50 mM
Mcilvaine buffer was prepared by mixing 50 mM Na2HPO4 and
50 mM citric acid. To observe the enzymatic reaction of α-Chy
in the presence of TCI, TCI was added (to a final concentration
of 17.5 μM) before addition of p-NAc. After removal of the
fibermats, A404, which is characteristic of deprotonated p-
nitrophenol, was measured to determine the amount of p-
nitrophenol produced based on its molar extinction coefficient
(18,384 mol cm-1 dm-3)23 at pH 8. In case of Lac-encapsulating
fibermats, after removal from the buffer solution, the fibermat
samples were placed in 25 mM Mcllvaine buffer (pH 4)
O
n
PCL
Figure 1. (a) Coaxial spinneret for fabrication of core-shell
fibermats of poly(AM/DAAM)/ADH and PCL. (b) Chemical
structures of poly(AM/DAAM), ADH, and PCL.
In the synthesized poly(AM/DAAM), the molar ratio of AM to
DAAM was set to be 8:2, assigning from 1H-NMR measurement,
and the Mn and PDI were determined to be 52,000 and 1.32 from
GPC analysis. The Mn of commercially available PCL was
80,000. The poly(AM/DAAM)/ADH solution was prepared
right before electrospinning by adding ADH to poly(AM/
DAAM) in phosphate buffer (pH 8). According to the
progression of the crosslinking reaction between the ketone
groups in poly(AM/DAAM) and the hydrazide groups in ADH,
the viscosity of the solution gradually increased, but the
electrospinning could be completed before the solution
transformed into a stiff hydrogel (within 1.5 h). Notably, solvent
volatilization (i.e., dehydration) during electrospinning can
3

facilitate the crosslinking reaction between poly(AM/DAAM)
and ADH. Thus, even in the absence of the PCL layer, the spun
nanofibers could be stacked without fusion of adjacent
nanofibers, and after further progression of the crosslinking
reaction within each nanofiber, the obtained fibermats became
water-insoluble. The feed rates of the poly(AM/DAAM)/ADH
solution and PCL were fixed respectively at 0.1 and 0.4 mL/h
(1:4 ratio); we also tested other spin conditions featuring distinct
feed-rate ratios, but the core-shell fibermats could be
successfully prepared only when this ratio was 1:3 – 1:4.
Therefore, we selected the feed-rate ratio of 1:4 for the
construction of the core-shell fibermats in this study.
different, but the estimated ratio of the core diameter and the
shell thickness should be similar to those of core-shell fibermats
collected onto a plate-collector.
We next used ATR-IR measurements to further
characterize the core-shell fibermats (Figure 4). In our previous
study on the poly(AM/DAAM)/ADH fibermat, according to the
formation of hydrazone bonds between the ketone groups in
poly(AM/DAAM) and the hydrazide groups in ADH, IR bands
emerged at 1540 and 1615 cm-1 in addition to the IR band for the
amide bond in AM and DAAM at 1650 cm-1.15 Since similar IR
bands were also observed for the core-shell fibermat in Figure 4,
progress of crosslinking reaction between the ketone groups in
poly(AM/DAAM) and the hydrazide groups in ADH was also
occurred. With the core-shell fibermats, besides the
aforementioned IR bands, we detected the coexistence of other
IR bands, at 1710 and 1180 cm-1, corresponding to the ester
bonds in PCL.27 The coexistence of these IR bands also
suggested successful construction of the target core-shell
fibermats of poly(AM/DAAM)/ADH and PCL.
Figure 2. SEM images of the core-shell fibermats of poly(AM/
DAAM)/ADH and PCL (a) before and (b) after immersion in
50 mM phosphate buffer (pH 8) for 1 h, and of (c)
poly(AM/DAAM)/ADH fibermats after immersion in the buffer.
Acceleration voltage, 10 kV.
Figure 3. TEM images of (a) poly(AM/DAAM)/ADH fibermat,
(b) PCL fibermat, and (c) the core-shell fibermat of poly(AM/
DAAM)/ADH and PCL. Acceleration voltage, 200 kV.
