Access to this full-text is provided by MDPI.
Content available from Gels
This content is subject to copyright.
Citation: Kyriakou, S.; Lubig, A.;
Sandhoff, C.A.; Kuhn, Y.;
Jockenhoevel, S. Influence of
Diameter and Cyclic Mechanical
Stimulation on the Beating Frequency
of Myocardial Cell-Laden Fibers. Gels
2023,9, 677. https://doi.org/
10.3390/gels9090677
Academic Editors: Guilhem Godeau
and Miguel A. Cerqueira
Received: 22 July 2023
Revised: 12 August 2023
Accepted: 21 August 2023
Published: 23 August 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
gels
Article
Influence of Diameter and Cyclic Mechanical Stimulation on
the Beating Frequency of Myocardial Cell-Laden Fibers
Stavroula Kyriakou 1, Andreas Lubig 1, Cilia A. Sandhoff 1, Yasmin Kuhn 1and Stefan Jockenhoevel 1, 2, *
1Department of Biohybrid & Medical Textiles (BioTex), AME-Institute of Applied Medical Engineering,
Helmholtz Institute, RWTH Aachen University, 52074 Aachen, Germany;
kyriakou@ame.rwth-aachen.de (S.K.); sandhoff@ame.rwth-aachen.de (C.A.S.);
kuhn@ame.rwth-aachen.de (Y.K.)
2AMIBM-Aachen-Maastricht-Institute for Biobased Materials, Maastricht University,
186260 Geleen, The Netherlands
*Correspondence: jockenhoevel@ame.rwth-aachen.de
Abstract:
Atrioventricular block (AVB) is a severe disease for pediatric patients. The repetitive
operations needed in the case of the pacemaker implantation to maintain the electrical signal at the
atrioventricular node (AVN) affect the patient’s life quality. In this study, we present a method of
biofabrication of multi-cell-laden cylindrical fibrin-based fibers that can restore the electrical signal
at the AVN. We used human umbilical vein smooth muscle cells (HUVSMCs), human umbilical
vein endothelial cells (HUVECs) and induced pluripotent stem cell cardiomyocytes (iPSC-CMs)
cultivated either statically or dynamically to mimic the native AVN. We investigated the influence
of cell composition, construct diameter and cyclic stretch on the function of the fibrin hydrogels
in vitro
. Immunohistochemistry analyses showed the maturity of the iPSC-CMs in the constructs
through the expression of sarcomeric alpha actinin (SAA) and electrical coupling through Connexin
43 (Cx43) signal. Simultaneously, the beating frequency of the fibrin hydrogels was higher and easy
to maintain whereas the concentration of iPSC-CMs was higher compared with the other types of
cylindrical constructs. In total, our study highlights that the combination of fibrin with the cell
mixture and geometry is offering a feasible biofabrication method for tissue engineering approaches
for the treatment of AVB.
Keywords: fibrin gel; iPSC-CMs; biofabrication; mechanical stimulation; atrioventricular block
1. Introduction
The atrioventricular block is a crucial pathological situation that specifically constitutes
a significant burden in pediatric heart surgery. The atrioventricular block of any origin-
immune-mediated, inherited, or apparently idiopathic- [
1
] is a rare [
2
] pathological situation
that belongs to the huge category of bradycardias [
3
]. Cardioneuroablation [
4
], pacemaker
implantation, and antibiotics have been used for the treatment of the atrioventricular
block [
5
]. Nevertheless, pacemaker implantations necessitate repetitive operations and
influence the life quality of the children [
6
,
7
]. Particularly, pacemaker implantation has
been associated with left ventricular (LV) dysfunction [
8
], dilated cardiomyopathy [
9
,
10
]
relatively high mortality rate in pediatric patients after the surgery [
11
], risk of syncope,
heart failure and sudden death [12].
Many studies have reported solutions that could be potentially useful for the cell-
mediated therapy of atrioventricular block. Gorabi et al. inserted a Tbx18 gene alone and
also Tbx18-inserted pacemaker-like stem cells in a complete heart block rat model achieving
a pacemaker capacity by comparing the outcomes of the gene and cell insertion [
13
].
Lu et al. injected mHCN4-modified cMSCs with support of an electronic pacemaker [
14
].
In another study, Yokokawa et al. inserted mesenchymal stem cells with antifibrotic action
in a rat model of complete atrioventricular block [
15
]. Moreover, Chaveau et al. used
Gels 2023,9, 677. https://doi.org/10.3390/gels9090677 https://www.mdpi.com/journal/gels
Gels 2023,9, 677 2 of 17
iPSC-CMs as therapy for the atrioventricular block, but the beating frequency and other
parameters related to pacing remain still unraveled [
16
]. To improve particular aims on
cell orientation and achieve adequate perfusion parameters for cardiovascular applications
many studies have developed bioreactors. Cingolani et al. created a strip of cardiomyocytes
by attaching the cells to magnetic beads that were then aligned over a magnetic field [
17
].
Using a rat model of a complete heart block, they implanted this strip outside of the heart,
restoring atrioventricular conduction. Miklas et al. [
18
] prepared collagen gels embedded
with neonatal rat heart-derived cardiomyocytes performing electrical and mechanical
stimulation. Additionally, Nunes et al. [
19
] created self-assembled electrically stimulated
cardiac biowires using iPSC-CMs. Moreover, Kensah et al. [
20
] applied cyclic and static
stretch in murine and human-induced cardiomyocytes. In terms of the geometry, Keijdener
et al. [
21
] established a method for the preparation of cylindrical fibrin gels using neural
Schwann cells, endothelial and smooth muscle cells showing orientation after applied
stretch of different degrees.
Nowadays the treatment of atrioventricular block necessitates a biological solution to
avoid the repetitive operations, which influence the quality of life of the children. Bioma-
terial science and cell delivery systems such as fibrin gel can be used as a state-of-the-art
method for this aim. Fibrin is a plasma membrane protein that has been widely used
because of its high biocompatibility [
22
,
23
]. In particular, fibrin gel has been widely used
for
in vitro
tissue formation [
24
,
25
] for different applications which include wound repara-
tion [
26
], drug delivery systems [
27
], cell delivery [
28
,
29
], gene delivery [
30
], differentiation
in tissue engineering [
31
] and patterning [
32
]. However, fibrin composites lack in terms of
mechanical properties compared to other polymers offering equal biocompatibility and
biodegradability [33,34].
In this study, we are describing the method optimization of embedding induced
pluripotent stem cell cardiomyocytes (iPSC-CMs) together with smooth muscle cells and
endothelial cells for the preparation of an autonomous cylindrical construct. This study
aims to assess the influence of the cell composition, the diameter and the cyclic mechanical
stimulation on the beating frequency of the BioPacer constructs and also on the expression
of important cell markers indicating the construct with the best performance. For the
preparation of the BioPacer constructs we used a micro-molding technology described
previously by Keijdener et al. The cell density of each construct is described in Table 1.
Table 1. BioPacer samples.
Sample Type Fiber Number [iPSC-CMs]
(Cells/Fiber)
[HUVSMCs]
(Cells/Fiber)
[HUVECs]
(Cells/Fiber)
Low cell density 500 µm (G1) 2 150,000 60,000 20,000
Mix 500 µm (G2) 1 300,000 60,000 20,000
500 µm (G3) 2 300,000 60,000 20,000
500 µm stretch (G4) 2 300,000 60,000 20,000
2×500 µm (G5) 2 300,000
3×500 µm (G6) 3 300,000
1 mm (G7) 2 500,000 120,000 60,000
1 mm stretch (G8) 2 500,000 120,000 60,000
Mix 1 mm (G9) 1 500,000 120,000 60,000
Different diameters were used for the molding of the hydrogels in the silicon tubes
and also different cell compositions of the BioPacer were prepared for the assessment of the
diameter and the stretch to the beating frequency of the constructs. Additionally, 2 types of
constructs were subject to cyclic mechanical stretch and were also tested for the expression
of iPSC-CM, HUVSMC and HUVEC markers.
