Access to this full-text is provided by Wiley.
Content available from Advanced Healthcare Materials
This content is subject to copyright. Terms and conditions apply.
RESEARCH ARTICLE
www.advhealthmat.de
Aqueous Two-Phase Emulsion Bioresin for Facile One-Step
3D Microgel-Based Bioprinting
Qingbo Wang, Özge Karadas, Oskar Backman, Luyao Wang, Tuomas Näreoja,
Jessica M. Rosenholm, Chunlin Xu, and Xiaoju Wang*
Microgel assembly as void-forming bioinks in 3D bioprinting has evidenced
recent success with a highlighted scaffolding performance of these bottom-up
biomaterial systems in supporting the viability and function of the laden cells.
Here, a ternary-component aqueous emulsion is established as a one-step
strategy to integrate the methacrylated gelatin (GelMA) microgel fabrication
and assembly through vat photopolymerization in situ using digital light
processing (DLP)-based bioprinting. The as-proposed aqueous emulsion is
featured with the partitioning of a secondary photo-crosslinkable
polysaccharide, methacrylated galactoglucomannan (GGMMA) derived from
plant source in both the dispersed phase of GelMA droplets and the
continuous phase of dextran (Dex). As an emulgator, GGMMA renders
enhanced stability of the aqueous emulsion bioresins. Strategically, the
photo-crosslinkable GGMMA adheres the GelMA microgels that are
conveniently converted from emulsion droplets to form hydrogel construct in
layer-by-layer curing to accommodate the laden cells directly mixed in the
aqueous emulsion. The spatially interconnected void space left by the removal
of Dex benefits the cell growth under the guidance of the microgel surface and
supports cell colonization within the macroscopic porous hydrogel. This work
amends a low-concentration and cost-effective bioresin that is highly
applicable for facilely fabricating microgel assembly as a porous hydrogel
construct in DLP-based bioprinting.
1. Introduction
3D bioprinting technology based on additive manufactur-
ing of cell-encapsulated biomaterials (bioinks) into complex
Q. Wang, O. Backman, L. Wang, C. Xu, X. Wang
Laboratory of Natural Materials Technology
Faculty of Science and Engineering
Åbo Akademi University
Henrikinkatu 2, Turku 20500, Finland
E-mail: xwang@abo.fi
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adhm.202203243
© 2023 The Authors. Advanced Healthcare 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.
DOI: 10.1002/adhm.202203243
architectures in a precisely controlled man-
ner is prevailing among the new develop-
ments on the biofabrication horizon. Un-
doubtedly, 3D bioprinting shows proven
establishments as a revolutionary toolbox
not only for creating in vitro 3D cell mod-
els in use for mechanistic disease under-
standing and drug screening but also for
engineering tissue mimics in regenera-
tive medicine.[1–3 ] Hydrogel scaffolds have
been exploited as material combinations
of biological or synthetic origins in con-
stituting versatile bioinks, as they pro-
vide a spatial and hydrated microenviron-
ment for cell residence.[4,5 ] Yet, bulk hy-
drogels have merely nanoporous meshes
inside the crosslinked network and thus
face inevitable challenges such as insuffi-
cient nutrient exchange, poor cell infiltra-
tion, and weak vascularization, mainly due
to the lack of micropores in the macro-
scopic construct.[6,7 ] More importantly, the
in vitro 3D cell models need to be biologi-
cally relevant in terms of recapitulating the
microenvironmental factors that resemble
native tissue or disease pathology.[8] The
composition and structural characteristics
of extracellular matrix (ECM) can also be
of heterogeneous nature in supporting a specific tissue func-
tion or under disease progression.[9,10 ] To reflect the complexity
of ECM, creating heterogeneity in in vitro models is remained
to tackle in the development of biomaterial systems. Microgel
Ö. Karadas, T. Näreoja, J. M. Rosenholm, X. Wang
Pharmaceutical Sciences Laboratory
Faculty of Science and Engineering
Åbo Akademi University
Tykistökatu 6A, Turku 20520, Finland
T. Näreoja
Molecular Biotechnology & Diagnostics
Department of Life Technologies
University of Turku
Kiinamyllynkatu 10, Turku 20520, Finland
T. Näreoja
Department of Laboratory Medicine
Karolinska Institute
Stockholm 17177, Sweden
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (1 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
assembly is now emerging as a prospective biomaterial for 3D
bioprinting to address the aforementioned challenges.[11] This
bottom-up approach assembles from the elementary microgel
to a packed macroscopic aggregate via precise control of the as-
sembly. The great advantage is that the size and shape of micro-
gels are highly tunable, and the space between the packed mi-
crogels provides abundant void space and interconnected micro-
pores to the hydrogel, which would improve the spatial activities
of cells, referring to cell morphology, spreading, infiltration, and
proliferation.[11,12 ]
Various fabrication methods like batch emulsion, microfluidic
emulsion, electrodynamic spraying, and mechanical fragmenta-
tion were investigated to fabricate microgels.[4,12–16 ] Droplet mi-
crofluidic devices have favored generating monodisperse sphere
microgels using different polymeric materials. The as-prepared
microgels are assembled either through microgel jamming in
random assembly or more specific assembly strategies, result-
ing in inter-microgel crosslinking via a physical or chemical
reaction.[17–20 ] In the context of 3D bioprinting, microgel jam-
ming has been a basis for integrating the microgel assembly as
an extrudable bioink since the densely packed material provides
the thixotropic and recovery properties to meet the rheological
requirements.[15,21 ] Generally, the cells encapsulated in the mi-
crogels show high viability after the bioprinting process. How-
ever, the long-term stability of the printed densely packed micro-
gels is challenged due to the weak physical interactions.[11] Thus,
a secondary crosslinking strategy is essential through the anneal-
ing process or dispersing microgels into a crosslinkable aque-
ous phase that improves the structural stability through post-
curing and reaches an extended biofabrication window.[4,12] Nev-
ertheless, the microgel-based bioink preparation processes are
still complex and limited in large-volume production since the
microgels need to be prepared separately involving the organic
phase to form oil/water emulsion, followed by crosslinking for
stabilization and repeat washing before mixing with cells. Hence,
research efforts have been focused on establishing facile meth-
ods of preparing microgels. Strategies such as mechanical siev-
ing of the bulk hydrogel into microscale particles or high-aspect-
ratio microstrands have been practiced in microgel-based 3D
bioprinting.[22] However, the void fraction of the microstrand-
based hydrogel is lower compared to the spherical microgel sys-
tems. Still, the workflow for these bioprinting processes is com-
plex, which possibly includes 1) hydrogel precursor preparation,
2) microgel fabrication, 3) bioink preparation, and 4) bioprinting
of microgel assembly in a format of macroscopic hydrogel con-
struct. Therefore, developing functional bioinks that enable one-
step microgel bioprinting is significant to simplify the process.
Recently, a pioneering work reported by Ying et al. has estab-
lished aqueous two-phase emulsion (ATPE) bioinks to construct
microporous hydrogels in 3D bioprinting.[23] ATPE is formed by
phase separation driven by entropic contribution in an aqueous
mixture containing two incompatible additives, which is readily
applied in microgel fabrication.[24] In their work, polyethylene
oxide (PEO) droplets were dispersed in a continuous phase of
methacrylated gelatin (GelMA) through the formation of ATPE
in phosphate buffer saline (PBS) upon pipette mixing. In prin-
ciple, the aqueous emulsions provide oil-free and biocompati-
ble environments for the suspended cells. Then, a 3D bioprinted
microporous GelMA hydrogel resulted from the photocrosslink-
ing of the continuous phase (i.e., GelMA) followed by the leach-
ing of the dispersed phase (i.e., PEO) from hydrogel construct
into the incubation medium. The interconnected microporous
structure in the 3D bioprinted hydrogel promoted cell viability,
spreading, and proliferation in comparison to their bulk hydro-
gel counterparts. Moreover, attributed to the sol-gel transition na-
ture of GelMA, this ATPE bioink could meet the rheological re-
quirements and could be directly applied in both extrusion-based
and digital light processing (DLP)-bioprinting. When applying
the ATPE bioink in DLP-bioprinting, the bioink must be main-
tained in liquid status to enable the layer-by-layer recoating pro-
cess. Still, for the practical workflow of cell-mixing and bioprint-
ing at ease, extended stabilization of these all-aqueous droplets
is highly appreciated to guarantee a reasonable batch-to-batch
consistency in biofabrication. Efforts have been made by adjust-
ing the types, concentration, and molecular weight of GelMA
and porogen polymer of the bioink to achieve better stability for
DLP-bioprinting.[25] Besides, additives that can enhance the sta-
bility of the ATPE bioink/resins were also studied. Biosurfactant
such as rhamnolipid was deployed to stabilize GelMA/PEO ATPE
bioinks to enhance their applicability towards 3D bioprinting.[26]
Pickering emulsion strategy using 𝛽-lactoglobulin nanoparticles
or interleukin-4-loaded Ag-coated Au-nanorods (IL-4@AgGNRs)
have also been studied to render a stable emulsion bioink of
dextran (Dex) droplets in a continuous phase of GelMA and ap-
plied in DLP-bioprinting to fabricate porous hydrogels featured
with complex structures.[27,28 ] More interestingly, the multifunc-
tional additive could also endow the bioink with antimicrobial
properties and anti-inflammatory regulation.[28] Compared with
extrusion-based bioprinting, DLP bioprinting requires preload-
ing the photo-crosslinkable bioresin onto the printing bed, which
leads to the limitation of materials choice, as well as bioresin
wastage and cost increase.[29] However, rapid fabrication, high
printing resolution, and less mechanical shear stress to the laden
cells count as advantages in DLP-bioprinting.[3] Even though
ATPE bioinks facilitate the above-mentioned superior strategies
for constructing microporous hydrogel, it is a challenge to ap-
ply ATPE in microgel bioprinting by simply inverting the dis-
persed/continuous phases, as the insufficient interconnectivity
would greatly hinder the assembly process of microgels in DLP-
bioprinting. Furthermore, a bottom-up approach for microgel as-
sembly that would give control over the biomaterial system in
DLP-bioprinting has not thus far been developed. From this per-
spective, a strategy where the aqueous emulsion bioink for for-
matting GelMA droplets in a DLP printing setup was missing in
the field of 3D bioprinted of void-forming hydrogels.