SEM and TEM images of the prepared core-shell fibermats
are shown in Figure 2(a) and Figure 3(c), respectively. To obtain
sufficient image contrast of the core unit against the shell unit in
TEM images, we added a small amount (0.001 wt% against
poly(AM/DAAM)) of sodium phosphotungstate to the precursor
solution of the core unit before electrospinning. As a reference,
we also performed TEM measurements of poly(AM/DAAM)
/ADH fibermats and PCL fibermats (prepared using only PCL)
(Figure 3(a), (b)). As indicated by the SEM images in Figures
2(a), we observed successful accumulation of isolated
nanofibers featuring an average diameter of 473 ± 88 nm.
Furthermore, a double-layered structure, consisting of the
independent core and shell layers, was clearly observed at the
level of the nanofibers of core-shell fibermats (Figure 3(c)). In
the reference poly(AM/DAAM)/ADH fibermats and the PCL
fibermats, this double-layered structure was not detected in TEM
images (Figure 3(a), (b)), which strongly suggested that core-
shell fibermats consisting of poly(AM/DAAM)/ADH and PCL
were successfully constructed. By using the TEM images, we
calculated the ratio of the average diameter (nm) of the core unit
to the thickness (nm) of the shell unit to be 3.69 ± 0.81. The fiber
diameters in TEM samples appeared smaller than those in SEM
samples, but this can be explained according to the method used
to prepare TEM samples; in order to get the TEM sample of
single nanofibers, we collected nanofibers at the interspace
between the spinneret needle and plate-collector to shorten the
collecting time of nanofibers. Due to difference of preparation
procedure the average diameter of the observed nanofibers was
Figure 4. ATR-IR spectra of poly(AM/DAAM)/ADH fibermats
(blue line), PCL fibermats (green line), and core-shell fibermats
(red line).
Measurement of the contact angle of water droplets is
another effective method to verify PCL wrapping of the core unit
of poly(AM/DAAM)/ADH in core-shell fibermats. If the
poly(AM/DAAM)/ADH core is successfully wrapped with the
PCL layer without any defect, the obtained core-shell nanofibers
would be expected to show altered hydrophobicity, originating
from the hydrophobic nature of PCL. Water droplets spotted on
top of fibermats and the calculated average contact angles are
shown in Figure 5 and Table 1, respectively. Reflecting the
hydrophilic nature of poly(AM/DAAM)/ADH, the contact angle
of the poly(AM/DAAM)/ADH fibermat was first measured to be
17° ± 6.3°, and according to adsorption of water, this value
decreased further. By contrast, the contact angle of the core-shell
fibermat was observed to be 119° ± 3.5° with no change, which
is comparable to that of PCL fibermats (125° ± 1.6°). These data
also suggested successful preparation of the core-shell fibermats
of poly(AM/DAAM)/ADH and PCL.
Because of the inherent properties of PCL, wrapping of the
nanofibers of poly(AM/DAAM)/ADH with PCL (i.e.,
preparation of the core-shell fibermats) was expected to
4

substantially enhance the mechanical strength of the fibermats
against tensile forces. Therefore, we next examined maximum
load value (N) under the application of tensile strain (%) to
Table 2. Results of Tensile Tests of Fibermats
Maximum
Elongation to
failure (%)
Fibermat sample
load/N
PCL fibermat
3.21 ± 0.14
112 ± 8.2
Poly(AM/DAAM)/ADH
fibermat
Core-shell fibermat
Not obtained
(extremely fragile)
0.67 ± 0.04 133 ± 5.3
Characterization of GFP-Encapsulating Core-Shell
Fibermats.
Figure 5. Images showing water droplets placed onto (a) core-
shell fibermat, (b) poly(AM/DAAM)/ADH fibermat, and (c)
PCL fibermat.