Gels 2023,9, 677 3 of 17
2. Results and Discussion
2.1. Cell Infiltration and Proliferation
For the investigation of the cell infiltration in the fibrin gel, three different BioPacer
mix types of samples were prepared. The 1 mm (G7) and the Mix 500
µ
m (G2) samples
had the same initial diameter on day 2 of cultivation. Even if the 3
×
500
µ
m (G6) sample
had a bigger initial diameter, on day 6 of cultivation a contraction of the fibrin matrix
was observed and that is why the diameter was decreased (Figure 1). This trend appeals
to all the 3 types of samples. On the 10th day of cultivation, the cells and particularly
the smooth muscle cells were expanded out of the cylindrical structure and they started
attaching and growing on the surface of the well plate. On day 14 of cultivation and mostly
for the G7 samples and for the G6 which contain the initial higher number of HUVSMCs,
the proliferation was also higher and the HUVSMCs created multiple layers around the
fibrin gels (right column of Figure 1). As a consequence, the percentage of cell integration
was effective for the different diameters of the BioPacer sample and the samples could be
functional in terms of integration in the fibrin gel for various configurations.
Gels 2023, 9, x FOR PEER REVIEW 3 of 17
2. Results and Discussion
2.1. Cell Infiltration and Proliferation
For the investigation of the cell infiltration in the fibrin gel, three different BioPacer
mix types of samples were prepared. The 1 mm (G7) and the Mix 500 µm (G2) samples
had the same initial diameter on day 2 of cultivation. Even if the 3 × 500 µm (G6) sample
had a bigger initial diameter, on day 6 of cultivation a contraction of the fibrin matrix was
observed and that is why the diameter was decreased (Figure 1). This trend appeals to all
the 3 types of samples. On the 10th day of cultivation, the cells and particularly the smooth
muscle cells were expanded out of the cylindrical structure and they started aaching and
growing on the surface of the well plate. On day 14 of cultivation and mostly for the G7
samples and for the G6 which contain the initial higher number of HUVSMCs, the prolif-
eration was also higher and the HUVSMCs created multiple layers around the fibrin gels
(right column of Figure 1). As a consequence, the percentage of cell integration was effec-
tive for the different diameters of the BioPacer sample and the samples could be functional
in terms of integration in the fibrin gel for various configurations.
Figure 1. Bright-field images of three different mix configurations of the BioPacer for days 2, 6, 10
and 14 of the cultivation period. First row: Mix 500 µm sample. Second row: 1 mm sample. Third
row: 3 × 500 µm sample.
2.2. iPSC-CM Markers for the BioPacer
One of the first steps for the effect of stretching of the iPSC-CMs was the investigation
of the effect of stretching for different cultivation periods on the samples. In Figure 2 the
BioPacer sample 500 µm and the 1 mm BioPacer are represented without and with cyclic
stress cultivated for 7 days and with cyclic stress for 14 days. Cross-sectional cuts are pre-
sented in Supplementary Figure S1.
Figure 1.
Bright-field images of three different mix configurations of the BioPacer for days 2, 6,
10 and 14 of the cultivation period. First row: Mix 500
µ
m sample. Second row: 1 mm sample.
Third row: 3 ×500 µm sample.
2.2. iPSC-CM Markers for the BioPacer
One of the first steps for the effect of stretching of the iPSC-CMs was the investigation
of the effect of stretching for different cultivation periods on the samples. In Figure 2the
BioPacer sample 500
µ
m and the 1 mm BioPacer are represented without and with cyclic
stress cultivated for 7 days and with cyclic stress for 14 days. Cross-sectional cuts are
presented in Supplementary Figure S1.
Gels 2023,9, 677 4 of 17
Gels 2023, 9, x FOR PEER REVIEW 4 of 17
Figure 2. DAPI (blue), SAA (green) and Cx43 (red) for the BioPacer samples without and with cyclic
stress cultivated for 7 days and with cyclic stress for 14 days.
The 500 µm sample showed that the expression of SAA from the iPSC-CMs was rel-
atively high comparing the 500 µm BioPacer sample with applied stretch. However, the
expression of Cx43 in the 1 mm stretch sample was higher in the 1 mm sample without
applied stretch, indicating that the stretching of the samples positively affects the electrical
coupling among the iPSC-CMs. Another important factor was the longer cultivation of the
samples. Moreover, the influence of stretching was much more visible in the 1 mm Bio-
Pacer as the nuclei were oriented to the direction of the stretch. Moreover, the SAA was
not equally expressed by all the cells embedded in the fibrin gel. Comparing the 500 µm
sample, the SAA production was distributed in a more regular paern, as expected from
the absence of stretch. At the same time, the Cx43 signal was higher for the samples culti-
vated for 14 days, as the iPSC-CMs needed some days to regulate their metabolism and
their normal function.
As a further step, we investigated the expression of the SAA, Cx43 and col I according
to the concentration change of the iPSC-CMs in the constructs, the composition and the
applied cyclic stretch. In Figure 3 (first row), the respective configurations are shown for
the SAA expression of the Low cell density 500 µm (G1), the Mix 500 µm (G2) and the 500
µm stretch sample (G4). Cross-sectional cuts are presented in Supplementary Figure S2.
Figure 2.
DAPI (blue), SAA (green) and Cx43 (red) for the BioPacer samples without and with cyclic
stress cultivated for 7 days and with cyclic stress for 14 days.
The 500
µ
m sample showed that the expression of SAA from the iPSC-CMs was
relatively high comparing the 500
µ
m BioPacer sample with applied stretch. However, the
expression of Cx43 in the 1 mm stretch sample was higher in the 1 mm sample without
applied stretch, indicating that the stretching of the samples positively affects the electrical
coupling among the iPSC-CMs. Another important factor was the longer cultivation of
the samples. Moreover, the influence of stretching was much more visible in the 1 mm
BioPacer as the nuclei were oriented to the direction of the stretch. Moreover, the SAA
was not equally expressed by all the cells embedded in the fibrin gel. Comparing the
500
µ
m sample, the SAA production was distributed in a more regular pattern, as expected
from the absence of stretch. At the same time, the Cx43 signal was higher for the samples
cultivated for 14 days, as the iPSC-CMs needed some days to regulate their metabolism
and their normal function.
As a further step, we investigated the expression of the SAA, Cx43 and col I according
to the concentration change of the iPSC-CMs in the constructs, the composition and the
applied cyclic stretch. In Figure 3(first row), the respective configurations are shown for
the SAA expression of the Low cell density 500
µ
m (G1), the Mix 500
µ
m (G2) and the
500
µ
m stretch sample (G4). Cross-sectional cuts are presented in
Supplementary Figure S2.
The SAA expression of the G1 sample was concentrated on one side of the fiber, where
the iPSC-CMs, which are responsible for the SAA-specific signal, are also located. In
these fibers, even if the concentration of the iPSC-CMs is half of the concentration in the
G2 sample, the SAA signal was still significant. In Figure 3, there was an overall distribution
of the SAA signal reflecting the difference in the molding process by mixing all the cell
types in a common matrix. In the samples where the number of the iPSC-CMs was low, the
collagen expression from the smooth muscle cells was more profound as there was more
space for the disposal of fibroblasts in the diameter of the cylindrical structure, unlike the
conventional BioPacer samples, where the collagen amount was a bit mitigated from the
presence of a higher number of iPSC-CMs which were present in the BioPacer samples.
At the same time, the Cx43 signal was stronger in the G1 than in the G2 sample, a fact
that is reflecting the effect of fibroblast proliferation in the G2 sample. Furthermore, as
for the effect of the applied cyclic stretch, the expression of collagen was the strongest
among all the other samples, as it constitutes the direct consequence of the applied stretch.
Simultaneously, the Cx43 signal was still present, but not as in a high amount as in the
G1 sample.