Herein, based on the ATPE bioresin where GelMA forms
droplets in a continuous phase of Dex, we developed a strat-
egy to introduce a photo-crosslinkable polysaccharide, methacry-
lated galactoglucomannan (GGMMA), a derivative of woody het-
eropolysaccharides, as a third component to partition into both
phases; thus acting as a “photo-adhesive” biopolymer to facili-
tate the cell-laden bioprinting in a format of microgel assem-
bly in one-step fashion. We compared the cell response in the
microgel-based porous hydrogel construct to that in the dense
bulk hydrogel, the applicability and facileness of this aqueous
emulsion bioresin, and thereby, validated in the workflow forfab-
ricating and assembling the GelMA microgels in cell-laden DLP
bioprinting.
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (2 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 1. Illustration of the one-step microgel bioprinting of ATPE-based bioresin via vat photopolymerization.
2. Results
2.1. Designing Rationale and Characterizations of the ATPE
Photoresins
Inspired by the most recent peer developments on void-forming
and pore-forming bioinks, we propose an ATPE-based bioresin
system for one-step microgel bioprinting via vat photopoly-
merization, as illustrated in Figure 1.[ 1,4,23] This bioresin could
be directly applied in DLP-bioprinting to fabricate high-fidelity
microgel-based hydrogel composed of in situ annealed micro-
gels. In this ternary-component bioresin formulation, all biopoly-
mers play complementary roles in the bioprinting process. Upon
tip pipetting, GelMA resembles itself as the emulsified micro-
droplet that can be converted as individual spherical microgel
upon photocrosslinking, and Dex solution presents as a con-
tinuous phase in the emulsion. A secondary photo-biopolymer,
GGMMA is a key-purpose additive that is miscible in both GelMA
and Dex phases, most likely with a higher partitioning coefficient
in the Dex-rich phase than in the GelMA droplets. In vat polymer-
ization, GGMMA is hypothesized to adhere the GelMA micro-
gels, resulting in an annealed assembly with stronger viscoelastic
properties of the macroscopic construct, and this is essential to
warrant the structural integrity and to define the printing accu-
racy in DLP-bioprinting. During the culturing, Dex would leach
out to form the voids between the microgels, which provide spa-
tial cues to promote cell activities.[25–27 ]
GelMA and Dex are chosen as the ingredient for the emul-
sion (Emul) photoresin due to their non-opposite charges and
thermodynamic incompatibility, which could form an ATPE at
critical concentrations.[24,25,27 ] The photoresin formulations are
summarized in Table 1 in the experimental section. As shown in
Figure 2a, the Emul photoresins are less transparent compared
to the Non-Emul photoresins (without Dex), presumably due to
Tabl e 1 . The formulation of Emul and Non-Emul photoresins.
Photoresin GelMA
[% w/v]
GGMMA
[% w/v]
Dex
[% w/v]
LAP
[% w/v]
Emul 6_0 6 0 5 0.25
Emul 5_1 5 1 5 0.25
Emul 4.5_1.5 4.5 1.5 5 0.25
Emul 3_3 3 3 5 0.25
Non-Emul 6_0 6 0 0 0.25
Non-Emul 5_1 5 1 0 0.25
Non-Emul 4.5_1.5 4.5 1.5 0 0.25
Non-Emul 3_3 3 3 0 0.25
Emul 5_0a5050
Emul 4.5_0a4.5 0 5 0
Emul 3_0a3050
aEmulsion without GGMMA for stability analysis.
the unmatched refractive index between GelMA and Dex, and the
light scattering through the dispersed phase. The fluorophore-
conjugated R-GelMA and F-GGMMA were used to characterize
the size distribution of the emulsion droplets and the spatial dis-
tribution of the GGMMA using ImageJ software. The majority of
GelMA is concentrated in the dispersed phase, and the emulsion
droplet size is reduced from 26.6 ±21.9 to 13.8 ±4.7 μmbyde-
creasing the GelMA content from 6% to 3% (Figure 2b,c). Never-
theless, owing to the gentle pipetting preparation method and the
instability nature of the ATPE, even though most of the droplet
sizes are within the range of 10–50 μm, there are still some bigger
droplets that exist, as shown in Figure 2b,c.
The main drawback of the ATPE is its instability caused by
the rapid coalescence between the droplets.[30] We have also
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (3 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 2. Characterizations of the photoresins. a) Stability of Non-Emul and Emul photoresins at room temperature (GelMA mixed with R-GelMA: pink;
GGMMA mixed with F-GGMMA: green). From left to right in each column: Non-Emul/Emul 6_0, 5_1, 4.5_1.5, and 3_3. b) Fluorescence microscopy
images of the Emul photoresins. (GelMA marked with red fluorescence; GGMMA marked with green fluorescence; overlapping GelMA and GGMMA
observed as orange; scale bars: 200 μm) c) Droplet size distribution of the dispersed phase in Emul photoresins. d) TSI curves of the photoresins with
and without GGMMA.
noticed that the GelMA dispersed phase is likely to coalesce due
to its poor stability. As observed in Figure 2a, Emul 6_0 showed
a significant phase separation even after 15 min at room temper-
ature. To be noticed, the stability of the Emul photoresins signif-
icantly increased with the decrease of GelMA and the addition of
GGMMA. Besides the effect of the concentration between GelMA
and Dex, we speculate that the additive GGMMA also contributed
to the increase in the stability of Emul photoresins, possibly as
an emulgator. It is confirmed that GGMMA could significantly
decrease the surface tension of PBS buffer, as shown in Fig-
ure S1, Supporting Information. GGMMA, a polysaccharide with
the derivation of methacryloyls, can disperse into both GelMA
and Dex phases, as seen in Figure 2b. Attributed to its structural
similarity to Dex and chemical similarity of functional groups to
GelMA, it might form a layer consisting of GelMA/GGMMA/Dex
complex at the surface of GelMA droplets that inhibits them from
coalescence. Other types of polysaccharides or their derivatives
(chitosan, diethyl aminoethyl dextran, or propylene glycol algi-
nate) were earlier reported to stabilize the ATPE (PEO/dextran)
through interactions between the polymer and interface.[31] We
further utilized a sugar analysis approach to quantitatively as-
sess the distribution of GGMMA in these two phases and their
interface of ATPE. Mannose, as a marker of GGMMA’s charac-
teristic component, was employed for quantitative analysis. Af-
ter standing for 3 days at 37 ˚C, phase separation of the GelMA
rich phase (GelMA phase) and Dex phase was clearly observed in
the Emul photoresins, and the volume fraction of each phase is
mainly determined by the GelMA/Dex concentration ratio (Fig-
ure S2a,b, Supporting Information). Compared to the GelMA
phase, GGMMA was partitioned slightly more in the dextran
phase and the difference did not change with respect to the
GelMA/dextran ratio (Figure S2c, Supporting Information). Ow-
ing to the larger volume of the ATPE occupied by the Dex phase,
more GGMMA was found in the Dex phase than in the GelMA
phase. Depending on the formulations, around 10.8% to 20.3% of
the GGMMA was accumulated in the interface between GelMA
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (4 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 3. Temperature sweeps of a) Emul and b) Non-Emul photoresins illustrating the gelation temperature (close symbol: G′; open symbol: G″). c,d)
Flow curves of Emul and Non-Emul photoresins at 25 and 30 °C. (close symbol: Emul photoresins; open symbol: Non-Emul photoresins).
and Dex phase (Figure S2d, Supporting Information). We further
used the turbiscan stability index (TSI) as the metric to evaluate
the emulsion stability and made a comparison across the series
of Emul photoresins. As displayed in Figure 2d, the Emul 5_0
was the least stable formulation, where the TSI increased rapidly
to 20 after 20 min. The dramatic increase in the transparency of
Emul 5_0 at the bottom part is due to the sedimentation of the
GelMA phase (Figure S3, Supporting Information). As expected,
the addition of GGMMA led to a significant decrease in the TSI
value of Emul 5_1 (Figure 2d), and the transparency of the bottom
part became more consistent (Figure S3, Supporting Informa-
tion). Meanwhile, the stability of Emul photoresins was greatly
affected by the concentration between GelMA and Dex. Within
the testing period, a negligible extent of droplet coalescence, in
another word, no clear phase separation was observed in Emul
4.5_1.5 and 3_3. All in all, this confirms that the partitioning of
GGMMA in both phases improves the emulsion stability of ATPE
bioresins.