By PCL wrapping (i.e. modifying the original
poly(AM/DAAM) fibermat to the core-shell fibermat with PCL),
we successfully enhanced the tensile strength of the
poly(AM/DAAM)/ADH nanofibers (i.e., of the core unit of the
core-shell fibermats) and the long-term stability of the stacked
nanofibrous structure after immersion in a buffer. Therefore, we
next move to protein encapsulation experiments. Because stable
retention of enzyme molecules in the core unit without leakage
is indispensable for evaluation of their biological activities, we
first assessed the release behavior of encapsulated proteins after
immersion in buffers of distinct pH. As a probe protein, we used
GFP (Figure 7),28 which allows easy chase of protein leakage by
monitoring GFP fluorescence. Moreover, GFP encapsulation in
polymeric materials has been widely attempted and discussed
the relationship between GFP fluorescence properties and the
Table 1. Contact Angles of Distinct Fibermats for Water
Droplets
Fibermat sample
Contact angle (°)
PCL fibermat
Poly(AM/DAAM)/ADH fibermat
Core-shell fibermat
125 ± 1.6
17 ± 6.3
119 ± 3.5
dumbbell-shaped samples of core-shell fibermats by using a
tensile test instrument. The obtained data are summarized in
Figure 6 and Table 2. The poly(AM/DAAM)/ADH fibermat was
extremely soft and fragile, and thus the maximum load value
could not be determined. Conversely, for the core-shell fibermat
of poly(AM/DAAM)/ADH and PCL, the maximum load value
was able to obtain and estimated to be 0.67 ± 0.04 N. As expected,
the tensile strength of the core-shell fibermat was adequately
raised by PCL wrapping.
To ascertain whether the use of the PCL wrapping would
enhance the long-term stability of stacked nanofibrous structures
in an aqueous buffer, we performed SEM analyses on
poly(AM/DAAM)/ADH fibermats and the core-shell fibermats
of poly(AM/DAAM)/ADH and PCL before and after immersion
in a buffer for 1 h (Figure 2(b), (c)). In the case of
poly(AM/DAAM)/ADH fibermats, although a stacked structure
of nanofibers was retained, partial fusion between adjacent
nanofibers was observed (Figure 2(c)); by contrast, no nanofiber
fusion was detected for the core-shell fibermat (Figure 2(b)).
Intriguingly, after immersion in the buffer, several increase (by
~30%) in the average diameter of the core-shell nanofibers was
observed, which could be caused by the swelling of the core unit
of poly(AM/DAAM)/ADH in the immersion buffer.
Figure 7. (a) SEM image of GFP-encapsulating core-shell
fibermat. Osmium oxide vapor was deposited onto the fibermat
sample. Acceleration voltage, 5 kV. (b) TEM image of GFP-
encapsulating core-shell fibermat. The core unit was stained with
phosphotungstate-Na, Acceleration voltage, 200 kV. (c)
Fluorescence emission of encapsulated GFP (red line) in core-
shell fibermats. The reference fluorescence spectrum of GFP in
a phosphate buffer (pH 7) was illustrated as blue line in the same
figure.
impact of encapsulation on protein structure and biological
activities.29 From the fluorescence emission properties, we could
estimate the effects of encapsulation in the core unit of core-shell
fibermat on protein structure and functions of encapsulating
enzyme molecules.
GFP was encapsulated in the core unit of core-shell
fibermats through simple addition of GFP to the precursor
phosphate buffer (pH 8) of 20 wt% poly(AM/DAAM)/ADH
(GFP content was fixed at 5 wt% relative to that of
poly(AM/DAAM)). Coaxial electrospinning was performed
using the same method as that used to construct the core-shell
Figure 6. Tensile-strength measurement of PCL fibermat and
core-shell fibermat of poly(AM/DAAM)/ADH and PCL.
5

fibermat without protein. A representative SEM image, TEM
image, and the ATR-IR spectrum of the obtained GFP-
encapsulating core-shell fibermat are illustrated in Figure 7(a),
Figure 7(b), and Figure 8, respectively. Although the average
diameter of the obtained core-shell fibermat was slightly
increased and the ratio of the average diameter (nm) of the core
unit to the thickness (nm) of the shell unit becoming larger than
that without protein (7.24 ± 1.50), a stacked nanofibrous
structure (Figure 7(a)) and the characteristic IR bands at 1710,
1650, 1615, 1540, and 1180 cm-1 (Figure 8) were observed.
These results indicate that GFP addition to the precursor solution
of the core unit did not affect the construction of the core-shell
fibermats of poly(AM/DAAM)/ADH and PCL. Furthermore,
examination of the emission behavior and fluorescence spectrum
of the encapsulated GFP (Figure 7(b)) revealed that the GFP λmax
(510 nm) in the core-shell fibermat was identical to that in a
buffer, which suggests that the original fluorescence property of
GFP was maintained even after encapsulation. Furthermore, the
homogeneous GFP emission from the fibermats implied
homogeneous distribution of GFP molecules in the core unit of
the core-shell fibermats. These data implied that other proteins
such as enzymes could also be encapsulated in the core unit of
core-shell fibermats without protein denaturation.
buffers of different pH, GFP likely remained entrapped in the
hollow tubular PCL shell layers.