Gels 2023,9, 677 5 of 17
Gels 2023, 9, x FOR PEER REVIEW 5 of 17
Figure 3. DAPI (blue), SAA (green), Cx43 (red) and collagen I (green) staining. Left: Low cell density
500 µm (G1), middle: Mix 500 µm (G2), right: 500 µm stretch (G4).
The SAA expression of the G1 sample was concentrated on one side of the fiber,
where the iPSC-CMs, which are responsible for the SAA-specific signal, are also located.
In these fibers, even if the concentration of the iPSC-CMs is half of the concentration in the
G2 sample, the SAA signal was still significant. In Figure 3, there was an overall distribu-
tion of the SAA signal reflecting the difference in the molding process by mixing all the
cell types in a common matrix. In the samples where the number of the iPSC-CMs was
low, the collagen expression from the smooth muscle cells was more profound as there
was more space for the disposal of fibroblasts in the diameter of the cylindrical structure,
unlike the conventional BioPacer samples, where the collagen amount was a bit mitigated
from the presence of a higher number of iPSC-CMs which were present in the BioPacer
samples. At the same time, the Cx43 signal was stronger in the G1 than in the G2 sample,
a fact that is reflecting the effect of fibroblast proliferation in the G2 sample. Furthermore,
as for the effect of the applied cyclic stretch, the expression of collagen was the strongest
among all the other samples, as it constitutes the direct consequence of the applied stretch.
Simultaneously, the Cx43 signal was still present, but not as in a high amount as in the G1
sample.
Additionally, the SAA and Cx43 markers for the iPSC-CMs appeared to be in a good
correlation with the CD31 marker for the endothelial cells that were present in the cylin-
drical structures of the BioPacer. In Figure 4, the SAA signal produced by the iPSC-CMs
was stronger than the Cx43 signal for the 3 types of samples analyzed. Cross-sectional cuts
are presented in Supplementary Figure S3. At the same time, the Cx43 seemed to be
stronger in the G4 sample, as the iPSC-CMs gathered in the middle of the cylindrical core.
Furthermore, the SAA expression in the G4 sample appeared to be stronger than in the
G1 and G2 samples. Moreover, the SAA spots were well oriented in the G4 sample, unlike
the G1 and the G2 samples. Comparing the G1 and the G2 samples, the CD31 cobblestone
structure was clear in the G2 sample, unlike the G1 sample where the CD31 expression
was too low. Additionally, the CD31 expression paern followed the direction of the
stretch in the G4 sample.
Figure 3.
DAPI (blue), SAA (green), Cx43 (red) and collagen I (green) staining. Left: Low cell density
500 µm (G1), middle: Mix 500 µm (G2), right: 500 µm stretch (G4).
Additionally, the SAA and Cx43 markers for the iPSC-CMs appeared to be in a
good correlation with the CD31 marker for the endothelial cells that were present in the
cylindrical structures of the BioPacer. In Figure 4, the SAA signal produced by the iPSC-
CMs was stronger than the Cx43 signal for the 3 types of samples analyzed. Cross-sectional
cuts are presented in Supplementary Figure S3. At the same time, the Cx43 seemed to be
stronger in the G4 sample, as the iPSC-CMs gathered in the middle of the cylindrical core.
Furthermore, the SAA expression in the G4 sample appeared to be stronger than in the
G1 and G2 samples. Moreover, the SAA spots were well oriented in the G4 sample, unlike
the G1 and the G2 samples. Comparing the G1 and the G2 samples, the CD31 cobblestone
structure was clear in the G2 sample, unlike the G1 sample where the CD31 expression was
too low. Additionally, the CD31 expression pattern followed the direction of the stretch in
the G4 sample.
In order to better assess the qualitative results from the two photon images of the sam-
ples, volume quantification was performed using the most important markers expressed
in the cells used in the present study (HUVSMCs, HUVECs and iPSC-CMs) such as SAA,
Cx43, Col, CD31 and DAPI.
Particularly, the applied stretch consists of an influencing factor for the percentage of
DAPI, as there is a significant difference between the 1 mm (G7) and 1 mm stretch (G8)
samples. Moreover, the Cx43 signal did not show significant differences. On the other
hand, the expression of SAA had the highest value for the G8 sample and was significantly
different from the respective value of the G7 sample. This could be explained by the fact
that the iPSC-CMs the G7 could not produce the same amount of SAA as the G8 sample,
because of the inhibitory proliferation of the HUVSCMs and the expansion of the latest
ones outside the fibrin matrix. The highest proliferation rate of the HUVSMCs and the
lack of proliferation of the iPSC-CMs prohibits the production of SAA, an indicator of
their maturity. However, the G8 sample represented a higher expression of SAA, as the
mechanical stimulation of the samples could trigger them and catalyze the production of
SAA from the side of the iPSC-CMs.
Gels 2023,9, 677 6 of 17
Gels 2023, 9, x FOR PEER REVIEW 6 of 17
Figure 4. DAPI (blue), SAA (green), Cx43 (red) and CD31 (red) staining. Left: Low cell density 500
µm (G1), middle: Mix 500 µm (G2), right: 500 µm stretch (G4).
In order to beer assess the qualitative results from the two photon images of the
samples, volume quantification was performed using the most important markers ex-
pressed in the cells used in the present study (HUVSMCs, HUVECs and iPSC-CMs) such
as SAA, Cx43, Col, CD31 and DAPI.
Particularly, the applied stretch consists of an influencing factor for the percentage of
DAPI, as there is a significant difference between the 1 mm (G7) and 1 mm stretch (G8)
samples. Moreover, the Cx43 signal did not show significant differences. On the other
hand, the expression of SAA had the highest value for the G8 sample and was significantly
different from the respective value of the G7 sample. This could be explained by the fact
that the iPSC-CMs the G7 could not produce the same amount of SAA as the G8 sample,
because of the inhibitory proliferation of the HUVSCMs and the expansion of the latest
ones outside the fibrin matrix. The highest proliferation rate of the HUVSMCs and the
lack of proliferation of the iPSC-CMs prohibits the production of SAA, an indicator of
their maturity. However, the G8 sample represented a higher expression of SAA, as the
mechanical stimulation of the samples could trigger them and catalyze the production of
SAA from the side of the iPSC-CMs.
Furthermore, col I and CD31 were also quantified according to their volume in the
samples to evaluate the function of HUVSMCs and HUVECs in the G1, G2 and G4 sam-
ples (Figure 5). As observed in Figure 5, the expression of the col I was the highest for the
G4 sample, as the applied stretch promoted the higher expression of collagen from the
HUVSMCs. This difference in the value was significantly different from the expression of
the col I in the G2 sample, even if it contained the same amount of HUVSMCs as the G4
sample. This gives more importance to the fact that the applied stretch affected the expres-
sion of the col I. The lowest collagen expression was observed for the G2 sample. The
reason is that the concentration of HUVSMCs in the G2 sample was lower than the G1
sample and the subsequent proliferation and production of col I from the HUVSMCs was
also lower. At the same time, in the mixture of the cells in the cylindrical fibrin matrix, the
geometric positioning of the HUVSMCs also affected the production of collagen as the
distance between the single cells is longer and this does not facilitate the synergistic action
for the production of extracellular matrix.
Figure 4.
DAPI (blue), SAA (green), Cx43 (red) and CD31 (red) staining. Left: Low cell density
500 µm (G1), middle: Mix 500 µm (G2), right: 500 µm stretch (G4).
Furthermore, col I and CD31 were also quantified according to their volume in the
samples to evaluate the function of HUVSMCs and HUVECs in the G1, G2 and G4 samples
(Figure 5). As observed in Figure 5, the expression of the col I was the highest for the
G4 sample, as the applied stretch promoted the higher expression of collagen from the
HUVSMCs. This difference in the value was significantly different from the expression
of the col I in the G2 sample, even if it contained the same amount of HUVSMCs as the
G4 sample. This gives more importance to the fact that the applied stretch affected the
expression of the col I. The lowest collagen expression was observed for the G2 sample.
