Access to this full-text is provided by Wiley.
Content available from Advanced Materials
This content is subject to copyright. Terms and conditions apply.
www.advmat.de
2211637 (1 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Fibro-Gel: An All-Aqueous Hydrogel Consisting of
Microfibers with Tunable Release Profile and its Application
in Wound Healing
Yanting Shen, Yuan Liu, Janine K. Nunes, Chenmin Wang, Miao Xu, Michael K.T. To,*
Howard A. Stone,* and Ho Cheung Shum*
Y. Shen, Y. Liu, M. Xu, H. C. Shum
Department of Mechanical Engineering
The University of Hong Kong
Pokfulam Road, Hong Kong SAR, China
E-mail: ashum@hku.hk
J. K. Nunes, H. A. Stone
Department of Mechanical and Aerospace Engineering
Princeton University
Princeton, NJ 08544, USA
E-mail: hastone@princeton.edu
C. Wang, M. K.T. To
Department of Orthopaedics and Traumatology
LKS Faculty of Medicine
University of Hong Kong
Pokfulam Road, Hong Kong SAR, China
E-mail: mikektto@hku.hk
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.202211637.
DOI: 10.1002/adma.202211637
1. Introduction
Hydrogels have various biomedical
applications such as tissue-engineered
constructs,[1] drug delivery systems,[2] cell-
based therapies,[3] wound dressings,[2b,4]
and antiadhesion materials.[5] The use of
hydrogels is a consequence of their being
composed of a large amount of water and
a crosslinked polymer network that pro-
vides physical similarity to the extracellular
matrix and the capability to easily encapsu-
late drugs.[6] Cell and cytokine therapeutics
can be significantly enhanced via encapsu-
lation within hydrogels,[7] due to stabiliza-
tion during delivery,[2b,8] protection from
the immune system in vivo,[9] and localiza-
tion to the intended delivery region,[10] ulti-
mately extending the therapeutic window.
Recently, the evolution of injectable
hydrogels has been driven by the need to
overcome the drawback of the pre-formed
hydrogel for recapitulating natural tissue
function with a minimally invasive implan-
tation procedure.[3b,11] The injectable hydro-
gels not only have the typical advantages of conventional hydrogels
but they can also be injected with minimal invasiveness into target
sites and used for irregularly shaped sites. Accordingly, they have
been developed as a promising and successful material system for
many biomedical applications including the delivery of therapeutic
agents for the treatment of infectious diseases[2a] and for the regen-
eration of tissues such as bone, cartilage, muscle, and skin.[12]
However, there still are several major challenges to enhancing
care with hydrogels and reducing costs and waste: i) Poorly
scalable fabrication of a functional hydrogel hinders the wider
clinical applications and wastes resources.[13] A simple and cost-
eective manufacturing system with high throughput would
provide a much-needed boost to enable the utility of functional
hydrogels in a variety of industrial settings. ii) It is dicult using
the existing hydrogels to precisely tailor mechanical properties
for individual cases to match specific cell regeneration require-
ments. The mechanical properties of hydrogels aect the fate
and phenotype of the extracellular matrix.[1a] For instance, the
elastic moduli of 0.1–1kPa influence mesenchymal stem cell dif-
ferentiation.[14] iii) Controlling the release of distinct molecules at
dierent rates from existing hydrogels remains a challenge. The
potential importance of tunable drug release rates is exemplified
Injectable hydrogels are valuable tools in tissue engineering and regenera-
tive medicine due to their unique advantages of injectability with minimal
invasiveness and usability for irregularly shaped sites. However, it remains
challenging to achieve scalable manufacturing together with matching phys-
icochemical properties and on-demand drug release for a high level of control
over biophysical and biomedical cues to direct endogenous cells. Here,
the use of an injectable fibro-gel is demonstrated, a water-filled network of
entangled hydrogel microfibers, whose physicochemical properties and drug
release profiles can be tailored to overcome these shortcomings. This fibro-
gel exhibits favorable in vitro biocompatibility and the capability to aid vascu-
larization. The potential use of the fibro-gel for advancing tissue regeneration
is explored with a mice excision skin model. Preliminary in vivo tests indicate
that the fibro-gel promotes wound healing and new healthy tissue regenera-
tion at a faster rate than a commercial gel. Moreover, it is demonstrated that
the release of distinct drugs at dierent rates can further accelerate wound
healing with higher eciency, by using a two-layer fibro-gel model. The
combination of injectability and tailorable properties of this fibro-gel oers a
promising approach in biomedical fields such as therapeutic delivery, medical
dressings, and 3D tissue scaolds for tissue engineering.
ReseaRch aRticle
© 2023 The Authors. Advanced Materials published by Wiley-VCH
GmbH. This is an open access article under the terms of the Creative
Commons Attribution License, which permits use, distribution and
reproduction in any medium, provided the original work is properly cited.
Adv. Mater. 2023, 35, 2211637
www.advmat.dewww.advancedsciencenews.com
2211637 (2 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
by the fact that tissue repair and regeneration involve the sequen-
tial signaling of several growth factors.[2a] Therefore, controllable
hydrogel dressings with the ability to provide customized treat-
ment settings and modify drug release profiles at dierent rates
for diverse tissue engineering applications are urgently needed.
Here, we develop an injectable fibro-gel with controllable
physicochemical properties and drug release rates by changing
the length of the individual fibers that make up the gel. The
fibro-gel refers to a gel that is formed upon extrusion of a sus-
pension of long, flexible microfibers.[15] Based on this flow-
induced gelation mechanism, we sought to synthesize an oil-
free, biocompatible, biomimetic, and fully scalable hydrogel
material with a wide range of properties for use. Benefiting from
tetracycline (TC) and epidermal growth factor (EGF) loaded
inside the microfiber, as shown in Figure 1a, the fibro-gel exhib-
ited biological active functions: antibacterial activity and vascu-
larization, which were verified with both in vitro and in vivo
experiments. In addition,weexplored the potential of fibro-gel
for clinical applications by using an excisional wound healing
model. In vivo, the drug-loaded fibro-gel served as a wound
dressing demonstrating significantly improved therapeutic out-
comes compared to a commercial gel (Hydrosorb Gel).
2. Results and Discussion
2.1. From Microfiber Suspension to Fibro-Gel
For the fabrication of the fibro-gel, instead of traditional oil/
aqueous systems,[15] we used the aqueous two-phase system
(ATPS),[16] poly(ethylene glycol) diacrylate (PEGDA) and potas-
sium phosphate tribasic, to fabricate the microfibers using
continuous flow microfluidic methods. After reaching phase
equilibria, the polymer-rich phase was mixed with a photo-
initiator, lithium phenyl-2,4,6-trimethylbenzoylphosphinate
(LAP), and drugs, e.g., tetracycline (TC) and/or epidermal
growth factor (EGF), and then flowed as the inner phase. The
salt-rich phase was used as the outer phase for forming the
co-flow configuration. The flow rates of the outer and inner
phases were fixed during the whole fabrication procedure,
which created a uniform inner flow with a central stream of
diameter 110µm. Pulsed UV illuminationwasused to segment
a continuous pre-gel aqueous phase into uniform microfibers,
as shown in Figure1a. The length of the fiber, Lfiber, is propor-
tional to the product of the flow rate and the exposure time (the
“pulse on” time). Inour study, Lfiberwas only adjusted by the
UV exposure time due to the fixed flow rates.