The temperature-dependent sol-gel transition of GelMA en-
ables it to formulate various bioinks/resins that could be applied
in extrusion-based printing and in vat photopolymerization.[32–35 ]
For the vat photopolymerization printing process, the flowability
of the photoresin is a primary prerequisite since the photoresin is
required to recoat onto the previous layer continuously.[3] In this
study, we first determined the gelation kinetics of the Emul pho-
toresins by oscillatory temperature sweep using a rheometer. As
shown in Figure 3a,b, the Emul photoresins showed higher gela-
tion temperature and lower G′at 5 ˚C compared to the Non-Emul
photoresins. The gelation temperature of GelMA bioink is largely
affected by the GelMA concentration.[36] The GelMA are evenly
distributed in Non-Emul photoresins, whereas the majority of
GelMA are concentrated in the dispersed phase with a higher lo-
cal concentration in Emul photoresins, thus resulting in a higher
gelation temperature. In regard to the hydrogel mechanics, the
Non-Emul photoresins are gelated into a homogenized network
with a higher G’ below gelation temperature. However, the het-
erogeneous network of Emul photoresins is mainly composed of
locally gelated GelMA phase and flowable Dex phase, thus result-
ing in a lower G′. In order to ensure the essential flowability as
required in vat photopolymerization, we further registered the
flow curves of the Emul and Non-Emul photoresins at 25 and
30 °C (Figure 3c,d). In general, the Emul photoresins showed a
higher viscosity than the Non-Emul counterparts, which might
be ascribed to the higher biopolymer concentration in total. At 25
°C, the Emul photoresins showed relatively high viscosities and
apparent shear-thinning properties when the GelMA concentra-
tion is high. At 30 °C, both Emul and Non-Emul photoresins ex-
hibited near Newtonian flow behavior and the series of all Emul
photoresins showed similar flow curves with a viscosity around
20 mPa·s within the testing range, regardless of the GelMA con-
centration. This low-viscosity Newtonian behavior of Emul pho-
toresins at 30 °C guarantees the demand on resin flowability in
vat photopolymerization.[3,37 ]
Rapid photo-crosslinking kinetics of the photoresin is crucial
for the vat photopolymerization bioprinting process, which en-
sures printability and reduces the damage to laden cells caused
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (5 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 4. Photo-rheological properties of the a) Emul and b) Non-Emul hydrogels (close symbol: G′; open symbol: G″). c) Gelation time of the Emul and
Non-Emul hydrogels. d) Representative images of the hydrogels cast by Emul and Non-Emul photoresins. Scale bars: 3 mm. e) Stress-strain curve and
f) Young’s modulus of the Emul and Non-Emul hydrogels (close symbol: Emul photoresins; open symbol: Non-Emul photoresins). g) Representative
3D reconstructed confocal images of FITC-dextran perfused microgel-based hydrogels. Scale bars: 60 μm. h) Representative confocal images of FITC-
dextran perfused bulk and microgel-based hydrogels. Scale bars: 200 μm. i) Porosity of the microgel-based hydrogels determined by the area occupied
by the FITC-dextran within the voids in 60 μm thick stacks. The data were collected from the confocal images. (n>3)
by prolonged exposure.[37,38 ] We thus characterized the photo-
crosslinking kinetics of photoresin using photo-rheology by ad-
justing the testing parameters (e.g., light intensity, temperature,
and gap distance) to stimulate the vat photopolymerization bio-
printing process. The change in storage modulus (G′)andloss
modulus (G″) was registered. As shown in Figure 4a–c, the
Emul and Non-Emul photoresins showed rapid crosslinking ki-
netics with gelation time (G′>G″)lessthan11s.Tobeno-
ticed, the gelation time of Emul photoresins is much shorter
than that of the Non-Emul photoresins. It is assumed that the
majority of GelMA is accumulated in the dispersed phase as
displayed in Figure S2a, Supporting Information, leading to
a higher local crosslinking density, thus accelerating the free-
radical chain polymerization and further improving the perfor-
mance in vat photopolymerization.[38] No significant differences
were observed in gelation time among the Emul photoresins.
However, we noticed that the gelation time increased with in-
creasing GGMMA content in the relatively homogeneous Non-
Emul photoresins.
The mechanical property of the hydrogels was investigated,
which is one of the critical biophysical cues that impact cell
fate.[39] The Non-Emul hydrogels with a more homogeneous net-
work exhibit higher transparency than the Emul hydrogels with
a heterogeneous network (Figure 4d). To be noticed, Emul 6_0
photoresin could also be fabricated into hydrogel with the ab-
sence of GGMMA. It is suggested that the small amount of
GelMA in the Dex phase participates in the photo-crosslinking
process and connecting of the GelMA microgels. We also noticed
that some irregularly shaped microgels appeared after photo-
crosslinking beside the sphere-shaped microgels (Figure S4,
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (6 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Supporting Information). Photopolymerization-induced phase
separation has been earlier reported in a PVA/Dex system.[40] It
is speculated that during the photo-crosslinking, GelMA in the
continuous phase might get separated and crosslinked with the
large GelMA droplets, which were likely to coalesce into a con-
tinuous and irregularly shaped hydrogel. Attributed to the dis-
placement of the microgels when a force was applied, the Emul
hydrogels showed better flexibility than the Non-Emul hydro-
gels, as shown in Figure 4e.[4] The elasticity was increased in
the Emul hydrogels with increasing the GGMMA content (Fig-
ure 4f and Table S1), where Young’s modulus of Emul 3_3 (4.2
±1.1 kPa) is almost twice of Emul 6_0 (2.3 ±0.3 kPa). In ad-
dition, a drastic decrease of G′of the crosslinked Emul pho-
toresins without the presence of GGMMA was confirmed when
compared to the Emul photoresins with GGMMA (Figure S5c,
Supporting Information). This evidence supports the hypothe-
sized role of GGMMA as a secondary polymeric network that in-
tegrates the microgels and limits the deformation and displace-
ment. It is necessary to determine whether the GGMMA occu-
pies the entire void between microgels and leaves sufficient space
for oxygen and nutrient transport to the cells and metabolic waste
removal. We further prepared Emul hydrogels and immersed
them into PBS for 24 h for Dex leaching, followed by perfusion
with fluorescein isothiocyanate-dextran (FITC-dextran) solution
before confocal microscope imaging. As shown in Figure 4g, the
FITC-dextran occupied the voids in the microgel-based hydrogels
formed by the Dex leaching. The porosities of the Emul hydro-
gels are in a range from 20% to 40%, as calculated by imaging
analysis (Figure 4i). The porosity variation of the Emul hydro-
gels strongly indicates their heterogeneous structure. The varia-
tion became smaller as the decrease of GelMA content from 6%
to 4.5%, which is consistent with a more homogenous GelMA
droplet size distribution (Figures 2c and 4i). We further recon-
structed confocal images to illustrate the porous structures of
the Emul hydrogels in hydrated status, as displayed in Figure 4g.
Continuous signals of FITC-dextran could be observed within
the testing area, indicating the interconnected pores. Whereas no
signal of FITC-dextran could be observed in the bulk hydrogels
(Figure 4h), suggesting the impermeability of the dense bulk hy-
drogel for a molecular substance that is in the equivalent size of
FITC-dextran (Mw =500 kDa). Ideally, the porosity of the Emul
hydrogels should be in line with the volume fraction of the Dex-
phase in the photoresins. Considering the fact that the presence
of GGMMA in the Dex-phase may occupy the void when Dex
leaches from the hydrogel, we further compared the cross-section
morphology of lyophilized Emul and Non-Emuls hydrogels. As
shown in Figure S6, Supporting Information, the microporous
structure of the Emul hydrogels was evident compared to the
Non-Emul hydrogels, which present a plain and non-porous mor-
phology. Even though it is worth mentioning that the morphology
of the hydrogel is difficult to preserve during the freeze-drying
process, and the pore structure visualized in the SEM imaging
may not be sufficient to reproduce the structure of the hydrogel
in the wet state,[32] the cross-section structure could still provide
indications. Also, we could observe that the pores in Emul 3_3
were much smaller than those in other Emul hydrogels, which
might owe to high GGMMA content and high GGMMA partition
in the dextran phase, as shown in Figures S2d and S6, Supporting
Information.
Emul photoresins were able to satisfy all prerequisites for
vat photopolymerization according to the rheological and photo-
rheological analysis. We further demonstrated that one-step
microgel assembly-based hydrogel printing could be facilely
achieved with the Emul photoresins in a DLP printer. The Emul
photoresins were gently pipetted for 10 s before loading to the
vat. The printing temperature was set at 30 °C, at which all
the Emul photoresins showed excellent flowability and rapid
photo-crosslinking kinetics, and the printing layer height was
set at 50 μm, which is larger than the majority of the emulsion
droplet size and thin enough for light penetration to minimize
the effect of sample transparency. 20-layered honeycomb struc-
tured hydrogels with structural integrity were printed using the
Emul photoresins, as shown in Figure 5. The microscopy im-
ages demonstrated that the printed hydrogels consisted of mi-
crogels as elementary building blocks. Some irregularly shaped
large microgels were observed in the Emul 6_0 and 5_1 hydrogels
(Figure 5a–III), which might owe to the phase separation that
occurred during photo-crosslinking. In comparison, the Emul
4.5_1.5 and 3_3 hydrogels consisted of microgels with a spher-
ical shape. We also noticed slight over-curing beyond the focal
plane in these hydrogels, whereas no over-curing was observed
in the printed Non-Emul hydrogels (Figure S7, Supporting Infor-
mation). Meanwhile, excess crosslinking was observed as much
less in Emul 4.5_1.5 and 3_3 hydrogels, as indicated in Figure 5b
where the Emul 4.5_1.5 and 3_3 hydrogels showed smaller beam
width compared to Emul 6_0 and 5_1 hydrogels. Presumably, the
relatively large GelMA microgels would lead to more severe light
diffraction, thus deteriorating the printing fidelity. We further
printed different structures with complex spatial connectivity us-
ing the Emul 4.5_1.5 and incubated them in a PBS buffer for 1
week to explore their long-shelf stability. All the printed hydrogel
scaffolds showed excellent stability that maintained the structural
integrity during the entire incubation period, and no swelling was
noticeable, as shown in Figure 5c.