The reason is that the concentration of HUVSMCs in the G2 sample was lower than the
G1 sample and the subsequent proliferation and production of col I from the HUVSMCs
was also lower. At the same time, in the mixture of the cells in the cylindrical fibrin matrix,
the geometric positioning of the HUVSMCs also affected the production of collagen as the
distance between the single cells is longer and this does not facilitate the synergistic action
for the production of extracellular matrix.
Furthermore, the profile of the expression of SAA had the same tendency as the
expression of col I. The SAA had the greatest value in the G4 sample, because of the
mechanical stimulation of the BioPacer. The result was significantly different from the
G1 sample, as the amount of cells is significantly different than the G4 sample. Similar
was the behavior of the G2 sample, with a highly significant difference compared with the
G4 sample. This is explained by the fact that the iPSC-CMs were located in the cylindrical
structure more isolated than in the G1 sample. Consequently, they act more individually
than they may do in the G1 and in the G4 sample, where the mechanical loading contributes
to the matrix remodeling.
Gels 2023,9, 677 7 of 17
Gels 2023, 9, x FOR PEER REVIEW 7 of 17
Figure 5. Quantification of DAPI, SAA, CD31 and Col I for the Low cell density 500 µm (G1), Mix
500 µm (G2), 500 µm stretch sample (G4) (Data are expressed as mean ± SD, n = 3, * p < 0.05,*** p <
0.001, **** p < 0.0001).
Furthermore, the profile of the expression of SAA had the same tendency as the ex-
pression of col I. The SAA had the greatest value in the G4 sample, because of the mechan-
ical stimulation of the BioPacer. The result was significantly different from the G1 sample,
as the amount of cells is significantly different than the G4 sample. Similar was the behav-
ior of the G2 sample, with a highly significant difference compared with the G4 sample.
This is explained by the fact that the iPSC-CMs were located in the cylindrical structure
more isolated than in the G1 sample. Consequently, they act more individually than they
may do in the G1 and in the G4 sample, where the mechanical loading contributes to the
matrix remodeling.
2.3. Beating Frequency
During the cultivation period, we evaluated the beating frequency of the constructs.
As observed during the cultivation period, the constructs started the synchronized beating
after the second day of cultivation, reaching their maximum level after the first week of
cultivation. The highest beating frequency was shown for the 2 × 500 µm sample (G5) and
was observed on the 8th day of cultivation, while the lowest beating frequency value was
shown for the G1 sample, where the concentration of the iPSC-CMs is half of the amount
in the rest of 500 µm samples (Figure 6). The samples in which the beating frequency had
a slightly increasing tendency were the G6 and the G2 sample. The beating frequency for
the G5 sample had also the same increasing tendency, but the value was slightly decreased
on the 12th day. On the contrary, for the G1, the G3 and the G7 samples the beating fre-
quency had been decreasing during the cultivation period (Figure 6).
Looking more deeply at the beating frequency capacities of the different samples,
until the 4th day, there were no significant changes among the samples, even if the sam-
ples started the homogeneous and synchronized beating until the 4th cultivation day. On
the contrary, on the 8th day, there was a significant difference between the G1 sample and
the G5 sample, as the laer contained a higher concentration of iPSC-CMs promoting the
increase in the beating frequency over the cultivation period. At the same time, the G7
Figure 5.
Quantification of DAPI, SAA, CD31 and Col I for the Low cell density 500
µ
m (G1), Mix
500
µ
m (G2), 500
µ
m stretch sample (G4) (Data are expressed as mean
±
SD, n= 3, * p< 0.05,
*** p< 0.001, **** p< 0.0001).
2.3. Beating Frequency
During the cultivation period, we evaluated the beating frequency of the constructs.
As observed during the cultivation period, the constructs started the synchronized beating
after the second day of cultivation, reaching their maximum level after the first week of
cultivation. The highest beating frequency was shown for the 2
×
500
µ
m sample (G5) and
was observed on the 8th day of cultivation, while the lowest beating frequency value was
shown for the G1 sample, where the concentration of the iPSC-CMs is half of the amount in
the rest of 500
µ
m samples (Figure 6). The samples in which the beating frequency had a
slightly increasing tendency were the G6 and the G2 sample. The beating frequency for the
G5 sample had also the same increasing tendency, but the value was slightly decreased on
the 12th day. On the contrary, for the G1, the G3 and the G7 samples the beating frequency
had been decreasing during the cultivation period (Figure 6).
Looking more deeply at the beating frequency capacities of the different samples, until
the 4th day, there were no significant changes among the samples, even if the samples
started the homogeneous and synchronized beating until the 4th cultivation day. On the
contrary, on the 8th day, there was a significant difference between the G1 sample and
the G5 sample, as the latter contained a higher concentration of iPSC-CMs promoting
the increase in the beating frequency over the cultivation period. At the same time, the
G7 sample presented also higher significant difference than the G5 sample, as the higher
proliferation of HUVSMCs prohibited the action of the high number of iPSC-CMs, by
isolating their signal at a higher extent than needed. However, most of the changes
appeared on the 12th day of the cultivation period. For instance, the G1 sample had a
significantly different beating frequency than the G5 and G6 samples. The reason is that the
higher concentration of iPSC-CMs stabilized the threshold of the beating frequency values
of the G5 and G6 samples (Supplementary Video S1). Moreover, the beating frequency of
the Mix 1 mm sample (G9) was significantly higher than the respective G7 value, indicating
that the mixing of the samples acts as a promoting factor for the maintenance of the beating
frequency. Additionally, there was a highly significant difference between the G1 and the
G9 sample. The main reason appeared to be the increased iPSC-CM concentration in the
samples in combination with the mixing during the molding. Simultaneously, the same
is the behavior of the G3 sample, as in this case only one of the fibers that constitute the
Gels 2023,9, 677 8 of 17
BioPacer was beating, while the other one acted as a mechanical support and signal isolator
of the electrical signal.
Gels 2023, 9, x FOR PEER REVIEW 8 of 17
sample presented also higher significant difference than the G5 sample, as the higher pro-
liferation of HUVSMCs prohibited the action of the high number of iPSC-CMs, by isolat-
ing their signal at a higher extent than needed. However, most of the changes appeared
on the 12th day of the cultivation period. For instance, the G1 sample had a significantly
different beating frequency than the G5 and G6 samples. The reason is that the higher
concentration of iPSC-CMs stabilized the threshold of the beating frequency values of the
G5 and G6 samples (Supplementary video S1). Moreover, the beating frequency of the
Mix 1 mm sample (G9) was significantly higher than the respective G7 value, indicating
that the mixing of the samples acts as a promoting factor for the maintenance of the beat-
ing frequency. Additionally, there was a highly significant difference between the G1 and
the G9 sample. The main reason appeared to be the increased iPSC-CM concentration in
the samples in combination with the mixing during the molding. Simultaneously, the
same is the behavior of the G3 sample, as in this case only one of the fibers that constitute
the BioPacer was beating, while the other one acted as a mechanical support and signal
isolator of the electrical signal.
Figure 6. Beating frequencies for the BioPacer samples (Low cell density 500 µm (G1), Mix 500 µm
(G2), 500 µm (G3), 2 × 500 µm (G5), 3 × 500 µm (G6), Mix 1 mm (G9) and 1 mm (G7) on the Day 4, 8
and 12 of cultivation (Data are expressed as mean ± SD, n = 3, * p < 0.05, ** p < 0.01.