After collecting a certain volume of microfibers from the
microfluidic device, the injection strategy was adapted to form
the fibro-gel from the microfiber suspensions, as shown in
Figure1c and Movie S1, Supporting Information. Triggered by
a flow-induced mechanism that produced physical entangle-
ments,[15] upon extrusion the fiber suspension was converted
into a viscoelastic gel, with no additional chemical reactions or
post-processing required. This approach can be extended to fab-
ricate fibers using most photosensitive hydrogel components,
as well as to fabricate the corresponding fibro-gels.
Using this method to form the fibro-gel also yielded a
high throughput with roughly 3 mL of fibro-gel (volume frac-
tion = 50%) every 1h operating a single channel device. The
Adv. Mater. 2023, 35, 2211637
Figure 1. Fabrication of fibro-gel. a) Schematic illustrations of the microfiber generation using a co-flow microfluidic device. The inner phase contains
polymer (PEGDA), photoinitiator (LAP), and drug (TC/EGF). The drug-loaded microfibers are synthesized by pulsed UV illumination. The flow rates
are fixed and the length of the fiber, Lfiber, is adjusted by the exposure time. b) Extrusion of the fibro-gel from a needle, where the fibro-gel is formed
from a suspension of fibers. The extruded fibro-gel is viscoelastic with a high viscosity. c) SEM image of a dried fibro-gel, with a magnified image (right)
showing the micromorphology of the fibro-gel.
www.advmat.dewww.advancedsciencenews.com
2211637 (3 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
throughput can be increased significantly and easily using mul-
tiple streams.[17] This laboratory-scale experiment paves the way
for the cost-eective and high-throughput fabrication of fibro-
gels at an industrial scale.
2.2. Characterization of the Fibro-Gel
2.2.1. Controllable Physicochemical Properties of Fibro-Gel
The micromorphology of the hydrogel was observed by scan-
ning electron microscopy (SEM). The structures revealed that
the fundamental process of gelation is mechanical, in which
fibers flex (bend, twist, form loops, etc.) and form topological
entanglements as a result of extrusion, as shown in Figure1d.
By changing Lfiberwe were able to generate fibro-gels with
distinct properties. The fibro-gels with the same fiber diam-
eter (110 µm) but dierent fiber lengths were formed with
the same fiber volume fraction (50%) from a needle with the
same extrusion rate. Individual confocal z-slices of fibro-gels
created with dierent Lfiber were taken (Figure 2a), showing
a 3D structure formed by coiled fibers. With an increase of
Lfiber from 8.8 to 220 mm, the number of looped and coiled
fibers increased, indicating that fibro-gels with longer fiber
lengths had more complex topological entanglements. Based
on the confocal images, the porosity of the fibro-gels for dif-
ferent Lfiber was explored. As shown in Figure2b, the porosity
of the fibro-gel decreased with an increase in the fiber length,
which was a direct consequence of more complex topological
entanglements.
Adv. Mater. 2023, 35, 2211637
Figure 2. Characterizations of the fibro-gel. The properties of the fibro-gel, such as swelling ratio and modulus, depending on the length of the indi-
vidual microfibers, Lfiber. a) Individual confocal z-slices showing the microfiber deformations in the fibro-gels for dierent fiber lengths and the same
fiber diameter produced with similar operating conditions. The number of looped and coiled fibers increases with Lfiber. Scale bar: 200µm. b) The
porosity of the fibro-gels for dierent fiber lengths decreases with the increase of Lfiber. c) Frequency-sweep rheological properties of the fibro-gel for
dierent fiber lengths. The elastic modulus G′ (solid pattern) and viscous modulus G″ (hollow pattern) are measured using a parallel-plate rheometer.
The fibro-gels exhibited an elastic-like behavior (i.e., G′> G″). Both G′ and G″ increase with Lfiber. The arrow highlights the increase for G’ and G’’ with
increasing Lfiber. d) Strain-sweep rheology study of fibro-gels for dierent fiber lengths. The arrow highlights the increase for G′ and G″ with increasing
Lfiber. e,f) The tunable swelling behavior of the fibro-gels for dierent fiber lengths. The fibro-gels reached the equilibrium swollen state within 5 minutes.
The fibro-gel with longer fiber lengths displayed a smaller swelling ratio.
www.advmat.dewww.advancedsciencenews.com
2211637 (4 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
To verify the mechanical properties of fibro-gels, rheological
analysis as a function of frequency and strain were conducted
using a parallel-plate rheometer.[15] To study the influence of
Lfiber on the mechanical properties of the fibro-gels, the rheo-
logical properties of the fibro-gels for dierent Lfiber lengths
were compared. As shown in Figure2c,d, the fibro-gels exhib-
ited an elastic-like behavior (i.e., G′> G″)[18] and both G′ and
G″ increased with an increase in fiber length (indicated by the
arrow); here G′ indicates the elastic modulus and G″ indicates
the viscous modulus. All the fibro-gels showed linear viscoelastic
regions for strains < 10%; G′ decreased at higher strain values
due to the onset of network breakdown. Both the micro-network,
which is the chemical crosslinking of the individual microfibers,
and the macro-network, which is the entanglement among
many microfibers, exist in the fibro-gels. With the same experi-
mental conditions, the molecular scale, chemical crosslinking of
the hydrogel should be the same. Hence, a qualitative explana-
tion for the dierence in the mechanical properties is that with
an increase in the fiber length, the entanglements inside the
fibro-gels become more complex with a larger number of looped
and coiled fibers, which suggests that the main mechanism for
the formation of the macro-network of the fibro-gels leads to
the dierences of G′ and G″. These easily tunable mechanical
properties allow the production of fibro-gels with a wide range
of properties to be constructed for clinical use.
The fibro-gels also showed dierent swelling behaviors for
dierent Lfiber, as shown in Figure2e,f. The swelling ratio (%)
of the fibro-gels was calculated as:
sw
elling ratio% 100%
t0
0
WW
W
()
=
−
× (1)
where Wt was the weight of the swollen fibro-gel at time t and W0
was the weight of the initial fibro-gel before swelling. Although
all of these fibro-gels reached an equilibrium swollen state within
5 min (Figure2e), which illustrates the favorable swelling ability,
the fibro-gel with longer fiber lengths displayed a smaller swelling
ratio as shown in Figure2f. Hydrogels with higher specific sur-
face areas and larger pore sizes have larger swelling rates and
lower swelling activation energies due to their faster response
behavior.[19] Thus, the measurements indicate that dierent fiber
lengths would not only lead to dierent degrees of physical entan-
glements (Figure2c,d), and so distinct rheological properties, but
also to dierent surface areas and pore sizes of the fibro-gel.