2.2. Cytocompatibility of the Emul Bioresins
To test the biocompatibility of the Emul hydrogels in terms of cell
attachment, viability, spreading, proliferation, and migration, we
further cultured the normal human dermal fibroblasts (NHDFs)
within Emul photoresins. To examine whether the porous hydro-
gels integrated with microgel assembly as a network are bene-
ficial for stimulating the activity of the cells, that is, prolifera-
tion, ECM production, and migration, we compared them with
the Non-Emul hydrogels of similar composition. We facilely pre-
pared Emul and Non-Emul bioresins by mixing the Emul and
Non-Emul photoresins with the NHDFs and photo-crosslinked
the bioresins into hydrogels in a non-cell culture treated 96-well
plate. The NHDFs-laden Emul and Non-Emul hydrogels were
cultured for 14 days, and the Cell Counting Kit-8 (CCK-8) and
Live/Dead assay were applied to test the metabolic activity and
cell viability within the hydrogels. Compared to the Non-Emul
hydrogels, microgel-based Emul hydrogels showed enhanced cell
activity, as shown in Figure 6a by CCK-8 assay. This is believed to
be associated with the leaching of Dex away from the microgel-
based network during cell culture, which provides microporos-
ity for more efficient mass transfer and cell spreading, thus
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (7 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 5. a) Representative optical micrographs of the printed hydrogels (20-layer) using Emul photoresins demonstrating the microgel-based network.
Scale bars: I) 5 mm, II) 500 μm, and III) 200 μm. b) Measured beam width of the 3D printed hydrogels. c) Printed microgel-based hydrogel scaffolds
with different structures showing long-term stability and structural integrity in PBS for 7 days. Scale bars: 2 mm.
leading to superior surface area for cell growth. According to our
results, decreasing the GelMA content in the hydrogel compo-
sition caused a decrease in cell viability, especially for Non-Emul
hydrogels. However, this phenomenon was not valid for the Emul
hydrogels on Day 7 and 14, where the activity of NHDFs was sim-
ilar except for the Emul 3_3 hydrogels. More cells were able to
attach to the hydrogels containing a higher content of GelMA in
the early days of the cell culture. This suggests that the high affin-
ity of cells to the RGD sequences presented by GelMA is crit-
ical for cell attachment at early time points. The voids formed
within the Emul hydrogels during the culturing provide a spa-
tial cue to guide the cell-biomaterial interactions, as the prolifera-
tion rate of NHDFs in the later stages of the culture compensated
the difference between different GelMA content in hydrogels by
spreading into the voids and growing on the available surface
provided by the assembled microgels. As displayed in Figure 6c,
the cells encapsulated in the Emul hydrogel showed better cell
spreading on Day 7, compared to the cells in the Non-Emul hy-
drogel (Figure 6d). To understand how cells organize themselves
in the Emul hydrogels, we analyzed the colocalization of NHDFs
in the microgels. As shown in Figure S9, Supporting Informa-
tion, Pearson’s cross-correlation of the cells in the hydrogel was
zero or slightly negative, indicating that the cells have no or nega-
tive association with the GelMA microgels already on Day 1. The
distribution did not change during a 7-day culture period, and
the cells mostly preferred to spread onto the surface of the mi-
crogels rather than penetrate into them. Meanwhile, cells in the
Emul 3_3 hydrogels showed less activity in comparison with the
Emul 5_1 and Emul 4.5_1.5 hydrogels, and the cells were also less
elongated. To quantify the observation, we measured the round-
ness of the cells by analyzing the microscope images with Im-
ageJ software to understand the viability and metabolic activity
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (8 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 6. Cytocompatibility of the Emul and Non-Emul bioresins. a) Metabolic activity of cell-laden Emul and Non-Emul hydrogels during 14 days culture
as determined by the CCK-8 assay. Columns from left to right represent the optical density (O.D.) value on Day 1, 3, 7, and 14. b) ALP activity of MC3T3-E1
cells in Emul and Non-Emul 4.5_1.5 hydrogels on Day 7. Representative fluorescence microscopy images showing the viability of NHDFs encapsulated
in c) Emul and d) Non-Emul hydrogels on Day 1 and 7. (live: green, dead: red; scale bars: 200 μm)
data. Fibroblasts present a round morphology in bulk hydrogels
and culture substrates that do not support their biological activity,
but when they attach and spread on the scaffold, their morphol-
ogy changes from a spherical shape to a more spindle-like shape.
It was observed in Figure S10, Supporting Information, that for
the cells grown in Emul 3_3, the majority of the cells showed a
spherical shape. In contrast, the cells in other Emul hydrogels re-
sembled more typical fibroblasts, which is in agreement with the
metabolic activity data. It is suspected that material characteris-
tics of the microgel assembly in the Emul 3_3 hydrogel, for ex-
ample, microgel dimension, matrix mechanics, or organization,
might not be the most suitable for supporting the NHDFs be-
havior, due to a relatively high content of GGMMA in Dex phase
occupying the void left by the leaching of Dex and result in a rel-
atively dense structure with small pores (Figures S2 and S6, Sup-
porting Information). This compact structure also restricts cell
migration and spreading, and leads to low cell activity and im-
mobile and inactive cells with higher roundness (Figure 6 and
Figure S10, Supporting Information). In addition to the viabil-
ity and morphology of the cells in Emul and Non-Emul hydro-
gels, we also evaluated the functionality of the MC3T3-E1 pre-
osteoblast cells by measuring the enzyme activity of bone alka-
line phosphatase (ALP) secreted to the medium. As shown in Fig-
ure 6b, cells seeded into Emul 4.5_1.5 hydrogels have produced
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (9 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Figure 7. a) Representative fluorescence microscopy images showing the viability of bioprinted microgel-based hydrogel using the Emul 4.5_1.5 bioresin
encapsulated with the NHDFs and MC3T3-E1 cells on Day 1, 3, and 7 (live: green, dead: red). Scale bars: 200 μm. b) Quantifications of cell viability. c)
Fluorescence microscopy images (i) and (3D reconstructed images (ii) of cytoskeleton staining demonstrating NHDFs and MC3T3-E1 spreading in the
bioprinted bulk (Non-Emul 4.5_1.5) and microgel-based (Emul 4.5_1.5) hydrogel on Day 1 and 7/14 (F-actin: green, nuclei: blue). Scale bars: (i) 100 μm,
(ii) 150 μm.
significantly more bone ALP than the cells in Non-Emul 4.5_1.5
hydrogels by Day 7.
2.3. Bioprinting of the Emul Bioresin using Vat
Photopolymerization
Twenty-layered honeycomb structured hydrogels (50 μm layer
height) were printed with the Emul 4.5_1.5 bioresin encapsu-
lated with two cell lines, respectively, using a DLP printer to
demonstrate the cytocompatibility of the bioprinting process by
Live/Dead assay. As shown in Figure 7a, the bioprinted structure
showed great structure integrity, and both NHDFs and MC3T3-
E1 cells showed a pronounced cell proliferation within the Emul
4.5_1.5 hydrogel after 7 days culturing period. Moreover, we
could clearly observe in Figure 7a that the MC3T3-E1 cells spread
along the surface of microgels on Day 7. The cytocompatibil-
ity of Emul 4.5_1.5 bioresin applied in vat photopolymerization
bioprinting was confirmed by Live/Dead staining. The NHDFs
showed great cell viability (over 90%) throughout the culturing
period with no significant difference between Day 1 (92.2 ±
3.7%), Day 3 (91.4 ±2.3%), and Day 7 (92.3 ±2.9%). Whereas, the
viability of MC3T3-E1 cells was lower compared to the NHDFs
on Day 1 (82.4 ±5.6%), but it was well maintained on Day 3
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (10 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
(92.4 ±2.4%) and Day 7 (90.9 ±2.0%), as indicated in Figure 7b.
Notwithstanding, the microporous hydrogels provided good 3D
culture conditions to two different cell types.
To demonstrate the advantages of porous microgel assembly-
based hydrogels for the construction of biomimicking microen-
vironments that provide spatial cues, we further used cytoskele-
ton staining to evaluate the spreading and morphology of en-
capsulated cells within the bioprinted Emul 4.5_1.5 hydrogels.
As clearly observed in Figure 7c, the encapsulated cell types
showed enhanced spreading within the porous hydrogel con-
structs compared to the bulk hydrogels. The cells encapsulated
in the microgel-based hydrogel showed a more elongated mor-
phology than those in the bulk hydrogel, even on Day 1. After a 7
or 14-day culturing period, in bulk hydrogels only the cells on the
surface of the printed scaffold struts showed elongated morphol-
ogy reminiscent of fibroblasts in vivo (Figure 7c), whereas the
cells embedded in the hydrogel remained round and proliferated
less. In comparison, in the microporous hydrogel, the cells pro-
liferated and migrated both along the surface and within the hy-
drogel space (Figure 7c). Especially the MC3T3-E1 cells showed
a high cell density and active fibroblastic morphology on Day 14,
both on surfaces and infiltrated within the microgel-based net-
work to form cell-cell connections, as shown in the 3D recon-
structed images (Figure 7c). Sustaining high cell density, sites
for bioadherence, and providing relevant matrix stiffness are key
characteristics of successful bioinks/bioresins in service of fabri-
cating 3D cell culture models via 3D bioprinting.