Figure 9. Release behavior of encapsulated GFP from GFP-
encapsulating core-shell fibermats upon immersion in buffers of
(a) pH 3–5 and (b) pH 6–10.
Characterization of Enzyme-Encapsulating Core-Shell
Fibermats.
Lastly, we constructed enzyme-encapsulating core-shell
fibermats and characterized their biological activities. For
evaluating the biological activities of the enzymes encapsulated
in core-shell fibermats, it was necessary to first assess the
permeability of substrate molecules for enzymes into the
fibermats, because molecular permeation could potentially be
impeded by the PCL wrapping. Therefore, we constructed core-
shell fibermats encapsulating the low-MW molecule fluorescein
(MW = 332) or rhodamine B (MW = 479) as dummy fluorescent
molecules in the core unit of the fibermats and assessed their
release behaviors. If robust release of these encapsulated
molecules from the core unit to immersion buffers was observed,
adequate permeation of the substrate molecules having similar
MW from the immersion buffers to the inside of the core unit
was expected. The observed release behaviors into an immersion
buffer are summarized in Figure 10. As compared with
poly(AM/DAAM) fibermats,15 which lacked the PCL wrapping,
the release rate and release amount were slightly decreased.
Nevertheless, with fluorescein, the release was found to be
sufficient (>60%) and to rapidly reach equilibrium (within 30
min) (Figure 10), which suggested that if low-MW substrates are
used, enzymatic activities of enzyme-encapsulating core-shell
fibermats can be evaluated.
Figure 8. ATR-IR spectra of core-shell fibermats incorporating
no protein (black line) or encapsulating GFP (red line), α-Chy
(green line), or Lac (blue line).
We next analyzed the leakage behavior of the encapsulated
GFP into immersion buffers of different pH. Based on our
previous analysis of the poly(AM/DAAM)/ADH fibermat,13
poly(AM/DAAM)/ADH in the core unit was expected to
dissolve in an acidic buffer (pH < 5) according to the hydrolysis
of hydrazone bonds. Conversely, GFP was expected to be stably
retained in neutral or slightly basic buffers because the network
structure of poly(AM/DAAM) /ADH layer is not damaged under
these conditions. The obtained profile of GFP leakage from the
core-shell fibermat into immersion buffers is summarized in
Figure 9. As expected, under neutral and slightly basic
conditions, no release of GFP was observed. While, even under
acidic conditions (pH < 5), GFP release was not observed.
Because the hydrophobic PCL layer was insoluble in all tested
Figure 10. Release behaviors of fluorescein and rhodamine B
from core-shell fibermats encapsulating either molecule in the
core unit.
Considering the aforementioned results, we next selected
two representative pairs of enzyme and substrate; α-Chy30 and
p-NAc (MW = 181), and Lac31 and p-NGlu (MW = 301) (Figure
11) and constructed the corresponding enzyme-encapsulating
core-shell fibermats. In both cases, measurement of A404
(characteristic of p-nitrophenolate) enabled spectrophotometric
monitoring of the amounts of products generated by the
enzymes,23 which allowed evaluation of apparent enzymatic
activities.
6

of α-Chy or Lac dissolved directly in the assay buffer. The
estimated enzymatic activities from the observed data are
summarized in Table 3. In our previous study on the
poly(AM/DAAM)/ADH fibermat, we found that α-Chy-
encapsulating poly(AM/DAAM)/ADH fibermats exhibited
>85% apparent enzymatic activity relative to that of the same
amount of enzyme dissolved in buffer.15 Because of the porous
network structure of poly(AM/DAAM)/ADH and the high
surface area of the stacked nanofibrous structure of α-Chy-
encapsulating poly(AM/DAAM)/ADH fibermats, apparent
enzymatic activity successfully reached >85%. Although
hinderance of substrate permeation due to the PCL wrapping was
a major concern, we unexpectedly found that the α-Chy-
encapsulating core-shell fibermats exhibited nearly 80%
apparent enzymatic activity relative to that of the same amount
of enzyme dissolved in buffer. In case of Lac-encapsulating core-
shell fibermats, the activity reached 70% relative to that in buffer.