2.2.2. Tunable Release Profile of Fibro-Gel
To explore the drug release profile of the fibro-gel,wefirst encap-
sulated tetracycline (TC) into the fibro-gels and measured the
released amount of TC from the same volume of fibro-gels with
dierent Lfiber. The cumulative drug release (%) and the drug
release rate (mg per day) of the hydrogels were calculated as:
where Pt was the percentage released at time t and Pt-∆t was the
percentage released at the time of the previous measurement.
As shown by the results in Figure 3a, fibro-gels with dierent
fiber lengths have dierent drug release profiles. A fibro-gel
with longer fibers provided a prolonged drug release time,
which was calibrated by using a UV absorption spectrometer
(see Experimental Section and Figure S1, Supporting Informa-
tion). The drug release rates of fibro-gels with dierent Lfiber
showed that there was a rapid release rate (burst release) at
earlier times as shown in Figure3b (upper-right panel), which
decreased with increasing fiber length, followed by a sustained
release rate as shown in Figure3b (bottom-right panel), which
increased a small fraction with increasing fiber length. To dem-
onstrate whether the fibro-gels can achieve tunable release of
various cargo, the release rates of EGF from fibro-gels with dif-
ferent Lfiber were explored in Figure3c,d, showing that the EGF
release profile changes with Lfiber as well.
As part of the drug administration strategy, burst release
has been used to deliver drugs at high release rates,[20] while
sustained release can provide the ability to maintain a constant
medication level within the body.[21] For instance, the drugs used
at the beginning of wound treatment with an initial burst pro-
vide immediate relief followed by prolonged release to promote
gradual healing.[2a] In accordance with the fiber-length-respon-
siveness, the fibro-gel could release the drug more rapidly when
the fiber length is shorter and more slowly when the fiber
length is increased (Figure3b,d).
When a drug is incorporated into a swellable gel, the dif-
fusivity of encapsulated molecules is directly aected by the
degree of swelling and porosity of the gel.[22] The drug release
kinetics is controlled by the increase in release area produced
by the swelling phenomenon. The fibro-gel with shorter fibers
has a faster response swelling behavior, higher swelling degree,
and larger pore size (Figure 2a,b,e,f ). Consequently, the drug
is released faster from the fibro-gel with shorter fibers. Based
on the fiber-length-dependent swelling property, the networks
inside the fibro-gels were expected to be dierent, with the
pore size inside the fibro-gel increasing with a decrease of fiber
length;[22] thus,wehave achieved on-demand, controlled delivery
of therapeutics by using the fiber-length-dependent properties.
For tissue engineering applications, dierent drugs (e.g., integ-
rins, growth factors, and small molecule medicines) should be
presented over dierent timescales during therapy so that coor-
dinated cellular response can be accurately triggered.[2a,23] With
the fiber-length-dependent drug release rate,wecan achieve the
controlled release of dierent drugs at dierent rates by loading
drugs into dierent microfibers whose fiber lengths are dif-
ferent. An illustration of how this approach can be used on an
actual wound is shown in Figure3e. To confirm the assumption
of the on-demand release of dierent drugs, 0.85mg mL−1 TC
and 42.5ng mL−1 EGF were encapsulated into the 8.8mm and
220 mm microfiber respectively to form a two-layer fibro-gel.
cumulative
percentage release(%) Volume of sample withdrawnmL
Bath volume mL
PP
tt t
()
()
=×
+
−∆ (2)
drug
releaseratemg/dayCumulative drug releaseamountmg
releasetimeday
()
()
()
= (3)
Adv. Mater. 2023, 35, 2211637
www.advmat.dewww.advancedsciencenews.com
2211637 (5 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
Figure 3. Drug release profile of the fibro-gel. The drug release rates of the fibro-gel can be adjusted by Lfiber. a) The TC release profile of fibro-gels for
dierent fiber lengths. The fibro-gel with longer fibers exhibits a more sustained drug release profile. Hydrolysis and/or the drug tightly entrapped in
the gel also result in the incomplete release of the drug. b) The TC release rates of the fibro-gels for dierent fiber lengths show that there is a rapid
release rate at earlier times (upper-right panel; burst release in the early stage), which decreases with increasing fiber length, followed by a sustained
release rate (lower-right panel; sustained release in the later stage), which increases a small fraction with increasing fiber length. The drug release
rates were calculated from the average cumulative drug release amount (Equation3). c) The EGF release profile of fibro-gel for dierent fiber lengths.
Hydrolysis and/or the drug tightly entrapped in the gel also result in the incomplete release of the drug. d) The EGF release rates of the fibro-gels for
dierent fiber lengths show that there is a rapid release rate at earlier times (upper-right panel), which decreases with increasing fiber length, followed
by a sustained release rate (lower-right panel), which increases a small fraction with increasing fiber length. e) Schematic illustration of a wound healing
application of a two-layer fibro-gel encapsulating dierent drugs. The TC was loaded into the bottom layer with shorter fibers (Lfiber= 8.8mm), while
the EGF was loaded in the upper layer with longer fiber (Lfiber= 220mm). Two layers of the fibro-gel were extruded sequentially. f ) The TC and EGF
release profiles of the two-layer fibro-gel as shown in Figure3e. The EGF that was loaded in the longer fibers has a prolonged release time compared
with the TC loaded in the shorter fibers. g) The TC and EGF release rates of the two-layer fibro-gel. The TC in the shorter fibers has a rapid release rate,
while the EGF in the longer fibers shows a sustained release rate. Release experiments were conducted in triplicate and results were presented as the
average ± standard deviation.
www.advmat.dewww.advancedsciencenews.com
2211637 (6 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
The drug release profile and the drug release rates were meas-
ured as shown in Figure3f,g. TC loaded into the shorter fiber
layer has a shorter release time with a faster release rate, while
EGF loaded into the longer fiber layer has a prolonged release
time with a sustained release rate.
2.3. Biocompatibility Test
To assess the potential of the fibro-gel as a dressing for tissue
engineering applications, in vitro cell culture experiments
were carried out using a fibro-gel consisting of fibers of diam-
eter 110 µm and Lfiber= 44 mm. Note that in the following in
vitro and in vivo tests, for the fibro-gel with TC and EGF group
(A), the TC and EGF were loaded together into the same fiber
(Lfiber= 44mm). While for the fibro-gel with TC and EGF (dif-
ferent Lfiber) group (B), TC was loaded into the shorter fibers
(Lfiber= 8.8mm), EGF was loaded into the longer fibers (Lfiber=
220 mm), respectively. Afterward, two layers of the fibro-gel
were extruded in sequence (see the schematic illustration in
Figure3e).