3. Discussion
As summarized in the Introduction section, the superior-
ity of microgel-based 3D bioprinting in biofabrication has
been evidenced by many established proof-of-concept research
showcases.[4,11,12 ] Importantly, the inherent void-forming feature
of microgel assembly rationally supports the outperforming ca-
pabilities of these bottom-up biomaterial systems in support-
ing the viability and functions of the laden cells. As elementary
building blocks, tunability to the material properties of microgels
also makes it a high-efficiency approach in constructing tissue
mimics in regard to reproducing the biologically relevant mor-
phology and functions. However, such challenges as the com-
plex preparation of microgel-based bioink and difficulty in main-
taining the porosity of the bioprinted hydrogels while ensuring
their stability in an aqueous medium, are still awaiting suitable
bioink developments and printing methods to address. ATPE
is an extremely easy-to-operate system in preparation of micro-
gels that can create the aqueous-aqueous interfaces facilely by
mild agitation like pipetting. In a desktop DLP printer, the ATPE
bioresins of GelMA/GGMMA/Dex can effortlessly realize a one-
step workflow of fabricating and assembling the GelMA micro-
gels upon the vat polymerization in situ while encapsulating the
laden cells within the hydrogel construct. To the best of the au-
thors’ knowledge, this is the first DLP-based demonstration of
a microgel assembly-integrated bioprinting to an ATPE bioresin
system, demonstrating satisfying printability and fidelity. It is
worth highlighting that a successful DLP bioprinting output is
delivered with a relatively low concentration of photo-biopolymer
at 6% w/v in the resin formulation. This is lower than the utiliza-
tion of 8–12% GelMA normally deployed in bioink formulations.
As reported elsewhere, the overly high viscosity of GelMA (10
wt%) prior to photocrosslinking will challenge the recoating pro-
cessing in DLP vat, leading to none processable DLP printing.[25]
Meanwhile, GGMMA, derived from a plant-based natural poly-
mer of volumetric availability, is introduced to partially replace
the GelMA (0-50% relative to GelMA) as a cost-effective additive
photo-biopolymer in the bioresin formulation. When the struc-
tural integrity of the 3D bioprinted construct is warranted, a low
concentration of photo-crosslinkable biopolymers in the hydro-
gel can favor the molecular diffusion and convectional fluid flow
of nutrients and soluble signaling mediators in cell culture.
The strategic purposes of introducing GGMMA into the
GelMA/Dex ATPE bioresin go far beyond the aforementioned
cost-effective consideration for bioresin development. GGMMA
is seen as a multi-function modulator in the APTE bioresin for-
mulation. First, GGMMA functions as an emulgator and sig-
nificantly improves the stability of the GelMA/Dex ATPE (Fig-
ure 2a,d). In general, instability is considered the main drawback
of the ATPE, owing to the low interfacial tension between the two
aqueous phases.[24] The emulsion droplets are very likely to coa-
lesce and lead to a phase separation in a short time, thus shorten-
ing the printing window.[23] Very recently, systematic efforts have
also been made to gain an in-depth understanding of how mul-
tiple parameters in GelMA-based ATPE bioresin system govern
the emulsion stability, for example, GelMA type, methacryloyl-
modification degree, concentration, and molecular weight of
porogen (PEO, polyvinyl alcohol, or Dex).[25] In recent studies,
biosurfactants and Pickering emulsion strategies have been stud-
ied to stabilize the ATPE bioresins.[26,27] The most intuitive en-
hancement of the stability is to improve their printability, and
meanwhile the functionalized stabilizer can endow value-added
functionality to the printed hydrogels.[26,28 ] Overall, these studies
all showcase the 3D bioprinting scenarios where microporous
hydrogel constructs were bioprinted in DLP printing with the
GelMA forming a continuous and crosslinked network, but the
porogen polymer leaching away to provide microporosity.[25,27]
Although we also adopt ATPE in microgel assembly-integrated
DLP bioprinting, the aforementioned aspects to improve the
emulsion stabilization cannot be directly translated into our
system because of the diametrically opposite basic logic. Sec-
ondly, from the perspective of biofabrication, the multifunctional
GGMMA is projected as photo-adhesive to sufficiently intercon-
nect the GelMA microgels and to guarantee the printability of
bioresin in DLP printing as well as the structural stability after
removal of Dex in culture medium. As shown in Figure 4e,f,
and Figure S5, Supporting Information, the content of GGMMA
upregulates the macroscopic mechanical properties of microgel-
based hydrogels when the GelMA content is decreased. Benefit-
ing from all the above-discussed perspectives, satisfactory print-
ability of the ATPE bioresins was demonstrated by DLP-based
printing to create multilayer constructs with complex patterns
and sharp edges (as displayed in Figure 5c). To be noticed, photo-
crosslinkable bioresins derived from ichthyic-origin gelatin were
developed and favored in DLP bioprinting attributed to their
flowability at room temperature with a concentration of 15%
w/v.[41] In future research work, it is also necessary to consider
using this type of gelatin-based material in the current ATPE sys-
tem to obtain microgel-based hydrogels with broad mechanical
strength and microgel size.
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (11 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
After the removal of Dex in the culturing, the porous struc-
ture of the microgel-based hydrogel was confirmed to support the
metabolic activities of cells and promote their migration within
the 3D hydrogel. As revealed by the FITC-dextran perfusion anal-
ysis (Figure 4g,i), the microgel-based hydrogel showed a superior
porosity of 20% to 40%, which might be inherited from the vol-
ume fraction occupied by the Dex phase in the ATPE bioresin.[42]
The porosity is higher than previously reported for the hydrogels
constructed with mechanically fragmented granular microgels
(9% to 21%) or hydrogel microstrands (2% to 8%).[12,22 ] Although
no conclusive comparison can be made among these reported
values due to the difference in the respective analysis method for
porosity, it still reflects the superior porosity due to the difference
in the bioresin and bioprinted method as proposed. No microgel
jamming occurs during the DLP-bioprinting, and the leaching
of Dex from microgel-based hydrogel could create native voids,
thus enabling the remarkable porosity of the bioprinted hydro-
gel. Our reported porosity is equivalent to what was reported for
microporous hydrogels prepared by annealing the spherical mi-
crogel particles fabricated through microfluidics or electrospray-
ing methods (20% to 40%).[20,43,44 ] It might raise concern that
the GGMMA distributed in the dextran phase might occupy the
void, thus affecting the microporosity of the Emul hydrogels.
We further applied SEM to validate whether the GGMMA oc-
cupies all the pores while connecting the GelMA microgels. As
shown in Figure S6, Supporting Information, all the Emul hy-
drogels showed a microporous structure. However, compared to
other Emul hydrogels, Emul 3-3 showed a rather plain structure
with small pores. This indicates that a higher concentration of
GGMMA would occupy the void between the microgels and lead
to a dense structure. We have demonstrated that the pores in
low GGMMA containing Emul hydrogels could allow more ef-
ficient exchanges of nutrients and wastes and provide sufficient
spaces for cell migration and spreading. The metabolic activity
of the NHDFs embedded in the microgel-based hydrogel was
higher than those embedded in bulk hydrogel, especially on Day
7, where the heterogeneous and porous microgel-based hydro-
gels showed a tremendous advantage over the bulk hydrogels.
In this study, the cells are directly mixed with Emul photoresins
to formulate the bioresins. We expected cells to be evenly dis-
persed in photoresins prior to the photocrosslinking of the hy-
drogel. The colocalization analysis performed based on the mi-
croscope images indicates that the cells are located within the
microgels. The cells grew in the Emul hydrogel mostly indepen-
dent of the GelMA microgels but often adhering to them. To be
noticed when applying GGMMA as a photo-adhesive, the dosage
should be adjusted appropriately. Indeed, within the Emul hydro-
gels, high GGMMA content did improve the mechanical prop-
erty by connecting the GelMA microgels, especially for the Emul
hydrogels, like Emul 3_3 with a low volume fraction of GelMA
phase, as shown in Figure S5c, Supporting Information. How-
ever, the relatively dense porous structure of Emul 3_3 might not
seriously affect the diffusion of nutrients and wastes. Still, it af-
fected the growth and migration of cells compared to other Emul
hydrogels. Overall, this one-step bioprinting process greatly sup-
ports the survival of the laden NHDFs or MC3T3-E1 cells as sug-
gested by Live/Dead assay. Inthe conventional microgel bioprint-
ing processes, the cells would undergo tedious microgel-based
bioink preparation and jamming, pre-cooling, and shearing dur-
ing the extrusion process, which might potentially harm the cell,
thus leading to a cell viability of 70% to 80%.[4,12,15 ] However, at-
tributed to the facile bioresin preparation, relatively high tem-
perature, and low shearing associated with this DLP-bioprinting,
the viability of MC3T3-E1 cells could reach 82.4 ±5.5%. More-
over, the viability of NHDFs could reach 92.2 ±3.7%. We further
demonstrate different spreading behaviors of the cells printed
with microgel-based and bulk hydrogels using cytoskeleton stain-
ing. The result correlates well with the cell metabolic activity data
and indicates that the microporosity alone in the microgel-based
hydrogels could permit cellular spreading at even Day 1. The cell
number was less than that in the bulk hydrogel, possibly due to
the cells releasing with the Dex phase. However, the difference
becomes increasingly significant, where the cells in bulk hydro-
gels only spread and proliferated along the outer surface that is
contacted with the culture medium, and the cells in the inner part
did not spread. However, cells in the microgel-based hydrogel not
only spread along the outer surface but also bridged multi micro-
gels and filled the voids. This provides a larger active area within
the hydrogels and higher cellular activity beneficial for applica-
tions that aim for ECM generation.