2.4. Glucose and Lactate Measurements
The glucose and lactate levels were measured for the different configurations of the
BioPacer for the constructs cultivated for 14 days. In Figure 7, the consumption of glucose
is ploed for the six different configurations of the BioPacer. The highest level of glucose
consumption was achieved for the G7 sample presenting significant differences among all
the other sample types. For the samples with different diameters, there are more signifi-
cant differences observed between the samples (G1 and G7, G2 and G7, G3 and G7, G4
and G7) than the differences when comparing samples with the same diameter (G3 and
G4, G7 and G8). As a result, the diameter plays an important role in the glucose consump-
tion of the samples. Furthermore, the lactate production value of the G4 sample was re-
ported higher than the G3 sample, reflecting the positive effect of the applied stress in the
metabolism of the small diameter samples. The cell concentration in the G1 samples is the
half value of the concentration in the G3 samples and one-fourth less than the value of the
Figure 6.
Beating frequencies for the BioPacer samples (Low cell density 500
µ
m (G1), Mix 500
µ
m
(G2), 500
µ
m (G3), 2
×
500
µ
m (G5), 3
×
500
µ
m (G6), Mix 1 mm (G9) and 1 mm (G7) on the Day 4, 8
and 12 of cultivation (Data are expressed as mean ±SD, n= 3, * p< 0.05, ** p< 0.01.
2.4. Glucose and Lactate Measurements
The glucose and lactate levels were measured for the different configurations of the
BioPacer for the constructs cultivated for 14 days. In Figure 7, the consumption of glucose
is plotted for the six different configurations of the BioPacer. The highest level of glucose
consumption was achieved for the G7 sample presenting significant differences among all
the other sample types. For the samples with different diameters, there are more significant
differences observed between the samples (G1 and G7, G2 and G7, G3 and G7, G4 and G7)
than the differences when comparing samples with the same diameter (G3 and G4, G7 and
G8). As a result, the diameter plays an important role in the glucose consumption of the
samples. Furthermore, the lactate production value of the G4 sample was reported higher
than the G3 sample, reflecting the positive effect of the applied stress in the metabolism of
the small diameter samples. The cell concentration in the G1 samples is the half value of the
concentration in the G3 samples and one-fourth less than the value of the cell concentration
in the cell concentration in the G7 samples. Therefore, the glucose provided by the new
media change was a lot more than the cells embedded in the G1 samples needed to consume.
Concerning the G4 sample there was higher glucose consumption and simultaneously
higher lactate production than the G3 sample and both metabolites followed the same
trend with slight differences in their values during the cultivation period. However, a
highly significant difference in glucose consumption was shown between the G7 and
the G8 samples. In the G7 samples, the glucose levels were decreasing while the lactate
consumption was increasing during the first week of the cultivation. This can be explained
by the fact that the iPSC-CMs were isolated by the high increase in the HUVSMCs in the
constructs after the first week of cultivation. Therefore, the proliferation of HUVSMCs
acted as an inhibitory factor to the effective iPSC-CM function, as the last ones seemed to
affect glucose consumption and lactate production to a high extent.
Gels 2023,9, 677 9 of 17
Gels 2023, 9, x FOR PEER REVIEW 9 of 17
cell concentration in the cell concentration in the G7 samples. Therefore, the glucose pro-
vided by the new media change was a lot more than the cells embedded in the G1 samples
needed to consume. Concerning the G4 sample there was higher glucose consumption
and simultaneously higher lactate production than the G3 sample and both metabolites
followed the same trend with slight differences in their values during the cultivation pe-
riod. However, a highly significant difference in glucose consumption was shown be-
tween the G7 and the G8 samples. In the G7 samples, the glucose levels were decreasing
while the lactate consumption was increasing during the first week of the cultivation. This
can be explained by the fact that the iPSC-CMs were isolated by the high increase in the
HUVSMCs in the constructs after the first week of cultivation. Therefore, the proliferation
of HUVSMCs acted as an inhibitory factor to the effective iPSC-CM function, as the last
ones seemed to affect glucose consumption and lactate production to a high extent.
Figure 7. Glucose and lactate levels for the BioPacer constructs. Percentage of glucose consumption
and lactate production for the Day 14 of the cultivation period. Data are expressed as mean ± SD, n
= 3 (* p < 0.05, *** p < 0.001, **** p < 0.0001).
2.5. Discussion
In this study, we presented the expression of the iPSC-CM markers for the ex-vivo
cultivation of fibrin hydrogels embedded with iPSC-CMs, HUVECs and HUVSMCs.
There are different configurations of the BioPacer sample that were tested for glucose and
lactate measurements to define the viability of the construct and for cell morphology to
visualize the cell expansion inside the fibrin matrix. Additionally, the expression of iPSC-
CM markers SAA, Cx43, col I and CD31 were investigated for the HUVSMCs and HU-
VECs, respectively. Moreover, the beating frequency of the iPSC-CMs in the fibrin hydro-
gels was evaluated. All these factors affect the performance of the BioPacer as a beating
construct in vitro.
Furthermore, the selection of the stretch sample confirmed that the stretch constitutes
an important factor in the preparation of the BioPacer constructs. Uniaxial stretch has been
Figure 7.
Glucose and lactate levels for the BioPacer constructs. Percentage of glucose consumption
and lactate production for the Day 14 of the cultivation period. Data are expressed as mean
±
SD,
n= 3 (* p< 0.05, *** p< 0.001, **** p< 0.0001).
2.5. Discussion
In this study, we presented the expression of the iPSC-CM markers for the ex-vivo
cultivation of fibrin hydrogels embedded with iPSC-CMs, HUVECs and HUVSMCs. There
are different configurations of the BioPacer sample that were tested for glucose and lactate
measurements to define the viability of the construct and for cell morphology to visualize
the cell expansion inside the fibrin matrix. Additionally, the expression of iPSC-CM markers
SAA, Cx43, col I and CD31 were investigated for the HUVSMCs and HUVECs, respectively.
Moreover, the beating frequency of the iPSC-CMs in the fibrin hydrogels was evaluated.
All these factors affect the performance of the BioPacer as a beating construct in vitro.
Furthermore, the selection of the stretch sample confirmed that the stretch constitutes
an important factor in the preparation of the BioPacer constructs. Uniaxial stretch has
been proven to positively affect the cytosolic calcium levels in iPSC-CMs [
35
] increase the
sarcomere length of iPSC-CMs [
36
] and successfully differentiated
in vitro
using mechan-
ical stretch [
37
]. In our two photon microscope pictures, we assume that the expression
of SAA, Cx43, CD31 and col I were respective to the cell concentration of the samples,
the molding technique and also the stretch. Particularly, the uniaxial cyclic stretch affects
the cell elongation of the cells along the fibrin gel tube in comparison to the static culture
(Figure 2)
. Moreover, the DAPI, SAA, Cx43, Col I and CD31 expression
in Figures 3and 4
were proportional to the low cell concentration of the iPSC-CMs, to the molding technique
and to the applied uniaxial cyclic stretch. Our results for the stretcher were in accordance
with another study where cardiac patches made of matrigel were subjected to cyclic me-
chanical stimulation. In this study, the expression of SAA and Cx43 is totally affected by the
applied stretch on the cardiac patches and is definitely affecting the cell alignment [
38
]. On
the other hand, another type of stretch, the cyclic sinusoidal strain increases their sarcomere
orientation perpendicular to the axis of strain [39].
Gels 2023,9, 677 10 of 17
As a further step, we believe that the combination of an electrical stimulation system
together with the mechanical stimulation achieved from our cyclic stretch bioreactor would
serve as a complete construct with both types of stimulation achieving the ideal function-
ality of the cylindrical hydrogel. Recently, Dou et al. created a biosensing platform for
the electrical stimulation of the iPSC-CMs with the simultaneous contraction and electro-
physiology measurements of the monolayers [
40
]. Therefore, the monolayers corresponded
well to the electrical stimulation, as the Cx43 signal for the electrical coupling and the
SAA signal, which indicates the maturity of the cells, were expressed accordingly to the
applied electrical stimuli. In another study, electrical and mechanical stimulation were also
combined for translational applications [41].