As our discussion of GFP-encapsulating core-shell fibermats in
the preceding section, α-Chy and Lac in the core unit of the core-
shell fibermats might retain their original structure and function
in the fibermats. Therefore, the small reduction observed here in
enzymatic activities (i.e., the reduction in apparent enzymatic
activities) is likely due to a decrease in the molecular
permeability of the substrate molecules.
(a)
(b)
O
OH
O
CH3
O
HO
HO
O
OH
NO2
MW 181.2
NO2
MW 301.3
Figure 11. Chemical structure of (a) p-nitrophenyl acetate and
(b) p-nitrophenyl-β-D-glucoside.
Core-shell fibermats encapsulating α-Chy or Lac were
constructed similarly as GFP-encapsulating fibermats. The
ATR-IR spectra, SEM images, and TEM images of these
fibermats are displayed in Figure 8 and Figure 12, respectively.
Although the ratio of the average diameter (nm) of the core unit
to the thickness (nm) of the shell unit became larger than that
without protein (6.14 ± 0.80 for α-Chy, 5.30 ± 0.48 for Lac), the
obtained data indicated that encapsulation of neither enzyme
markedly affected the construction of core-shell fibermats of
poly(AM/DAAM)/ADH and PCL. On the other hand, that the
ratio of core diameter to shell thickness is over 5 means that,
even if the protein-loaded nanofibers had 1 µM of diameter, still
shell thickness is less than 150 nm. In terms of enzyme reactivity,
this characteristic was expected to have advantages.
One of the advantages of encapsulating enzymes in the
nanofibers of fibermats is that the enzymes are protected from
undesirable interactions with contaminating molecules in
biological samples, such as inhibitory peptides and proteases.
Because of the filtering effect originating from the PCL
wrapping in the shell unit and the network structure of poly(AM
Figure 13. Plots showing p-nitrophenol amounts (nmol)
produced in 10 min against amounts of (a) α-Chy or (b) Lac in
the core unit of core-shell fibermats.
Table 3. Apparent Enzymatic Activities of Core-Shell Fibermats
Encapsulating α-Chy in the Presence/Absence of TCI or
Encapsulating Lac: Comparison with Enzymes Dissolved in
Assay Buffer
Apparent enzymatic activity (nmol min-1 mg-1)
α-Chymotrypsin
Lactase
No TCI
8.95 ± 0.91
Presence of TCI
Figure 12. SEM and TEM images of the core-shell fibermats
encapsulating α-Chy ((a), (c)) and Lac ((b), (d)). For TEM
measurements, the core unit was stained with phosphotungstate-
Na, Acceleration voltage, 200 kV.
Buffer
0.28 ± 0.03
(97% deactivated)b
5.34 ± 0.29
7.73 ± 0.20
---
Poly(AM/DAAM) 7.83 ± 0.25
/ADH fibermat
Core-shell
fibermat
(87%)a
7.24 ± 1.08
(81%)a
(32% deactivated)b
6.84 ± 0.81
5.38 ± 0.52
(70%)a
Enzymatic activities of the encapsulated enzymes were
evaluated from the product amounts in a 10-min reaction by
using different weights of enzyme-encapsulating core-shell
fibermats (i.e., by using different weights of enzyme samples
encapsulated in the core-shell fibermats). In linear-fitting
analyses of the data collected from these samples (Figure 13),
the enzymatic activities were estimated as the slopes of the
obtained plots, and nmol min-1 mg-1 was used as the activity unit
(1 U = 1 mg of enzyme sample that produces 1 nmol of
enzymatic product within 1 min). In control experiments, we
also evaluated enzymatic activity by using an identical amount
(6% deactivated)b
aThese values were calculated in comparison with values measured in
buffer. bThese values were calculated in comparison with values,
measured in the absence of TCI.