When the cultured cells grow along the surface of the micro-
gels in their assembly, the microgel size is a critical metric to
consider. It determines the surface-to-volume ratio of microgels
and acts as a microtopographical cue that can guide cell spread-
ing and migration.[20,45 ] Changing microgel size on one side can
control scaffold architecture and, on the other side, would also
modify the topographical curvature of the surface that cells ad-
here to. As reported previously, NHDFs showed a significant dif-
ference in cell spreading and proliferation, when they were cul-
tured on the microgel assemblies comprising different-sized mi-
crogel populations of hyaluronic acid: cells wrapped around mi-
crogels and adopted a more flattened shape in the large microgel
group (diameter in 60–200 μm) but showed less spreading and
lower levels of proliferation in small microgel group (diameter
in 20–60 μm).[20] In this regard, various microgel size distribu-
tions could be obtained by simply adjusting the GGMMA/GelMA
content in the as-proposed ATPE bioresins, as displayed in Fig-
ure 2b,c. For these ATPE bioresins, a heterogeneous nature in
the size distribution of GelMA droplets resulting from gently
pipetting has to be underlined and the same observation has
also been evidenced in similar systems.[16,46 ] However, the het-
erogeneity was observed to decrease with respect to the increased
ratio of GGMMA/GelMA content. The microgel size heterogene-
ity is inherent to the constructed microgel assembly, which pro-
vides a specific structural heterogeneity that might be of inter-
est when developing in vitro co-culture models involving multi-
cell types. Meanwhile, the material stiffness for both individ-
ual microgel and macroscopic hydrogel constructs is regulated
by the ratio of GGMMA/GelMA content. Thus, it is interesting
to use the as-proposed ATPE bioresins in multi-resin/multi-vat
DLP bioprinting to verify the possibilities of rapidly constructing
microgel-based hydrogel constructs with heterogeneous material
properties.[47]
4. Conclusion
Creating a void-forming hydrogel on the macroscopic scale while
keeping a continuous polymeric network for crosslinking is a
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (12 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
viable strategy for expanding the biofabrication window, with bal-
ancing between the printing fidelity of a bioink and compatible-
ness of the hydrogel’s ultrastructure with the metabolic activities
of the encapsulated cells. In our ternary-component bioresin, the
partitioning of GGMMA in both the dispersed phase of GelMA
droplets and the continuous phase of Dex is key to improving
the stability of the aqueous emulsion. In DLP printing, GGMMA
provides a photopolymerizable network to adhere the GelMA mi-
crogels converted upon photocrosslinking as a macroscopic hy-
drogel construct and it warrants a good output of printing fidelity
in the scenario of using a low content of bio-photopolymers in
bioresin formulation. Importantly, the removal of Dex by leach-
ing to the culture medium provides the microporosity within the
hydrogel construct, which is proven to better support the pro-
liferation and infiltration of the encapsulated cells to colonize
the 3D porous hydrogel matrices, mainly residing on the sur-
face of the microgels. High content of GGMMA would influ-
ence the pore structure and lead to relatively small pore size,
and suppress the cells’ metabolic activity. Meanwhile, the in-
creased ratio of GGMMA/GelMA resulted in the diameter of
GelMA microgels shifting downwards as well as a stronger me-
chanical stiffness of the printed hydrogel construct, which col-
lectively dictates the printing fidelity and in-growth of the laden
cells within the porous hydrogel construct. All in all, the aqueous
emulsion bioresin of GelMA/GGMMA/Dex amends a functional
and facile-to-operate strategy that can realize a one-step workflow
of fabricating and assembling GelMA microgels conveniently in
the DLP-based cell-laden bioprinting.
5. Experimental Section
Chemicals:All the chemicals were purchased from Sigma-Aldrich un-
less otherwise stated.
Synthesis of GelMA and Rhodamine B Conjugated GelMA:GelMA with
a degree of methacryloylation (DM) of 0.26 mmol g−1was synthesized
from porcine gelatine (Type A, ≈300 g bloom, Sigma-Aldrich) according to
the previously reported method.[48] Rhodamine B conjugated GelMA (R-
GelMA) was synthesized according to a reported method.[49] Briefly, 1.0 g
of GelMA was dissolved in 100 mL sodium bicarbonate solution (0.1 M)
at 37 °C overnight. After cooling to room temperature, 1 mL rhodamine
B isothiocyanate solution (10 mg mL−1in DMSO) was added into the
GelMA solution and reacted for 24 h in the dark.
Synthesis of GGMMA and Fluoresceinamine Conjugated GGMMA:
GGMMA with a DM of 0.46 mmol g−1was synthesized from galactogluco-
mannan (GGM) (number-averaged molecular weight was 9 kDa, hot water
extracted) according to our previous study with minor modification.[50]
Briefly, 1.0 g of GGM was dissolved in 100 mL distilled water at 50
°C overnight. 0.33 mL of methacrylic anhydrate was added and reacted
at 50 °C for 40 min in the dark. The pH was maintained at 8 us-
ing a 5 M NaOH solution. Fluoresceinamine conjugated GGMMA (F-
GGMMA) was synthesized by dissolving 1 g of GGMMA into 20 mL 4-
morpholineethanesulfonic acid solution (0.1 M) at room temperature for
1 h. The GGMMA was activated by reaction with 1 mL activation solu-
tion (0.1 g N-hydroxysuccinimide and 0.3 g N-(3-dimethylaminopropyl)-
N’-ethylcarbodiimide hydrochloride in 1 mL DMSO) for 15 min under vig-
orous stirring. Then, 0.5 mL fluoresceinamine, isomer I solution (20 mg
mL−1in DMSO) was added into the GGMMA solution and reacted for 24
h in the dark.
All the as-synthesized photo-crosslinkable polymers were dialyzed
against distilled water for 7 days in the dark (GelMA and R-GelMA at
40 °C). GelMA and GGMMA solutions were filtrated through a filter
(0.22 μm, nylon, sterile, Sartorius Lab Instruments) in advance. The ob-
tained lyophilized photo-crosslinkable polymers were stored under −20
°C in the dark before use. The DM of GelMA and GGMMA was character-
ized by 1H-NMR, and the results were listed in the Supporting Information
(Figure S8, Supporting Information).
ATPE Photoresin Preparation and Characterizations:The lyophilized
GelMA and GGMMA, Dex (Mw =250 kDa, Alfa Aesar), and lithium phenyl-
2,4,6-trimethylbenzoylphosphinate (LAP, 98%, TCI) were dissolved into
PBS separately to obtain the stock solutions with concentrations of 12, 12,
13.3, and 2% w/v, respectively. Then the stock solutions were mixed to fi-
nal concentrations, as shown in Table 1 below and pipetted gently for 10
s to obtain the Emul and Non-Emul photoresins. The surface tension of
GGMMA solution against air was analyzed by a Force Tensiometer (K100,
KRÜSS GmbH) using a Wilhelmy plate at room temperature.
Stability Analysis:The stability of the photoresins at 25 °C within
90 min was characterized by the Turbiscan Lab Expert stability analyzer
(Formulaction) equipped with a pulsed near-infrared light source (𝜆=
880 nm) with 25 s per scan. The transmitted signals of the test sam-
ples were registered along the sample height and used to calculate their
TSI using the Turbiscan software to compare their stability. TSI value
greater than 10 indicates the destabilization corresponding to sedimen-
tation/creaming and phase separation of the photoresins is visible.
Fluorescent photoresins were prepared by replacing GelMA and
GGMMA with R-GelMA and F-GGMMA (5% w/w), respectively. A 3i Mar-
ianas CSU-W1 spinning disk confocal microscope (50 μm pinholes, Intel-
ligent Imaging Innovations GmbH) was used to obtain the fluorescence
microscopy images of the morphology of photoresins.
GGMMA Distribution Analysis:The GGMMA containing Emul pho-
toresins were demulsified by centrifuging at 8000 rpm for 10 min and
incubated at 37 °C for 3 days to reach the equilibrium. Testing samples
were carefully taken from each phase and freeze-dried for quantitative
sugar analysis by acid methanolysis and gas chromatography.[51] Man-
nose, as a marker of GGMMA’s characteristic component, was employed
to quantitatively determine the distribution of GGMMA in each phase. The
GGMMA accumulated in the interphase was calculated by subtracting the
GGMMA amount in the GelMA and Dex phases from the total amount.
The main components in GGMMA are attached in Table S2, Supporting
Information.
Rheological Characterization:Rheological measurements of the pho-
toresins were registered by an Anton Paar Physica MCR 702 rheometer
(Anton Parr GmbH) with a plate-plate geometry (25 mm diameter) with a
gap distance of 0.5 mm at 25 and 30 °C. The viscosity curves were recorded
by shear flow measurement with a shear rate ramping from 1 to 1000 s−1.
The thermal gelation point was determined by an oscillatory temperature
sweep from 37 to 5 °Cwith2°Cmin
−1. The shear strain and frequency were
fixed at 5% and 1.5 Hz, respectively, as selected in the linear viscoelastic
range. The photoresins were pre-sheared at 200 s−1for 30 s and rested
for 60 s to reach equilibrium before testing. Photo-crosslinking kinetics
were registered under oscillation mode with a gap distance of 50 μmata
constant oscillatory strain and frequency of 5% and 1.5 Hz, respectively.
A 405 nm light source (32 mW cm−2, bluepoint LED eco, Hönle Group)
started irradiation upon the samples through a transparent bottom plate
from 30 s. The change in modulus was recorded at 30 °C. The gelation time
was determined from the beginning of the exposure to the time point when
G’ >G’’. All measurements were carried out at least in triplicate.
Mechanical Testing:The mechanical compression test was carried out
on cast hydrogel discs (diameter: 4.65 mm; height: 2 mm) using a dynamic
mechanical analyzer (MCR 702 Multidrive, Anton Paar GmbH) at 25 °C.
A 1 mN preload was applied to ensure complete contact between the hy-
drogel discs and the measuring plate. Hydrogel discs were compressed at
a speed of 10 μms
−1. Young’s modulus was calculated by linear fitting of
the initial elastic region (5–15%) of the stress-strain curves.