Concerning the beating frequency of the samples, the G2 sample together with the
G5 sample and the G6 sample performed more efficiently. As it was recently investigated,
the presence of fibroblasts in the culture is beneficial for the electrical pacing of the con-
structs [
42
]. In our case, we used the fibroblasts to stabilize the fibrin core via collagen
production, which was additionally increased after the applied stretch (Figure 6). Con-
sequently, the increased fibroblast concentration led to decreased beating frequency over
the culture. Moreover, the beating frequency of our constructs could also increase in case
of electrical stimulation as reported in a study where iPSC-CM spheroids were mechani-
cally and uniaxial cyclically stimulated [
43
]. Another study used special electrodes for the
electrical stimulation of iPSC-CMs with the further aim to synchronize the beating [44].
As for the glucose and lactate measurements, the diameter of the samples plays
an important role in glucose consumption. This is reflected by the fact that the glucose
consumption of the G7 sample was significantly higher than the G3 sample value. At the
same time, the glucose consumption of the G3 sample was significantly lower than the
respective value of the G4 sample, indicating that the cell metabolism has been increased
because of the stretch. However, the glucose consumption in the G7 samples is significantly
higher than the respective value of the G8 sample, demonstrating that the effect of stretch on
the consumption of glucose also depends on the diameter of the constructs. Moreover, the
simultaneous comparison of the lactate results represented a highly significant difference
between the G3 and the G4 sample. Consequently, for lactate production, only the stretch
is a determinant factor that affects the cell metabolism and not the diameter. In other
studies investigating the interactions between smooth muscle cells and endothelial cells,
the glucose and lactate levels varied during the culture [
45
]. Considering only the first
week of the cultivation period, the glucose levels changed slightly and the same tendency
was observed for the lactate levels as stated in a study of the biological vasculature using
endothelial cells in a microfluidic device [
46
]. Particularly, our results for the different
tendencies between glucose consumption and lactate production (Figure 7) over the entire
culture period may be explained by the different initial cell concentrations in the constructs.
For example, Heywood et al. showed that the glucose and lactate rates were affected by
the cell concentration in particular chondrocytes [
47
]. Specifically in our study, the lactate
levels were either slightly increasing (G4 sample) or abruptly increasing after the first week
of the culture period. However, there is further investigation needed for these results, as
it has been proven that the high levels of lactate in the culture are correlated with fibrin
clotting in neutrophil extracellular trap formation [
48
]. Last but not least, considering the
cultivation time, Zheng et al. proved that the glucose and lactate levels increased during
the first minutes of the cultivation in a fibrin gel acting as a barrier for tumor cell migration
based on the property of fibrin to clot [
49
]. Furthermore, our glucose and lactate results
for the G4 sample during the culture period followed a similar tendency to a study where
fibrin gels in the geometry of the heart valve were dynamically cultivated and the rates of
both glucose and lactate do not steeply change [
50
]. Particularly for the iPSC-CMs which
were the main component of all the constructs, they have been reported to maintain the
glucose levels high in the culture and the lactate levels relatively low during the first days
of cultivation in a study for drug testing [
51
]. That is why we concluded that the iPSC-CMs
Gels 2023,9, 677 11 of 17
cannot provide specific information for the explanation of these results when they are
embedded in the fibrin gel.
3. Conclusions
We managed to fabricate cylindrical constructs which contain three cell types
(HUVSMCs, HUVECs and iPSC-CMs) and subject them to cyclic stretch. The glucose
and lactate levels of the constructs and the presence of iPSC-CM markers together with
HUVEC and HUVSMC markers indicate the function of the constructs according to the di-
ameter and the cyclic stretch. By adjusting the cell composition and concentration together
with the cyclic stretch iPSC-CM embedded fibrin cylindrical hydrogels can be produced
in vitro
with the further aim to be used in the treatment of the atrioventricular block in
pediatric patients.
4. Methods
4.1. Cylindrical Fibrin Gel Constructs
500
µ
m and 1 mm diameter translucent silicone tubing was incubated overnight in 1%
pluronic solution (Sigma-Aldrich, Saint Louis, MO, USA) at room temperature. The air was
removed using a needle and a syringe (B. Braun, Melsungen, Germany). The concentra-
tion of the cell suspension was 3 million/mL, 1 million/mL and 15 million/mL for HU-
VSMCs, HUVECs and iPSC-CMs, respectively. Lyophilized fibrinogen from human plasma
(Sigma-Aldrich, Saint Louis, MO, USA) was dissolved in purified water and dialyzed
against Tris-buffered saline (TBS) using Spectra/Por 1 tubing (Spectrum) with a molecular
weight cut-off of 6000–8000 Da. After sterile filtration, the fibrinogen concentration was
determined by measuring the absorbance at 280 nm using an Infinite M200 spectropho-
tometer (Tecan Group Ltd., Männedorf, Switzerland). The cells were detached from the
flasks and resuspended in a solution consisting of 50% Tris-buffered saline (TBS) solution,
25% 40 U/mL thrombin (Sigma-Aldrich, Saint Louis, MO, USA) and 25% 50 mM CaCl
2
.
The polymerization took place via a dual injection head and mixer (Medmix Systems AG,
Switzerland) connected to two syringes containing the fibrinogen and the cell suspension,
respectively (Figure 8). Additionally, the syringes were held together with Duploject double
syringe holder (Tisseel, Baxter, United Kingdom). After 5 min, the fibers were extruded
from the tubes by adding DMEM and were collected in a DMEM bath in a Petri dish. The
fibers were cut into different lengths and were placed either in 6 well plates for the static
cultivation or in the cyclic stretch bioreactor for the dynamic cultivation condition. The
samples were cultivated with a medium consisting of 25% DMEM, 25% EGM-2 and 50%
Plyricyte medium supplemented with L-ascorbic acid-2-phosphate (1.0 mM; Sigma) and
antibiotic medium (ABM, Pan Biotech, Aidenbach, Germany). The medium was changed
on days 2, 4, 6, 8, 10 and 12 and the constructs were cultivated either for 7 or 14 days to
study the appropriate functionality of the BioPacer constructs. The procedure described
above was followed for the preparation of the BioPacer configurations listed in Table 1.
4.2. Cell Culturing
Human umbilical vein smooth muscle cells (HUVSMCs) and human umbilical vein
endothelial cells (HUVECs) were handled as described previously [52]. Human umbilical
cords were obtained after written consent at the University Hospital Aachen, Aachen,
Germany, and were provided by the RWTH Aachen University Centralized Biomaterial
Bank (cBMB) according to its regulations, following RWTH Aachen University, Medical
Faculty Ethics
Gels 2023,9, 677 12 of 17
Gels 2023, 9, x FOR PEER REVIEW 12 of 17
described above was followed for the preparation of the BioPacer configurations listed in
Table 1 .
Figure 8. Conduction of the experiment.
4.2. Cell Culturing
Human umbilical vein smooth muscle cells (HUVSMCs) and human umbilical vein
endothelial cells (HUVECs) were handled as described previously [52]. Human umbilical
cords were obtained after wrien consent at the University Hospital Aachen, Aachen, Ger-
many, and were provided by the RWTH Aachen University Centralized Biomaterial Bank
(cBMB) according to its regulations, following RWTH Aachen University, Medical Faculty
Ethics
Commiee approval (cBMB project number 323). The medium used was Dulbecco´s
Modified Eagle’s Medium (DMEM, Thermofischer, Waltham, MA, USA) supplemented
with 10% FCS for the HUVSMCs and EGM (PromoCell, Heidelberg, Germany) supple-
mented with 1% FCS, basic Fibroblast Growth Factor, Insulin-like Growth Factor, Vascular
Endothelial Growth Factor 165, Ascorbic Acid, Heparin and Hydrocortisone for the HU-
VECs, respectively. Both cell types were cultured in flasks and trypsin (Thermo Fisher,
Waltham, MA, USA) was used for cell dissociation. For the HUVECs flasks were coated
with 2% gelatin (Sigma-Aldrich, Saint Louis, MO, USA) before culture. The cells were cul-
tured at 37 °C, 5% CO2, the medium was changed every 3 days and they were up to pas-
sage 5.