The biocompatibility of the fibro-gel was assessed using a
Live/Dead assay and the Cell Counting Kit 8 (CCK8) during
a 5-day culture of NIH 3T3 cells. In Figure 4a, fluorescence
images of cell cytotoxicity testing for the control group, the
commercial gel (Hydrosorb Gel) group, the fibro-gel without
TC and EGF group, and the fibro-gel group with TC and EGF
group (A) showed the condition of live cells (green) and dead
cells (red) on days 1, 3, and 5. Over the course of the study,
most cells retained their traditional spindle-like morphology,
and on days 3 and 5 they still showed strong proliferative
activity. Another control, cells treated with free EGF, was also
performed (see Figure S4, Supporting Information), showing
that the fibro-gel improves growth factor eectiveness over a
simple bolus treatment. The mean optical density (OD, absorb-
ance) was used to calculate the percentage of cell viability based
on the quantitative results verified by the CCK-8 kit:
re
lative cell viability(%) 100%
material DI water
cell medium DI water
OD OD
OD OD
()
()
=
−
−× (4)
The cell viability values were plotted by averaging triplicate
results. The relative cell viability values in all groups are shown
in Figure4b. For the first day, the cell viability in the fibro-gel
with TC and EGF group was significantly higher than in the
other three groups. Although there was no significant dier-
ence in the growth trend between control group and the fibro-
gel group with TC and EGF on the third day, by the fifth day it
was clear that the cells were growing faster in the fibro-gel with
TC and EGF group. The results of the fibro-gel without TC and
EGF group also further demonstrated the favorable biocom-
patibility of the pure fibro-gel. All results confirmed that the
NIH 3T3 cells used in this experiment maintained a classical
spindle-like morphology throughout the experiment and had
good viability and proliferation despite dierent concentrations
of fibro-gels, which demonstrates the good biocompatibility of
the fibro-gel (Figure4c; Figure S2, Supporting Information).
The eects on cells of Lfiber and fiber arrangement and ori-
entation in the fibro-gels with dierent fiber lengths were also
investigated (see Figure S3, Supporting Information). The
results showed that Lfiber of fibro-gel did not aect the cell
viability, as most of the cells still showed strong proliferative
activity in the fibro-gel after 96h, which demonstrates the good
biocompatibility of the fibro-gel for dierent fiber lengths.
2.4. Angiogenesis Assay in Vitro
Vasculogenesis is a major element in tissue engineering appli-
cations,[24] so tube-formation experiments were used to assess
the eects of the fibro-gel on the angiogenesis process in vitro.
In particular, the cell nucleus was stained by 4′,6-diamidino-
2-phenylindole (DAPI), and the actin filaments of the cells were
stained by Phalloidin. The results in Figure4d,e showed that the
human umbilical vein endothelial cells (HUVECs) treated with
the fibro-gel with TC and EGF (14.7±1.5 mm) or the fibro-gel
without TC and EGF (11.1 ± 0.3mm) exhibited more junctions
and longer tube lengths than the control group (1.5± 0.5mm)
and the commercial gel group (Hydrosorb Gel) (7.3 ± 1.3mm).
Another control, HUVECs treated with free EGF, has been
conducted (see Figure S5, Supporting Information), indicating
that the sustained delivery improves vascularization. Although
vascular tube formation appeared in the control group and the
commercial gel group, the tubular framework seemed less
complete and with a smaller number of junctions and tube
lengths compared to the fibro-gel group.
2.5. In Vivo Proof-of-Concept of the Ecacy of the Fibro-Gel
The excisional wound healing model[25] is an animal model in
which a small piece of full-thickness skin is completely removed
from the wound bed. The assessment of epithelialization, gran-
ulation tissue formation, scar formation, and angiogenesis can
be achieved in this single model. Thus, it is a good model for
assessing the fibro-gel’s in vivo biocompatibility and the poten-
tial of accelerating the repair of tissues,[26] Therefore, the tissue
regeneration eciency of the commercial gel, the fibro-gel
without TC and EGF, and the fibro-gel with TC and EGF were
investigated in vivo by a mice excision skin model as shown in
Figure 5a. The healing results at various times are shown in
Figure5b. In the control group, dressings were not applied to
the wounds while in the commercial gel group and the fibro-gel
group, the mice were dressed, respectively, with corresponding
hydrogels. In the free TC and EGF group, the mice were treated
with 0.85mg TC and 42.5 ng EGF on day 0. On day 4, a lot of
yellow pus (indicated by the dashed line) was seen scattered in
the wound bed in the control group indicating a wound infec-
tion, while it was rarely detected in the other groups over the
course of the study. On the same day, compared with other
groups, the wound healing area was minimized in the fibro-
gel with TC and EGF group (B) (dierent Lfiber), indicating that
infection had been managed eectively, the inflammation was
significantly reduced, and the wound repair process acceler-
ated wound healing at the early stage, which is an advantage
of the faster release of TC from the first-layer fibro-gel, as
shown in Figure 5c. The wound healing area in the control
group expanded slowly in the first 8 days, which could mean
that the infection was not under control. In contrast, in the
Adv. Mater. 2023, 35, 2211637
www.advmat.dewww.advancedsciencenews.com
2211637 (7 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
fibro-gel with TC and EGF group (B) (dierent Lfiber), antibiotic
(TC) release in the early stages eectively shortened the anti-
inflammatory time and accelerated wound healing into the later
stages. The sustained release of EGF from the second layer also
accelerated cell proliferation and remodeling. The healing rate
on day 8 of the fibro-gel with TC and EGF group (B) (dierent
Lfiber) (98.3 ± 0.7%) was higher than that of the control group
(48.7 ± 1.9%), the commercial gel group (61.8 ± 2.5%), the free
Adv. Mater. 2023, 35, 2211637
Figure 4. Cell cytotoxicity and angiogenesis assay of the fibro-gel. a) Live/dead fluorescence results of the control group, the commercial gel group
(Hydrosorb Gel), the fibro-gel without TC and EGF group, and the fibro-gel with TC and EGF group (A). Fluorescence images of cell cytotoxicity testing
for the four groups showing the condition of live cells (green) and dead cells (red) on days 1, 3, and 5. Scale bar: 500µm. b) Cell viability as determined
with NIH 3T3 cell counting CCK-8 assay for the control group, the commercial gel group, and the fibro-gel group for 1, 3, and 5 days. The concentra-
tion of hydrogel is the same in each gel group at 10mg mL−1. c) Cell viability using CCK-8 assay following treatment with dierent concentrations of
fibro-gels with TC and EGF (A) for 24h. d) Vascular tube-formation in vitro of HUVECs for the control group, the commercial gel group, and the fibro-
gel group at 6h. Scale bar: 40µm. e) The total vascular tube length for the control group, the commercial gel group, and the fibro-gel group at 6h.