Porosity Analysis:The porosity of hydrogels was analyzed by FITC-
dextran (Mw =500 kDa) perfusion. Hydrogels were incubated in PBS for
24hat37°C until the Dex phase was leached out. Then, hydrogels were
perfused with FITC-dextran solution (1 mg mL−1)for24hat4°Cbe-
fore taking confocal images using a 3i Marianas CSU-W1 spinning disk
confocal microscope (50 μm pinholes, Intelligent Imaging Innovations
GmbH). The ratio of the fluorescent area in each scan step on the z-stack of
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (13 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
confocal images was determined using ImageJ software and averaged to
the whole stacks (60 μm in depth with a 1 μm gap between the adjacent
steps) for porosity calculation.[6,52]
SEM imaging was applied to reveal the microstructure of the Emul and
Non-Emul hydrogels. Emul and Non-Emul hydrogels were incubated in
PBS for 24 h at 37 °C until the Dex phase leached out and reached equilib-
rium. Then cross-section samples were prepared using the freeze-fracture
method in liquid nitrogen. The SEM analysis was carried out with an LEO
1530 scanning electron microanalyzer at an accelerating voltage of 5.0 kV.
Cell Culture:All cell culture reagents were purchased from Gibco un-
less otherwise stated.
NHDFs (adult, cryopreserved, C-12302) were purchased from Promo-
Cell. MC3T3-E1 cell line from mouse (99072810) was purchased from
Sigma-Aldrich. NHDFs were cultured in a complete cell culture medium
containing Dulbecco’s modified Eagle’s medium (DMEM) with fetal
bovine serum (FBS, 10%) and penicillin-streptomycin (Pen-strep, 100 U
mL−1). NHDFs were passaged about twice a week, and the cell culture
medium was replaced every other day. MC3T3-E1 cells were cultured in
a complete cell culture medium containing minimum essential medium
(MEM 𝛼) with FBS (10%), glutaMAX (2 mM), and Pen-strep (100 U mL−1).
MC3T3-E1 cells were passaged about once a week, and the cell culture
medium was replaced every other day. All the cells were cultured in a hu-
midified incubator with a CO2atmosphere (5%) at a constant temperature
of 37 °C. NHDFs before passage 12 and MC3T3-E1 cells before passage
20 were used in this work.
In Vitro Cytocompatibility Evaluation of the Emul and Non-Emul Bioresin:
NHDFs were harvested and resuspended into the Emul and Non-Emul
photoresin by gently pipetting to achieve well-suspended bioresins with
a cell density of 8.3 ×105cells per mL−1. Then bioresins were loaded
into a non-cell culture treated 96 well-plate to eliminate the metabolic ac-
tivity signals from the cells that attached to the plate surface (Nunc Mi-
croWell 96-Well Microplates, Thermo Scientific; 60 μL bioresin containing
1×105cells per well) and crosslinked into cell-laden hydrogels using a
large-format curing light source (405 nm, EFL-LS1602, Yongqinquan Intel-
ligent Equipment) at 45 mW cm−2for 90 s. 150 uL of complete cell culture
medium was added to each well and replaced every other day. The hydro-
gels were cultured for 1, 3, 7, and 14 days and NHDFs proliferation within
the Emul and Non-Emul hydrogels was determined using Cell Counting
Kit-8 (CK04-11, Dojindo). Briefly, the CCK-8 testing reagent was prepared
by diluting the CCK-8 agent by 10 times with the complete cell culture
medium. At certain time points, the cell culture medium in each well was
replaced by CCK-8 testing reagent-containing medium and the cell-laden
hydrogels were incubated for 4 h in the incubator. The medium was trans-
ferred to another plate for absorbance reading at 450 nm using spectral
scanning multimode reader (Varioskan Flash, Thermo Scientific). Cell vi-
ability was determined by the Live/Dead assay using the Live/Dead Cell
Staining Kit II (PromoKine). Briefly, the hydrogels were immersed in 200 μL
testing reagent (0.5 μM Calcein-AM/1.6 μM EthD-III diluted with PBS) at
room temperature for 3 h after washing the hydrogel with PBS. A 3i Mar-
ianas CSU-W1 spinning disk confocal microscope (50 μm pinholes, Intel-
ligent Imaging Innovations GmbH) was used to obtain the fluorescence
microscopy images of cells encapsulated in the inner part of the hydro-
gels (3.01 μm between each step, 50 steps in Z direction). Cell viability,
roundness, and colocalization were quantified using Fiji ImageJ software,
by particle analysis and calculating shape descriptors and Coloctest2, re-
spectively.
ALP Activity Analysis:MC3T3-E1 cells were harvested and resus-
pended separately into the Emul and Non-Emul 4.5_1.5 photoresin by
gently pipetting to achieve well-suspended Emul and Non-Emul 4.5_1.5
bioresins with a cell density of 8.3 ×105cells per mL. The bioresins were
crosslinked into non-cell culture treated 96 well-plate and cultured using
(MEM 𝛼with 10% FBS, 10 μM𝛽-glycerophosphate, 30 ng mL−1BMP-2,
0.25 mM ascorbic acid, and 10−5mM dexamethasone) for 7 days. At each
desired time point, 25 μL of culture medium was collected and reacted with
the 100 μL p-nitrophenol phosphate substrate at 37 °Cfor1h.Thereac-
tion was then terminated by adding 50 μL of 1 M NaOH solution and the
resulting samples were further reading the absorbance at 405 nm using
spectral scanning multimode reader (Varioskan Flash, Thermo Scientific).
Bioprinting:NHDFs and MC3T3-E1 cells were harvested and resus-
pended separately into the Emul and Non-Emul 4.5_1.5 photoresin by
gently pipetting to achieve well-suspended Emul and Non-Emul 4.5_1.5
bioresins with a cell density of 1.4 ×106cells per mL. 0.5 mM tartrazine
was involved as the photo absorber to prevent excess crosslinking. The
bioprinting was performed with a DLP 3D printer (405 nm, M-One Pro
30, MAKEX) equipped with a customized temperature control device that
maintains at 30 °C. A honeycomb-structured hydrogel was printed with
the printing parameters: light intensity of 32 mW cm−2, exposure time of
10–11 s, and a layer height of 50 μm. After printing, the printed constructs
were washed with PBS and incubated in the corresponding cell culture
medium in a humidified incubator with a CO2atmosphereof5%atacon-
stant temperature of 37 °C. The cell culture medium was replaced every
other day. The cell viability of the printed construct at certain time points
was quantified by the Live/Dead assay, as mentioned above using Fiji Im-
ageJ software.
Cytoskeleton Staining:Cytoskeleton staining was used to investigate
the morphologies and spreading of NHDFs and MC3T3-E1 cells within the
printed constructs. Briefly, cell-laden hydrogels printed by Emul and Non-
Emul 4.5_1.5 bioresins were cultured to certain time points and fixed with
4% paraformaldehyde for 25 min at room temperature. The cells within the
hydrogels were permeabilized with 0.1% Triton X-100 solution for 5 min at
4°C, followed by incubation in 1% bovine serum albumin solution at room
temperature for 1 h. The fixed hydrogels were subjected to F-actin staining
with Alexa 488-phalloidin dye (330 nM, Cell Signaling Technology) for 1 h
and nuclei staining with DAPI (300 nM, Cayman Chemical) for 10 min.
The stained hydrogels were washed with PBS and stored at 4 °Cinadark
place. A 3i Marianas CSU-W1 spinning disk confocal microscope (50 μm
pinholes, Intelligent Imaging Innovations GmbH) was used to obtain the
fluorescence microscopy images of stained cells (3.01 μm between each
step, 50 steps in Z direction).
Statistical Analysis:The statistical analysis of all the data has been per-
formed using one-way or two-way analysis of variance (ANOVA) by Graph-
Pad Prism 9. (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001). All
the experimental data were repeated at least three times and are reported
as the mean ±SD.
Supporting Information
Supporting Information is available from the Wiley Online Library or from
the author.
Acknowledgements
Q.W. would like to acknowledge the financial support from the China
Scholarship Council (Student ID 201907960002) and the Finnish Ministry
of Education and Culture funded international pilot project “Finland-
China Food and Health Network” to his doctoral study at Åbo Akademi
University (ÅAU), Finland. X.W. would like to thank the Academy of
Finland (333158) as well as the Jane and Aatos Erkko Foundation for
their funds for her research at ÅAU. T.N. and Ö.K. would like to thank the
Olins Foundation within Swedish Cultural Foundation in Finland and the
Finnish Society for Sciences for supporting their research. J.M.R.and C.X.
acknowledge the REACT EU funding received from the European Regional
Development Fund (ERDF) for the project “AMBioPharma” (Centre for
Additive Manufacturing for Life Science and Pharmaceutical Industry,
project code A77805).
The copyright line for this article was changed on 13 October 2023 after
original online publication.
Conflict of Interest
The authors declare no conflict of interest.
Data Availability Statement
The data that support the findings of this study are available from the cor-
responding author upon reasonable request.
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (14 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
www.advancedsciencenews.com www.advhealthmat.de
Keywords
3D bioprinting, aqueous emulsion, microgel assembly, microporous hy-
drogels, vat photopolymerization
Received: December 13, 2022
Revised: February 28, 2023
Published online: March 28, 2023
[1] L. Ouyang, J. P. K. Armstrong, Y. Lin, J. P. Wojciechowski, C. Lee-
Reeves, D. Hachim, K. Zhou, J. A. Burdick, M. M. Stevens, Sci. Adv.
2020,6, 5529.
[2] S. Mehrotra, B. A. G. de Melo, M. Hirano, W. Keung, R. A. Li, B. B.
Mandal, S. R. Shin, Adv. Funct. Mater. 2020,30, 1907436.
[3] K.S.Lim,J.H.Galarraga,X.Cui,G.C.J.Lindberg,J.A.Burdick,T.B.