Induced pluripotent stem cell cardiomyocytes (iPSC-CMs) (Ncardia, Leiden, The
Netherlands) and were cultured in 10 µg/mL fibronectin (Sigma-Aldrich, Saint Louis, MO,
USA). TrypLE Select Enzym (1×, Thermofischer, Waltham, MA, USA) was used for the
trypsinization. Plyricyte medium (Ncardia, Leiden, The Netherlands) was used for the
cultivation of the iPSC-CMs.
4.3. Bioreactor Set-Up
A housemade cyclic stretch bioreactor was constructed in the workshop of the Ap-
plied Medical Engineering Department of the Helmhol Institute Aachen. The bioreactor
consisted of a chamber with positions for 6 samples held by two edges using a metal
holder and a tiny screw on one side. The bioreactor was sealed in a box whose walls were
made of transparent PTFE for the less complicated observation of the experiment. The
samples were stretched to 15% after 7 days of static cultivation. The setup is presented in
Figure 9. The software and the electrical supply setup were provided by Igus (Cologne,
Germany). The whole bioreactor system is presented in Supplementary Figure S4.
Figure 8. Conduction of the experiment.
Committee approval (cBMB project number 323). The medium used was Dulbecco’s
Modified Eagle’s Medium (DMEM, Thermofischer, Waltham, MA, USA) supplemented
with 10% FCS for the HUVSMCs and EGM (PromoCell, Heidelberg, Germany) supple-
mented with 1% FCS, basic Fibroblast Growth Factor, Insulin-like Growth Factor, Vascular
Endothelial Growth Factor 165, Ascorbic Acid, Heparin and Hydrocortisone for the HU-
VECs, respectively. Both cell types were cultured in flasks and trypsin (Thermo Fisher,
Waltham, MA, USA) was used for cell dissociation. For the HUVECs flasks were coated
with 2% gelatin (Sigma-Aldrich, Saint Louis, MO, USA) before culture. The cells were
cultured at 37
◦
C, 5% CO
2
, the medium was changed every 3 days and they were up to
passage 5.
Induced pluripotent stem cell cardiomyocytes (iPSC-CMs) (Ncardia, Leiden, The
Netherlands) and were cultured in 10
µ
g/mL fibronectin (Sigma-Aldrich, Saint Louis, MO,
USA). TrypLE Select Enzym (1
×
, Thermofischer, Waltham, MA, USA) was used for the
trypsinization. Plyricyte medium (Ncardia, Leiden, The Netherlands) was used for the
cultivation of the iPSC-CMs.
4.3. Bioreactor Set-Up
A housemade cyclic stretch bioreactor was constructed in the workshop of the Ap-
plied Medical Engineering Department of the Helmholtz Institute Aachen. The bioreactor
consisted of a chamber with positions for 6 samples held by two edges using a metal holder
and a tiny screw on one side. The bioreactor was sealed in a box whose walls were made
of transparent PTFE for the less complicated observation of the experiment. The samples
were stretched to 15% after 7 days of static cultivation. The setup is presented in Figure 9.
The software and the electrical supply setup were provided by Igus (Cologne, Germany).
The whole bioreactor system is presented in Supplementary Figure S4.
4.4. Glucose and Lactate Measurements
During the cultivation period, the glucose and lactate levels of the culture medium
were measured using an Epoc Reader (Epoca, Ottawa, ON, Canada).
Gels 2023,9, 677 13 of 17
Gels 2023, 9, x FOR PEER REVIEW 13 of 17
Figure 9. The BioPacer sample in the bioreactor system.
4.4. Glucose and Lactate Measurements
During the cultivation period, the glucose and lactate levels of the culture medium
were measured using an Epoc Reader (Epoca, Oawa, Canada).
4.5. Immunohistochemistry
After cultivation, the samples were fixed in a 4% PFA solution (Carl–Roth, Karlsruhe,
Germany) for 1 h and then washed 2 times with PBS (Thermofischer, Waltham, MA, USA).
The cell membrane was permeabilized using 5% normal goat serum and 0.1% Triton X-
100 (Sigma-Aldrich, Saint Louis, MO, USA) in PBS. The samples were incubated with the
primary antibody and incubated overnight at 37 °C. Afterward, the samples were washed
2 times with PBS and incubated for 8 h at 37 °C. Furthermore, the samples were incubated
Figure 9. The BioPacer sample in the bioreactor system.
4.5. Immunohistochemistry
After cultivation, the samples were fixed in a 4% PFA solution (Carl–Roth, Karlsruhe,
Germany) for 1 h and then washed 2 times with PBS (Thermofischer, Waltham, MA, USA).
The cell membrane was permeabilized using 5% normal goat serum and 0.1% Triton X-
100 (Sigma-Aldrich, Saint Louis, MO, USA) in PBS. The samples were incubated with
the primary antibody and incubated overnight at 37
◦
C. Afterward, the samples were
washed 2 times with PBS and incubated for 8 h at 37
◦
C. Furthermore, the samples were
incubated with the secondary antibody following the same conditions applied for the
primary antibody incubation. After the last washing step, the nuclei were stained with
DAPI and incubated for 15 min at 37
◦
C and washed 3 times. The samples were stored at
4
◦
C in PBS containing ABM until visualization in a 1% antibiotic-antimycotic solution. In
Table 2, the primary and secondary antibodies used are listed. Samples were embedded in
2% agarose gel in PBS and visualized by two-photon laser-scanning microscopy (TPLSM)
Gels 2023,9, 677 14 of 17
using an Olympus FluoView 1000MPE with a 25
×
water objective (NA 1.05, Olympus,
Tokyo, Japan), a mode-locked MaiTai DeepSee Titanium-Sapphire Laser (Spectra-Physics,
Stahnsdorf, Germany) and FluoView FV 10 4.2 acquisition software.
Table 2. Antibodies for the staining of the samples.
Antibody Dilution Supply
Anti-α-Actinin (sarcomeric), mouse monoclonal 1:50 Sigma-Aldrich
Connexin 43 (polyclonal) 1:50 Thermofischer
Connexin 43 Monoclonal Antibody (3D8A5) 1:50 Thermofischer
Anti-CD31 (PECAM-1), mouse monoclonal 1:100 Sigma-Aldrich
CD31 rabbit polyclonal 1:100 Abbiotec
α-smooth muscle actin (monoclonal, clone 1A4) 1:1000 Sigma-Aldrich
Anti-Collagen I antibody (ab34710) 1:200 Abcam
4.6. Beating Frequency
Samples were placed in a Nikon Ti-Eclipse epifluorescence microscope TI-S-CON
(Tokyo, Japan) at 37
◦
C and 5% CO
2
using an Okolab heating box (Ottaviano, NA, Italy). A
video with a duration of 1 min was recorded for each sample and the beating frequency
per minute was calculated. The iPSC-CM spots beating simultaneously were selected for
these measurements.
4.7. Quantification of the Markers Content
The Imaris software was used to evaluate the images captured at the two-photon
laser-scanning microscope. The pixels were counted using the masking option in the
surface’s menu for the three different colors (blue, red, green) and the whole set of volume
and surfaces was used for the quantification of the different markers. The consumption of
glucose and the production of lactate values were expressed as percentages based on the
difference between the initial and the final value of every marker, respectively.
4.8. Statistics
All data were represented as mean-SD from three experiments. For the glucose and
lactate measurements, uncorrected Fisher’s LSD test with one-way analysis of variance
(ANOVA) was performed. For the quantification of the TPLSM results and the beating
frequency measurements, Tukey’s multiple comparisons test with two-way analysis of
variance (ANOVA) was selected. A p-value < 0.05 was considered statistically significant.