/DAAM)/ADH in the core unit, we expected the access of
middle- or high-MW molecules (>1 kDa) to the encapsulated
enzyme in the core unit to be suppressed. As a pilot experiment,
we evaluated the effect of an inhibitory peptide, TCI,20 on the
enzymatic activity of α-Chy either encapsulated in the core-shell
fibermat or dissolved in a buffer. TCI is a linear 7.9-kDa peptide
7

that can form a 1:1 complex with α-Chy in solution. After this
1:1 binding, α-Chy enzymatic activity was strongly inhibited
(97% of the original activity was suppressed; Table 3). In our
previous study on the poly(AM/DAAM)/ADH fibermat, α-Chy
encapsulation in the fibermat protected the original α-Chy
activity by 68%; inclusion of α-Chy in the network structure of
poly(AM/DAAM)/ADH hindered the TCI approach, but a
certain level of TCI binding and the subsequent inhibition still
remained.15 By comparison, in the case of the core-shell
fibermats, only 6% of the original enzymatic activity of α-Chy
was inhibited. The double-layered fibrous structure of the core-
shell fibermats, consisting of the PCL shell and the network
structure of poly(AM/DAAM)/ADH, provided clearly superior
protection of α-Chy against TCI binding. This type of protection
of the biological functions of proteins against the effects of
middle- or high-MW molecules (>1 kDa) is likely to be a general
behavior encountered when using the core-shell fibermats.
Additional optimization of fibermat nanostructure and selection
of polymer species can potentially further enhance the
performance of these fibermats relative to that demonstrated for
protein-encapsulating single-layered fibermats.
3.
4.
5.
6.
M. G. Lancina III, R. K. Shankar, H. Yang, J Biomed
Mater Res A. 2017, 105, 1252-1259.
Y. Dai, J. Niu, J. Liu, L. Yin, J. Xu, Bioresour. Techol.
2010, 101, 8942-8947.
,
H. Qi, P. Hu, J. Xu, A. Wang, Biomacromolecules 2006, 7
2327-2330.
M. Monfared, S. Taghizadeh, A. Z.-Hoseinabadi, S. M.
Mousavi, S. A. Hashemi, S. Ranjbar, A. M. Amani, Drug
Metab. Rev. 2019, 51, 589-611.
7.
8.
9.
S. Zupančič, Acta Pharm. 2019, 69, 131-153.
Y. Yang, X. Li, W. Cui, S. Zhou, R. Tan, C. Wang, J
Biomed Mater Res A. 2008, 86, 374-85.
G. Ren, X. Xu, Q. Liu, J. Cheng, X. Yuan, Y. Wan, React.
Funct. Polym. 2006, 66, 1559-1564.
10. H. Bang, K. Watanabe, R. Nakashima, W. Kai, K.-H. Song,
J. S. Lee, M. Gopiraman, I.-K. Kim, RSC Advances 2014,
4, 59571-59578.
11. P. Gupta, S. Trenor, T. E. Long, G. L. Wilkes,
Macromolecules 2004, 37, 9211-9218.
12. M. Kurecic, T. Mohan, N. Virant, U. Maver, J. Stergar, L.
Gradisnik, K. S. Kleinschek, S. Hribernik, RSC Advances
2019, 9, 21288-21301.
13. S. Koeda, K. Ichiki, N. Iwanaga, K. Mizuno, M. Shibata,
A. Obata, T. Kasuga, T. Mizuno, Langmuir 2016, 32, 221–
229.
14. K. Mizuno, S. Koeda, A. Obata, J. Sumaoka, T. Kasuga, J.
R. Jones, T. Mizuno, Langmuir 2017, 33, 4028-4035.
15. Y. Ido, A. L. B. Maçon, M. Iguchi, Y. Ozeki, S. Koeda, A.
Obata, T. Kasuga, T. Mizuno, Polymer 2017, 132, 342-352.
16. Z. Sun, Z. E. Alexander, A. L. Yarin, J. H. Wendroff, A.
Greiner, Adv. Mater. 2003, 15, 1929-1932.
17. N. Han, P. A. Bradley, J. Johnson, K. S. Parikh, A. Hissong,
M. A. Calhoun, J. J. Lannutti, J. O. Winter, J. Biomater.
Sci. Polym. Ed. 2013, 24, 2018-2030.
18. S. Srouji, D. Ben-David, R. Lotan, E. Livne, R. Avrahami,
E. Zussman, Tissue Eng. A 2011,17, 269-277.
19. C. L. McCormick, A. B. Lowe, Acc. Chem. Res. 2004, 37,
312-325.