(NS, not significant, *p< 0.05, **p< 0.01, ***p< 0.001, ****p< 0.0001).
www.advmat.dewww.advancedsciencenews.com
2211637 (8 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
Figure 5. In vivo proof-of-concept of the ecacy of the fibro-gel. a) Schematic of the mice excision skin model. A full-thickness circle of skin with a
1cm diameter was excised. In the control group, the mice were not dressed while in the commercial gel group (Hydrosorb Gel), the fibro-gel without
TC and EGF group, the fibro-gel with TC and EGF group (A) (same Lfiber), and the fibro-gel with TC and EGF group (B) (dierent Lfiber), the mice were
dressed with the commercial gel and the corresponding fibro-gel respectively. In the free TC and EGF group, the mice were treated with 0.85mg TC
and 42.5ng EGF on day 0. b) Photographs of mice skin wound tissues for dierent groups on days 0, 4, 8, and 12, showing that the wound site of the
fibro-gel group with TC and EGF in dierent fiber lengths is substantially reduced. The inner diameter of the rubber ring is 1cm. c) Measured wound
healing rate of dierentgroups for 4, 8, and 12 days, showing the fastest wound healing rates occur with the fibro-gel with TC and EGF group (B).
d) Hematoxylin and eosin (H&E) staining of skin tissue for the control group, the commercial gel group, the free TC and EGF group, the fibro-gel
without TC and EGF group, the fibro-gel with TC and EGF group (A) on day 12 and the fibro-gel with TC and EGF group (B) on day 8. In the fibro-gel
with TC and EGF group (B), the magnified images showed the complete epithelium and dermis structures, while in the control group, the free TC
and EGF group, and the commercial gel group the regenerated skin tissues displayed the typical appearance of collagenous scar tissue. e) Epidermal
thickness of skin tissue for dierent groups at the end of the experiment for the control group, the commercial gel group, the free TC and EGF group,
the fibro-gel without TC and EGF group, the fibro-gel with TC and EGF group (A) on day 12 and the fibro-gel with TC and EGF group (B) on day 8.
www.advmat.dewww.advancedsciencenews.com
2211637 (9 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
TC and EGF group (63.2 ± 1.5%), and the fibro-gel with TC and
EGF group (A) (88.1 ± 4.8%); these indicate a significant eect
on promoting wound healing as shown in Figure5b,c. When
the treatment was extended, the healing area of the wound
continued increasing, and a 98.3 ± 0.7% wound closure rate
was reached on day 8 in the fibro-gel with TC and EGF group
(B) (dierent Lfiber) (Figure5b,c), when compared with that in
the free TC and EGF group (90.5± 0.9%), the commercial gel
group (91.7 ± 1.0%), the fibro-gel with TC and EGF group (A)
(95.8 ± 1.3%) and the control group (82.5 ± 3.2%) on day 12. As
evidenced by the macroscopic observations, the fibro-gel exhib-
ited significant potential to expedite the complex process of
tissue regeneration, which required both antibacterial activity
and vascularization.
2.6. Histological Analysis
2.6.1. Hematoxylin and Eosin Staining
The wound samples at each time point were harvested and the
frozen sections were utilized for hematoxylin and eosin (H&E)
staining. As illustrated in Figure5d, in the control group the
regenerated skin tissue displayed the typical appearance of scar
tissue with a flattened epidermis after 12 days of treatment. In
the commercial gel group, the regenerated tissue displayed a
similar appearance, but with a thicker overall tissue compared
with the control group (Figure5d,e). However, in the fibro-gel
with TC and EGF group (B) (dierent Lfiber), the defect was
almost closed after 8 days of treatment and the regenerated
skin tissue revealed a de novo regenerated appearance that
almost has no dierence from normal tissue (Figure 5d–f ).
Overall, there was poor epithelialization in both the control and
commercial gel groups histologically, whereas the fibro-gel with
TC and EGF group (B) (dierent Lfiber) showed complete re-epi-
thelialization after 8 days of treatment.
2.6.2. Immunofluorescence Staining
Furthermore, we also used an immunofluorescence stain to
determine if the potential to vascularize was present. The fluo-
rescence expressions of vascular smooth muscle cells marker
α-SMA and vascular endothelial-specific marker CD31 were
tested as shown in Figure 6a,b. α-smooth muscle actin (α-SMA)
is often used as a marker of myofibroblast formation and CD31
is a protein that in humans is encoded by the PECAM1 gene
which can help to evaluate the degree of tissue angiogenesis.
The corresponding results on day 12 for the control group, the
commercial gel group, and the free TC and EGF group, shown
in Figure6c,d, revealed minimal α-SMA and CD31 expression
during the whole period. Conversely, the fluorescence inten-
sity of α-SMA and CD31 expression intensity in the fibro-gel
with TC and EGF group (B) (dierent Lfiber) on day 8 was 25.1
± 2.7 and 27.3 ± 2.3 respectively, which were significantly higher
than the expression intensity in the other groups. The higher
expression of α-SMA in the fibro-gel group also indicates more
myofibroblast formation during the wound healing process.
Myofibroblast can be derived from the transition of fibroblast
through mechanical and physical regulation. The fibro-gel may
have provided the physical stimuli. The faster wound healing
or closure of the wound in the fibro-gel group may also be the
result of the myofibroblast contraction. Overall, we conclude
that the fibro-gel provided an excellent 3D sterile microenviron-
ment platform for reconstructing a vascular network and was
more eective for improving tissue regeneration based on all of
the aforementioned findings.
3. Conclusion
We have designed and fabricated an oil-free, biocompatible,
biomimetic, and fully scalable fibro-gel for tissue engineering
applications. A considerable range of physicochemical proper-
ties and drug release rates can be modulated for our fibro-gel
by simply adjusting the microfiber lengths. Thus, fibro-gels
have a wide range of properties that can be constructed for clin-
ical use. Combining the tunable properties and the “plug and
play” nature of this microfluidically generated strategy allows
the incorporation of diverse already established materials (for
example, fibrin or hyaluronic acid), signals (for example, growth
factors), and cell populations (for example, stem cells). Complex
combinations of microfibers with dierent fiber lengths with
deterministic chemical and physical properties may enable
tissue regeneration in a variety of distinct physiological niches
(for example, neural, cardiac, skin, and so on). We also demon-
strated, by using a mice excision skin model, that the fibro-gel
has favorable biocompatibility and accelerated tissue regenera-
tion ability. The preliminary in vivo test results showed the con-
siderable advantages of the fibro-gel with a faster new tissue
regeneration rate and healthier de novo regenerated tissue
when compared with the commercial gel. In the future, we
need to further evaluate the tunable properties and the eects
of complex combinations of microfibers with dierent drugs or
cells in dierent animal models. Overall, the unique combina-
tion of scalable, injectable, and tunable physicochemical prop-
erties of fibro-gel has the potential to introduce a new approach
toward a precisely tailored gel for in situ tissue engineering.