F. Woodfield, Chem. Rev. 2020,120, 10662.
[4] Y. Fang, Y. Guo, M. Ji, B. Li, Y. Guo, J. Zhu, T. Zhang, Z. Xiong, Adv.
Funct. Mater. 2021, 2109810.
[5] P. Dorishetty, N. K. Dutta, N. R. Choudhury, Adv. Colloid Interface Sci.
2020,281, 102163.
[6] H. Liu, P. Chansoria, P. Delrot, E. Angelidakis, R. Rizzo, D. Rütsche,
L. Ann Applegate, D. Loterie, M. Zenobi-Wong, Adv. Mater. 2022,
2204301.
[7] A. K. Gaharwar, I. Singh, A. Khademhosseini, Nat. Rev. Mater. 2020,
5, 686.
[8] M. Machour, N. Hen, I. Goldfracht, D. Safina, M. Davidovich-Pinhas,
H. Bianco-Peled, S. Levenberg, Adv. Sci. 2022,9, 2200882.
[9] B. R. Freedman, N. D. Bade, C. N. Riggin, S. Zhang, P. G. Haines, K. L.
Ong, P. A. Janmey, Biochim. Biophys. Acta, Mol. Cell Res. 2015,1853,
3153.
[10] A. Malandrino, M. Mak, R. D. Kamm, E. Moeendarbary, Extreme
Mech. Lett. 2018,21, 25.
[11] Q. Feng, D. Li, Q. Li, X. Cao, H. Dong, Bioact. Mater. 2022,9,
105.
[12] K. Flégeau, A. Puiggali-Jou, M. Zenobi-Wong, Biofabrication 2022,14,
034105.
[13] A. C. Daly, L. Riley, T. Segura, J. A. Burdick, Nat. Rev. Mater. 2019,5,
20.
[14] S. Xin, D. Chimene, J. E. Garza, A. K. Gaharwar, D. L. Alge, Biomater.
Sci. 2019,7, 1179.
[15] C. B. Highley, K. Hoon Song, A. C. Daly, J. A. Burdick, Adv. Sci. 2019,
6, 1801076.
[16] V. G. Muir, T. H. Qazi, J. Shan, J. Groll, J. A. Burdick, ACS Biomater.
Sci. Eng. 2021,7, 4269.
[17] A. Harada, R. Kobayashi, Y. Takashima, A. Hashidzume, H. Yam-
aguchi, Nat. Chem. 2010,3, 34.
[18] Y. L. Han , Y. Ya n g , S . L i u , J . W u , Y. C h e n , T. J . L u, F. X u , Biofabrication
2013,5, 035004.
[19] D. Rommel, M. Mork, S. Vedaraman, C. Bastard, L. P. B. Guerzoni, Y.
Kittel, R. Vinokur, N. Born, T. Haraszti, L. De Laporte, Adv. Sci. 2022,
9, 2103554.
[20] N. F. Truong, E. Kurt, N. Tahmizyan, S. C. Lesher-Pérez, M. Chen, N.
J. Darling, W. Xi, T. Segura, Acta Biomater. 2019,94, 160.
[21] M. Shin, K. H. Song, J. C. Burrell, D. K. Cullen, J. A. Burdick, Adv. Sci.
2019,6, 1901229.
[22] B. Kessel, M. Lee, A. Bonato, Y. Tinguely, E. Tosoratti, M. Zenobi-
Wong, Adv. Sci. 2020,7, 2001419.
[23] G.-L. Ying, N. Jiang, S. Maharjan, Y.-X. Yin, R.-R. Chai, X. Cao, J.-
Z. Yang, A. K. Miri, S. Hassan, Y. S. Zhang, Adv. Mater. 2018,30,
1805460.
[24] Y. Chao, H. C. Shum, Chem.Soc.Rev.2020,49, 114.
[25] S.Yi,Q.Liu,Z.Luo,J.J.He,H.-L.Ma,W.Li,D.Wang,C.Zhou,
C. E. Garciamendez, L. Hou, J. Zhang, Y. S. Zhang, Small 2022,18,
2106357.
[26] X. S. Qin, M. Wang, W. Li, Y. S. Zhang, Regener. Eng. Transl. Med. 2022,
8, 471.
[27] J. Tao, S. Zhu, N. Zhou, Y. Wang, H. Wan, L. Zhang, Y. Tang, Y. Pan, Y.
Yang, J. Zhang, R. Liu, Adv. Healthcare Mater. 2022,11, 2102810.
[28] M. Wang, W. Li, Z. Luo, G. Tang, X. Mu, X. Kuang, J. Guo, Z. Zhao,
R. S. Flores, Z. Jiang, L. Lian, J. O. Japo, A. M. Ghaemmaghami, Y. S.
Zhang, Biofabrication 2022,14, 024105.
[29] J. Zhang, Q. Hu, S. Wang, J. Tao, M. Gou, Int. J. Bioprint. 2020,6, 12.
[30] A. Madadlou, A. Saint-Jalmes, F. Guyomarc’h, J. Floury, D. Dupont,
Food Hydrocolloids 2019,93, 351.
[31] L. Tea, T. Nicolai, F. Renou, Langmuir 2019,35, 9029.
[32] L. Ouyang, J. P. Wojciechowski, J. Tang, Y. Guo, M. M. Stevens, Adv.
Healthcare Mater. 2022,11, 2200027.
[33] H.Li,Y.J.Tan,R.Kiran,S.B.Tor,K.Zhou,Addit. Manuf. 2021,37,
101640.
[34] F. Zhou, Y. Hong, R. Liang, X. Zhang, Y. Liao, D. Jiang, J. Zhang, Z.
Sheng, C. Xie, Z. Peng, X. Zhuang, V. Bunpetch, Y. Zou, W. Huang,
Q. Zhang, E. V. Alakpa, S. Zhang, H. Ouyang, Biomaterials 2020,258,
120287.
[35] E. A. Guzzi, R. Bischof, D. Dranseikiene, D. V. Deshmukh, A.
Wahlsten, G. Bovone, S. Bernhard, M. W. Tibbitt, Biofabrication 2021,
13, 044105.
[36] M. Y. Shie, J. J. Lee, C. C. Ho, S. Y. Yen, H. Y. Ng, Y. W. Chen, Polymers
2020,12, 1930.
[37] S. Bertlein, G. Brown, K. S. Lim, T. Jungst, T. Boeck, T. Blunk, J. Tess-
mar, G. J. Hooper, T. B. F. Woodfield, J. Groll, Adv. Mater. 2017,29,
1703404.
[38] K.Yu,X.Zhang,Y.Sun,Q.Gao,J.Fu,X.Cai,Y.He,Bioact. Mater.
2022,11, 254.
[39] Y. Ma, M. Lin, G. Huang, Y. Li, S. Wang, G. Bai, J. Lu, F. Xu, Adv. Mater.
2018,30, 1705911.
[40] M. Z. Müller, R. W. Style, R. Müller, X.-H. Qin, bioRxiv, https://doi.
org/10.1101/2022.01.29.478338, submitted: January, 2022.
[41] R. Levato, K. S. Lim, W. Li, A. U. Asua, L. B. Peña, M. Wang, M. Falandt,
P. N. Bernal, D. Gawlitta, Y. S. Zhang, T. B. F. Woodfield, J. Malda,
Mater. Today Bio 2021,12, 100162.
[42] G. Ying, N. Jiang, C. Parra-Cantu, G. Tang, J. Zhang, H. Wang, S. Chen,
N.-P. Huang, J. Xie, Y. S. Zhang, Adv. Funct. Mater. 2020,30, 2003740.
[43] S. Xin, C. A. Gregory, D. L. Alge, Acta Biomater. 2020,101, 227.
[44] A. Isaac, F. Jivan, S. Xin, J. Hardin, X. Luan, M. Pandya, T. G. H. Diek-
wisch, D. L. Alge, ACS Biomater. Sci. Eng. 2019,5, 6395.
[45] D. R. Griffin, W. M. Weaver, P. O. Scumpia, D. Di Carlo, T. Segura,
Nat. Mater. 2015,14, 737.
[46] G. Ying, J. Manríquez, D. Wu, J. Zhang, N. Jiang, S. Maharjan, D. H.
Hernández Medina, Y. S. Zhang, Mater. Today Bio 2020,8, 100074.
[47] M. Wang, W. Li, L. S. Mille, T. Ching, Z. Luo, G. Tang, C. E. Garcia-
mendez, A. Lesha, M. Hashimoto, Y. S. Zhang, Adv. Mater. 2022,34,
2107038.
[48] C. D. O’Connell, C. D. Bella, F. Thompson, C. Augustine, S. Beirne,
R. Cornock, C. J. Richards, J. Chung, S. Gambhir, Z. Yue, J. Bourke,
B. Zhang, A. Taylor, A. Quigley, R. Kapsa, P. Choong, G. G. Wallace,
Biofabrication 2016,8, 015019.
[49] X. Chen, Z. Yue, P. C. Winberg, Y. R. Lou, S. Beirne, G. G. Wallace,
Biomater. Sci. 2021,9, 2424.
[50] Q. Wang, W. Xu, R. Koppolu, B. van Bochove, J. Seppälä, L. Hupa, S.
Willför, C. Xu, X. Wang, Carbohydr. Polym. 2022,276, 118780.
[51] A. Sundberg, K. Sundberg, C. Lillandt, B. Holmbom, Nord. Pulp Pap.
Res. J. 1996,11, 216.
[52] N. Broguiere, A. Husch, G. Palazzolo, F. Bradke, S. Madduri, M.
Zenobi-Wong, Biomaterials 2019,200, 56.
Adv. Healthcare Mater. 2023,12, 2203243 2203243 (15 of 15) © 2023 The Authors. Advanced Healthcare Materials published by Wiley-VCH GmbH