20. Y. Birk, Int. J. Pept. Protein Res. 1985, 2, 113-131.
21. T. Zor, Z. Selinger, Anal. Biochem. 1996, 236, 302-308.
22. T. C. McIlvaine, J. Biol. Chem. 1921, 49, 183-186.
23. A. E. Beg, J. Chem. Soc. Pak. 1984, 6, 55-61.
24. Y. Zhang, Z.-M. Huang, X. Xu, C. T. Lim, S. Ramakrishna,
Chem. Mater. 2004, 16, 3406-3409.
25. N. Han, J. K. Johnson, P. A. Bradley, K. S. Parikh, J. J.
Lannutti, J. O. Winter, J. Funct. Biomater. 2012, 3, 497-
513.
26. P. Wen, Y. Wen, X. Huang, M. H. Zong, H. Wu, J. Agric
Food Chem. 2017, 65, 4786-4796.
27. T. Elzein, M. Nasser-Eddine, C. Delaite, S. Bistac, P.
Dumas. J. Colloid Interf. Sci. 2004, 273, 381–387.
28. J. D. Pédelacq, S. Cabantous, T. Tran T. C. Terwilliger, G.
S. Waldo, Nat. Biotechnol. 2006, 24, 79-88.
4. Conclusion
In this study, we characterized enzyme-encapsulating core-
shell fibermats in which enzyme molecules (of α-Chy or Lac)
were included in the core unit of poly(AM/DAAM)/ADH and a
PCL layer was wrapped around this core unit. Distinct from the
enzyme-encapsulating
poly(AM/DAAM)/ADH
fibermat
composed of only the poly(AM/DAAM)/ADH layer including
enzyme molecules, the core-shell fibermat exhibited sufficient
mechanical strength and stability of the stacked nanofibrous
structure in a neutral-pH buffer, with both of these characteristics
originating from the hydrophobic shell unit of PCL. Furthermore,
the thin PCL wrapping (<150 nm thick) of the
poly(AM/DAAM)/ADH layer did not markedly impede the
permeation of low-MW substrates. Thus, the enzyme -
encapsulating core-shell fibermats functioned effectively in
allowing entry of low-MW substrates while concurrently
suppressing the adverse impact of middle- or high-MW (>1 kDa)
contaminating molecules such as inhibitor peptides and
proteases; the apparent enzymatic activities of α-Chy-
encapsulating and Lac-encapsulating fibermats reached 81% and
70%, respectively, relative to activities of the same amounts of
the enzymes dissolved in buffer. These results imply that core-
shell fibermats of poly(AM/DAAM)/ADH (core unit) and
PCL or other hydrophobic polymers (shell unit) could emerge as
a novel class of enzyme-immobilizing platforms. Our group is
now comprehensively evaluating enzyme-encapsulating core-
shell fibermats generated using distinct combinations of
polymers.
Acknowledgement
29. J. N. Brantley, C. B. Bailey, J. R. Cannon, K. A. Clark, D.
A. Vanden Bout, J. S. Brodbelt, A. T. Keatinge-Clay, C.
W. Bielawski, Angew. Chem. Int. Ed. 2014, 53, 5088-5092.
30. F. J. Kezdy, M. L. Bender, Biochemistry 1962, 1, 1097-
1106.
31. M. M. Maksimainena, A. Lampio, M. Mertanen, O.
Turunen, J. Rouvinen,, Int. J. Biol. Biomacromol. 2013, 60,
109-115.
This work was supported by JSPS KAKENHI (Grant
Number 17K05932), and the TOYOAKI SCHLORSHIP
FOUNDATION, and the NIMS Molecule & Material Synthesis
Platform in the “Nanotechnology Platform Project” operated by
the Ministry of Education, Culture, Sports, Science and
Technology (MEXT).
References
1.
R. Jayakumar, S. Nair, Ed. Biomedical Applications of
Polymeric Nanofibers; Springer: Berlin, Heidelberg,
Germany, 2012.
2.
L. Tian, M. P. Prabhakaran, X. Ding, D. Kai, S.
Ramakrishna, J. Mater. Sci.: Mater. Med. 2013, 24, 2577-
2587.
8