4. Experimental Section
Fabrication of the Fibro-Gel: Microfibers were made by an aqueous
two-phase system (ATPS), poly(ethylene glycol) diacrylate (PEGDA,
Mw = 575 g mol−1, 25 wt.%; Sigma), and potassium phosphate tribasic
(6 wt%; Sigma). The polymer-rich phase was mixed with 1 wt.% of the photo-
initiator, lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP; Sigma),
f) Quantitative analysis of the regenerated tissue thickness for the control group, the commercial gel group, the free TC and EGF group, the fibro-gel
without TC and EGF group, the fibro-gel with TC and EGF group (A) on day 12 and the fibro-gel with TC and EGF group (B) on day 8. The recovery
of regenerated tissue was evaluated by measuring the thickness of regenerated tissue. Paired t was calculated and values with a p-value ≤ 0.05 are
considered statistically significant. (NS, not significant, *p< 0.05, **p< 0.01, ***p< 0.001, ****p< 0.0001).
www.advmat.dewww.advancedsciencenews.com
2211637 (10 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
www.advmat.dewww.advancedsciencenews.com
2211637 (11 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
and the drugs, 1 mg mL−1 tetracycline (TC; Sigma) and/or 50 ng mL−1
epidermal growth factor (EGF; Thermo Fisher); this solution was used
as the inner phase of the co-flow. Pulsed UV illumination was used to
polymerize the aqueous jet into uniform microfibers. By tuning the UV
curing time, the length of the microfiber can be changed.
After collecting a certain volume of microfiber, the microfibers were
washed three times with DI water. Then the drug-loaded microfiber
suspension (fiber volume fraction = 50%) was extruded from a standard
needle (0.5″ long 27 G needle, inner diameter = 0.21mm) at a constant
extrusion rate 65 mL h−1 by syringe pump to form a fibro-gel. The
concentrations of the TC and EGF in the fibro-gel after washing were
calibrated at 0.85 mg mL−1 and 42.5ng mL−1.
Characterization of the Fibro-Gel: Scanning electron microscopy:
Freeze-dried fibro-gel was adhered on an aluminum substrate using
carbon tape and coated with a small amount of gold for 30 s. The coated
samples were observed by SEM (HITACHI S-4800).
Confocal z-slices were performed on a laser-scanning confocal
microscope (Zeiss). Fluorescent dye, rhodamine 6G (sigma), was used
for imaging.
Rheological Measurements: The rheological properties of the fibro-gel
were measured using a stress-controlled rheometer.[15] All experiments were
carried out at room temperature (23°C). To minimize the eect of wall slip,
we used a rough parallel plate with a diameter of 50mm. The measurement
gap was fixed at 1mm, and the shear stress was set at a constant 0.7Pa.
Drug Release Profile: The released amount of tetracycline (TC) and/
or epidermal growth factor (EGF) from the fibro-gels for dierent fiber
lengths was detected in PBS (pH = 7.4). The fibro-gels (1 mL) were
immersed in 15mL of PBS and the samples were shaken at 25 rpm on
a rotary shaker at 37 °C throughout the release study.[27] At certain time
intervals (∆t= 1, 2, 3, 4, 5,10 h, and every 24 h), 1 mL of the solution
was removed for measurement and an equal volume of fresh PBS was
added. UV–vis spectrophotometer (Shimadzu) was used to measure the
amount of released TC. The released amount of EGF in the collected PBS
solution was quantified using an EGF ELISA kit (Thermo Fisher). For the
two-layer fibro-gel, the TC-loaded layer was extruded from the shorter fiber
suspensions (Lfiber= 8.8mm; 1mL) followed by extruding the second layer
of EGF-loaded fibro-gel (Lfiber= 220mm; 1mL) on the top of the first layer.
Due to the high viscosity of the fibro-gel itself, there was no additional
reaction between the two layers. The two-layer fibro-gels were immersed in
15mL of PBS and the samples were shaken at 25rpm on a rotary shaker
at 37 °C throughout the release study. At certain time intervals (∆t= 1,
2, 3, 4, 5,10 h, and every 24 h), release experiments were conducted in
triplicate and results were presented as the average ± standard deviation.
The cumulative drug release (%) and the drug release rates of the TC and/
or EGF were calculated according to Equations2 and3.
Cell Experiments: NIH 3T3 and HUVEC cells (Thermo Fisher) were
used for in vitro experiments. The cells were incubated in DMEM
(Gibco) supplemented with 10% fetal bovine serum (Gibco), 1%
penicillin-streptomycin solution (Hyclone) and incubated in a CO2
constant temperature incubator.
Biocompatibility Test: NIH 3T3 cells were seeded in 96-well plates at
proper density and cultured with the DMEM buer in the control group, with
10 mg mL−1 commercial hydrogel in the commercial group, and with
10mg mL−1 microfiber-based hydrogel in the fibro-gel group. After 1, 3, and 5 days
of incubation, CCK-8 reagent (Dojindo) was mixed with fresh DMEM medium
for culturing for 4h and cell viability was quantified by a microplate reader at
450nm (Molecular Devices). Additionally, the Live/Dead kit (Invitrogen) was
comprised of calcein-AM (green fluorescence) and ethidium homodimer-1
(red fluorescence) that were employed to assess NIH 3T3 cell viability. The
fluorescent images were acquired usinga Nikon Ti2-E Widefield microscope.
Also, dierent concentrations (0, 8, 16, 32, and 64 mg mL−1) of fibro-gels
were cultured with the NIH 3T3 cells for 24h to verify the cell viability.
Angiogenesis Assay In Vitro: Growth factor-reduced basement
membrane extract Matrigel (200 µL; Corning) was used as a culture
substratum. A thin layer (10 µL per well) of commercial gel, fibro-gel
without TC and EGF, fibro-gel with TC and EGF (A) were laid over
the prechilled 24-well culture plates in the commercial gel group, the
fibro-gel without TC and EGF group, and the fibro-gel with TC and EGF
group (A) respectively. The concentrations of the TC and EGF in the
fibro-gel with TC and EGF group (A) were 0.85 mg mL−1 and 42.5 ng
mL−1. The free EGF group was added with 42.5ng mL−1 EGF while the
control group received nothing (all in triplicate). HUVEC cells were
seeded in the prepared 24-well plates at proper densities and cultured in
a CO2 constant temperature incubator. The morphology was monitored
under the optical microscope regularly. After 6h, the wells were rinsed
with PBS three times and then fixed with 4% polyformaldehyde (PFA;
Sangon Biotech), permeabilized with 0.2% Triton X-100 (Sigma),
and stained with phalloidin (Invitrogen) and DAPI (Sigma) solution.
Fluorescence images were acquired using a laser scanning confocal
microscope (Zeiss), and vascular tube lengths were quantified by Image
J Software in five randomly selected fields of view.
Animal Tests: The animal experiments were carried out according to
a protocol approved by the Committee on the Use of Live Animals in
Teaching and Research (CULATR, the University of Hong Kong: 5805-21),
which adheres to The International Guiding Principles for Biomedical
Research Involving Animals and The Hong Kong Code of Practice for Care
and Use of Animals for Experimental Purposes.In total, 36 male, 8-week-old
C57BL/6N mice weighing ≈20 grams were used in the experiment (The
University of Hong Kong; CULATR no. 5805–21). The mice were separated
into six groups (the control group, commercial gel group, free TC and EGF
group, fibro-gel without drug TC and EGF group, fibro-gel with TC and
EGF (same Lfiber) group (A), and fibro-gel with TC and EGF (dierent Lfiber)
group (B) (see Figure5) for 4 dierent days (day = 0, 4, 8, and 12; n= 3
for each timepoint for each group). Two circles with 1.0cm diameter full-
thickness skin were excised on the back of mice. The control group and
the fibro-gel with TC and EGF group (A) were tested by the same mice
for 12 days. The commercial gel group and the fibro-gel without TC and
EGF group were tested by the same mice for 12 days. The free TC and EGF
group and the fibro-gel with TC and EGF group (B) were tested by the same
mice for 12 days. Anesthesia is achieved with an intraperitoneal injection
of pentobarbital sodium solution at a dose of 50 mg kg−1. In the control
group, the mice were not dressed, while in the commercial gel group, the
fibro-gel without TC and EGF group, and the fibro-gel with TC and EGF
group (A and B), the mice were dressed in the corresponding hydrogel with
the same volume (1mL) respectively. In the free TC and EGF group and the
fibro-gel with TC and EGF group (A and B), the concentrations of TC and
EGF were 0.85 mg mL−1 and 42.5ng mL−1 respectively.
In Vivo Wound Healing: The criteria of complete healing of the wound
were defined by the denuded wound being completely covered by layers
of keratinocytes, and a newly stratified epidermis with underlying basal
lamina re-established from the margins of the wound. The skins in
various groups were harvested immediately after the wounds healed
completely. H&E trichrome stain and immunohistochemical test were
carried out in each group. The closure rate of the wound in dierent
groups at each time point was calculated by the formula: healing rate
(%) = (A0[initial wound area] − At [residual wound area at each time
point])/(A0[initial wound area]) × 100%.
Histological Analysis: The wound samples at each time point were
harvested and the frozen sections were utilized for hematoxylin and
eosin (H&E) staining. The epithelialization of the control group,
the commercial gel group, the free TC and EGF group, the fibro-gel
without TC and EGF group, and the fibro-gel with TC and EGF group
(A) on day 12 were calculated. While the epithelialization of the fibro-gel
with TC and EGF group (B) on day 8 was calculated. The percentage
Figure 6. Immunofluorescence staining of α-SMA and CD31 for dierent groups. a,b) Histological section images of mice skin wound tissues with
immunohistochemistry staining of protein α-SMA (green) and CD31(red) for the control group, the commercial gel group, the free TC and EGF group,
the fibro-gel without TC and EGF group, the fibro-gel with TC and EGF group (A) on day 12 and the fibro-gel with TC and EGF group (B) on day 8.
Scale bars are 200µm in the top row and 50µm in the bottom row. c) Measured α-SMA density in (a) for dierent groups (n= 3). d) Measured CD31
density in (b) for dierent groups. (n= 3; NS, not significant, *p< 0.05, **p< 0.01, ***p< 0.001, ****p< 0.0001).
www.advmat.dewww.advancedsciencenews.com
2211637 (12 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
of epithelialization was calculated based on the H&E staining images
according to the equation: 1 − (pixels open wound area/pixels total
wound area) × 100%. Means of epithelialization were calculated based
on 15–20 random site measurements. Immunofluorescence, such
as α-SMA (Abcam) and CD31 (Abcam), was used to analyze collagen
deposition and angiogenesis. All images were captured using a
fluorescence microscope (Zeiss).
Statistical Analyses: All experiments were repeated at least three times.
The data were analyzed by one-way ANOVA followed by Tukey’s test
(GraphPad Prism software) where p< 0.05 meant a significant dierence.
(NS, not significant, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.)
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
Acknowledgements
This project is supported by Research Grant Council of Hong Kong
through the Research Impact Fund (No. R7072-18), with H.C.S. as the
Project Coordinator, and J.K.N., H.A.S., and M.K.T.T. as co-principal-
investigators under a collaborative project. This research is partially
supported by the Health@InnoHK program of the Innovation and
Technology Commission of the Hong Kong SAR Government as well as
by NSF through the Princeton University (PCCM) Materials Research
Science and Enigneering Center DMR-2011750. The authors thank
Aviva S.F. Chow, Qingchuan Li, and Yi Pan for careful reading of the
manuscript and their critical feedback. The authors also thank Jiaying Liu
and Sihan Liu for their help with the cell experiment.
Conflict of Interest
H.C.S. is a scientific advisor of EN Technology Limited, and
MicroDiagnostics Limited, in both of which he owns some equity, and
also a managing director of the research center, namely Advanced
Biomedical Instrumentation Centre Limited. The works in the paper are
however not directly related to the works of these two entities, as far as
the authors know. The other authors declare no conflict of interest.
Author Contributions
Y.S. and Y.L. contributed equally to this work. Y.S. and Y.L. designed
research; Y.S. and Y.L. performed research; Y.S., Y.L., J.K.N., M.K.T.T.,
H.A.S., and H.C.S. analyzed data; Y.S., Y.L., J.K.N., H.A.S., and H.C.S.
wrote the paper and all authors commented on the paper.
Data Availability Statement
The data that support the findings of this study are available from the
corresponding author upon reasonable request.
Keywords
biomaterials, drug deliveries, injectable hydrogels, microfluidics, wound
healing
Received: December 12, 2022
Revised: February 6, 2023
Published online: March 24, 2023
[1] a) Z. Wang, H. Cui, M. Liu, S. L. Grage, M. Homann,
E. Sedghamiz, W. Wenzel, P. A. Levkin, Adv. Mater. 2022, 34,
2107791; b) F. M. Yavitt, T. E.Brown, E. A. Hushka, M. E. Brown,
N.Gjorevski, P. J.Dempsey, M. P. Lutolf, K. S. Anseth, Adv. Mater.
2020, 32, 1905366; c) T. Yu, L. Zhang, X. Dou, R. Bai, H. Wang,
J.Deng, Y.Zhang, Q.Sun, Q.Li, X.Wang, B.Han, Adv. Sci. (Weinh).
2022, 9, 2203734.
[2] a) J.Li, D. J.Mooney, Nat. Rev. Mater. 2016, 1, 16071; b) G. Kiaee,
P. Mostafalu, M. Samandari, S. Sonkusale, Adv. Healthcare Mater.
2018, 7, 1800396.
[3] a) Y. S. Zhang, A. Khademhosseini, Science 2017, 356, 6337;
b) D. R.Grin, W. M.Weaver, P. O.Scumpia, D.Di Carlo, T.Segura,
Nat. Mater. 2015, 14, 737.
[4] a) Z. Tu, Y. Zhong, H. Hu, D. Shao, R. Haag, M. Schirner,
J. Lee, B. Sullenger, K. W. Leong, Nat. Rev. Mater. 2022, 7,
557; b) Y. Liu, J. Liu, S. Chen, T. Lei, Y. Kim, S. Niu, H. Wang,
X.Wang, A. M.Foudeh, J. B.Tok, Z. Bao, Nat. Biomed. Eng. 2019,
3, 58.
[5] Y. Wang, L. Li, Y.Ma, Y. Tang, Y. Zhao, Z. Li, W.Pu, B. Huang,
X. Wen, X. Cao, J. Chen, W.Chen, Y. Zhou, J. Zhang, ACS Nano
2020, 14, 8202.
[6] P. S. Briquez, L. E. Clegg, M. M. Martino, F. M. Gabhann,
J. A.Hubbell, Nat. Rev. Mater. 2016, 1, 1.
[7] L. J.Macdougall, T. E. Homan, B. E. Kirkpatrick, B. D. Fairbanks,
C. N. Bowman, S. L. Spencer, K. S. Anseth, Adv. Mater. 2022, 34,
2202882.
[8] X.Hao, T.Du, H.He, F.Yang, Y.Wang, G.Liu, Y.Wang, Adv. Mater.
Interfaces 2022, 9, 2102184.
[9] a) J.Kim, D. M.Francis, L. F.Sestito, P. A.Archer, M. P.Manspeaker,
M. J.O’Melia, S. N.Thomas, Nat. Commun. 2022, 13, 1479; b) X.Li,
Y.Shou, A.Tay, Adv Nanobiomed Res 2021, 1, 2000073.
[10] C. C.Huang, S.Ravindran, Z.Yin, A.George, Biomaterials 2014, 35,
5316.
[11] a) H. D.Lu, D. E.Soranno, C. B.Rodell, I. L.Kim, J. A.Burdick, Adv.
Healthcare Mater. 2013, 2, 1028; b) K.Lee, J. A.Hubbell, Curr. Opin.
Biotechnol. 2013, 24, 827.
[12] a) L. Yu, J. Ding, Chem. Soc. Rev. 2008, 37, 1473; b) E. A.Appel,
M. W.Tibbitt, M. J. Webber, B. A.Mattix, O. Veiseh, R.Langer, Nat.
Commun. 2015, 6, 6295.
[13] A. C.Yu, H.Chen, D.Chan, G.Agmon, L. M.Stapleton, A. M.Sevit,
M. W. Tibbitt, J. D. Acosta, T. Zhang, P. W. Franzia, R. Langer,
E. A.Appel, Proc Natl Acad Sci 2016, 113, 14255.
[14] a) T. R.Cox, J. T.Erler, Dis Model Mech 2011, 4, 165; b) D. T.Butcher,
T.Alliston, V. M.Weaver, Nat. Rev. Cancer 2009, 9, 108.
[15] A.Perazzo, J. K.Nunes, S.Guido, H. A.Stone, Proc. Natl. Acad. Sci.
2017, 114, E8557.
[16] a) GY.Chao, H. C.Shum, Chem. Soc. Rev. 2020, 49, 114; b) H. Y.Lo,
Y.Liu, S. Y. Mak, Z. Xu, Y. Chao, K. J. Li, H. C. Shum, L. Xu, Phys.
Rev. Lett. 2019, 123, 134501.
[17] a) M. B.Romanowsky, A. R.Abate, A.Rotem, C.Holtze, D. A.Weitz,
Lab Chip 2012, 12, 802; b) J.Kim, M.Hegde, S. H.Kim, T. K.Wood,
A.Jayaraman, Lab Chip 2012, 12, 1157.
[18] J.Yu, K.Wang, C.Fan, X.Zhao, J.Gao, W.Jing, X.Zhang, J.Li, Y.Li,
J.Yang, W.Liu, Adv. Mater. 2021, 33, 2008395.
[19] a) R. Censi, W.Schuurman, J. Malda, G. di Dato, P. E.Burgisser,
W. J. A. Dhert, C. F.van Nostrum, P. di Martino, T. Vermonden,
W. E. Hennink, Adv. Funct. Mater. 2011, 21, 1833; b) Y. Huang,
M.Zeng, J.Ren, J.Wang, L.Fan, Q.Xu, Colloids Surf. A 2012, 401,
97.
[20] a) X. Huang, C. S. Brazel, J. Control Release 2001, 73, 121;
b) S. R. Benhabbour, M. Kovarova, C. Jones, D. J. Copeland,
R. Shrivastava, M. D. Swanson, C. Sykes, P. T. Ho, M. L. Cottrell,
A. Sridharan, S. M. Fix, O. Thayer, J. M. Long, D. J. Hazuda,
P. A. Dayton, R. J. Mumper, A. D. M. Kashuba, J. V.Garcia, Nat.
Commun. 2019, 10, 4324.
www.advmat.dewww.advancedsciencenews.com
2211637 (13 of 13) © 2023 The Authors. Advanced Materials published by Wiley-VCH GmbH
Adv. Mater. 2023, 35, 2211637
[21] a) Y.Wang, X.Yang, X.Chen, X.Wang, Y.Wang, H.Wang, Z.Chen,
D. Cao, L. Yu, J. Ding, Adv. Funct. Mater. 2022, 32, 2206554;
b) Y. Lvov, W. Wang, L.Zhang, R.Fakhrullin, Adv. Mater. 2016, 28,
1227.
[22] K. J. De France, F. Xu, T. Hoare, Adv. Healthcare Mater. 2018, 7,
1800396.
[23] K. J.Rambhia, P. X.Ma, J Control Release 2015, 219, 119.
[24] a) J.Vajda, M.Milojevic, U. Maver, B. Vihar, Biomedicines 2021, 9,
589; b) F.Simunovic, G.Finkenzeller, Cells 2021, 10, 1749.
[25] L.Chen, R.Mirza, Y.Kwon, L. A.DiPietro, T. J.Koh, Wound Repair
Regen 2015, 23, 874.
[26] a) J. E.Neil, M. B. Brown, A. C.Williams, Sci. Rep. 2020, 10, 21192;
b) A. L. Clement, G. D. Pins, Wound Heal. Biomater. Woodhead
Publishing, 2016, pp. 253–275.
[27] a) J.Koo, S. B.Kim, Y. S.Choi, Z.Xie, A. J.Bandodka, J.Khalifeh,
Y.Yan, H.Kim, M. K.Pezhouh, K.Doty, G.Lee, Y.-Y.Chen, S. M.Lee,
D.D’Andrea, K.Jung, K.Lee, K.Li, S.Jo, H.Wang, J.-H.Kim, J.Kim,
S.-G. Choi, W. J. Jang, Y. S. Oh, I. Park, S. S. Kwak, J.-H. Park,
D. Hong, X. Feng, C.-H. Lee, et al., Sci Adv. 2020, 6, eabb1093;
b) K. Hirota, A. C. Doty, R. Ackermann, J. Zhou, K. F. Olsen,
M. R. Feng, Y. Wang, S. Choi, W. Qu, A. S. Schwendeman,
S. P.Schwendeman, J Control Release 2016, 244, 302.