Supplementary Materials:
The following supporting information can be downloaded at: https://
www.mdpi.com/article/10.3390/gels9090677/s1, Figure S1: SAA (red), Cx43 (green) and DAPI (blue)
for the Biopacer samples without and with cyclic stress (cross-sectional cuts), Figure S2: DAPI (blue),
SAA (green), Cx43 (red) and collagen I (green) staining. Left: Low cell density 500 µm, middle: Mix
500 µm, right: 500 µm stretch (cross-sectional cuts), Figure S3: DAPI (blue), SAA (green), Cx43 (red)
and CD31 (red) staining. Left: Low cell density 500
µ
m, middle: Mix 500
µ
m, right: 500
µ
m stretch
(cross-sectional cuts), Figure S4. Bioreactor set-up, Video S1: G6 (3 sample at Day 8).
Author Contributions:
Writing-original draft preparation: S.K. Writing-review editing: S.J. Concep-
tualization: S.J. and S.K. Data curation: S.K. Methodology: S.K. and A.L. Investigation: S.K. and A.L.
Resources: S.J., A.L., C.A.S. and Y.K. Supervision: S.J. Software: S.K. Validation: S.K. Formal Analysis:
S.K. Visualization: S.K. Project Administration: S.J. Funding Acquisition: S.J. All authors have read
and agreed to the published version of the manuscript.
Funding: This research was funded by Stiftung Kinderherz (Project number 2511-5-00-002).
Gels 2023,9, 677 15 of 17
Institutional Review Board Statement:
Human umbilical cords were obtained in accordance with
the Declaration of Helsinki after written consent at University Hospital Aachen, Aachen, Germany,
and were provided by the RWTH Aachen University Centralized Biomaterial Bank (cBMB) according
to its regulations, following RWTH Aachen University, Medical Faculty Ethics Committee approval
(cBMB project number 323).
Informed Consent Statement: Not applicable.
Data Availability Statement:
Data associated with this study is available upon request to the corre-
sponding authors.
Acknowledgments:
This work was funded by the Stiftung Kinderherz. Furthermore, the authors
acknowledge the support by the core facility Two-Photon Imaging, a core facility of the Interdis-
ciplinary Center for Clinical Research (IZKF) Aachen within the Faculty of Medicine at RWTH
Aachen University.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Baruteau, A.E.; Pass, R.H.; Thambo, J.B.; Behaghel, A.; Le Pennec, S.; Perdreau, E.; Combes, N.; Liberman, L.; McLeod, C.J.
Congenital and childhood atrioventricular blocks: Pathophysiology and contemporary management. Eur. J. Pediatr.
2016
,175,
1235–1248. [CrossRef]
2.
Bartakian, S.; Fagan, T.E.; Schaffer, M.S.; Darst, J.R. Device closure of secundum atrial septal defects in children <15 kg:
Complication rates and indications for referral. JACC Cardiovasc. Interv. 2012,5, 1178–1184.
3.
Vogler, J.; Breithardt, G.; Eckardt, L. Bradyarrhythmias and Conduction Blocks. Rev. Española Cardiol. (Engl. Ed.)
2012
,65, 656–667.
[CrossRef]
4.
Aksu, T.; Gopinathannair, R.; Bozyel, S.; Yalin, K.; Gupta, D. Cardioneuroablation for Treatment of Atrioventricular Block. Circ.
Arrhythmia Electrophysiol. 2021,14, e010018. [CrossRef]
5.
Yeung, C.; Baranchuk, A. Systematic Approach to the Diagnosis and Treatment of Lyme Carditis and High-Degree Atrioventricular
Block. Healthcare 2018,6, 119. [CrossRef]
6.
Pyngottu, A.; Werner, H.; Lehmann, P.; Balmer, C. Health-Related Quality of Life and Psychological Adjustment of Children and
Adolescents with Pacemakers and Implantable Cardioverter Defibrillators: A Systematic Review. Pediatr. Cardiol.
2019
,40, 1–16.
[CrossRef]
7.
Gheissari, A.; Hordof, A.J.; Spotnitz, H.M. Transvenous pacemakers in children: Relation of lead length to anticipated growth.
Ann. Thorac. Surg. 1991,52, 118–121. [CrossRef] [PubMed]
8.
Song, M.K.; Kim, N.Y.; Bae, E.J.; Kim, G.B.; Kwak, J.G.; Kim, W.H.; Lee, J.R. Long-term Follow-up of Epicardial Pacing and Left
Ventricular Dysfunction in Children With Congenital Heart Block. Ann. Thorac. Surg.
2020
,109, 1913–1920. [CrossRef] [PubMed]
9.
Silvetti, M.S.; Drago, F.; Ravà, L. Determinants of early dilated cardiomyopathy in neonates with congenital complete atrio-
ventricular block. Europace 2010,12, 1316–1321. [CrossRef] [PubMed]
10.
Cate, F.E.U.T.; Breur, J.M.; Cohen, I.M.; Boramanand, N.; Kapusta, L.; Crosson, E.J.; Brenner, J.I.; Lubbers, L.J.; Friedman, A.H.;
Vetter, V.L.; et al. Dilated cardiomyopathy in isolated congenital complete atrioventricular block: Early and long-term risk in
children. J. Am. Coll. Cardiol. 2001,37, 1129–1134. [CrossRef] [PubMed]
11.
Liberman, L.; Silver, E.S.; Chai, P.J.; Anderson, B.R. Incidence and characteristics of heart block after heart surgery in pediatric
patients: A multicenter study. J. Thorac. Cardiovasc. Surg. 2016,152, 197–202. [CrossRef]
12. McLeod, K.A. Cardiac pacing in infants and children. Heart 2010,96, 1502–1508. [CrossRef]
13.
Gorabi, A.M.; Hajighasemi, S.; Khori, V.; Soleimani, M.; Rajaei, M.; Rabbani, S.; Atashi, A.; Ghiaseddin, A.; Saeid, A.K.; Tafti, H.A.;
et al. Functional biological pacemaker generation by T-Box18 protein expression via stem cell and viral delivery approaches in a
murine model of complete heart block. Pharmacol. Res. 2019,141, 443–450. [CrossRef]
14.
Lu, W.; Yaoming, N.; Boli, R.; Jun, C.; Changhai, Z.; Yang, Z.; Zhiyuan, S. mHCN4 Genetically Modified Canine Mesenchymal
Stem Cells Provide Biological Pacemaking Function in Complete Dogs with Atrioventricular Block. Pacing Clin. Electrophysiol.
2013,36, 1138–1149. [CrossRef] [PubMed]
15.
Yokokawa, M.; Ohnishi, S.; Ishibashi-Ueda, H.; Obata, H.; Otani, K.; Miyahara, Y.; Tanaka, K.; Shimizu, W.; Nakazawa, K.;
Kangawa, K. Transplantation of mesenchymal stem cells improves atrioventricular conduction in a rat model of com-plete
atrioventricular block. Cell Transplant. 2008,17, 1145–1155. [CrossRef] [PubMed]
16.
Chauveau, S.; Anyukhovsky, E.P.; Ben-Ari, M.; Naor, S.; Jiang, Y.-P.; Danilo, P.; Rahim, T.; Burke, S.; Qiu, X.; Potapova, I.A.;
et al. Induced Pluripotent Stem Cell–Derived Cardiomyocytes Provide In Vivo Biological Pacemaker Function. Circ. Arrhythmia
Electrophysiol. 2017,10, e004508. [CrossRef] [PubMed]
17.
Cingolani, E.; Ionta, V.; Cheng, K.; Giacomello, A.; Cho, H.C.; Marbán, E. Engineered electrical conduction tract restores
conduction in complete heart block: From
in vitro
to
in vivo
proof of concept. J. Am. Coll. Cardiol.
2014
,64, 2575–2585. [CrossRef]
[PubMed]
Gels 2023,9, 677 16 of 17
18.
Miklas, J.W.; Nunes, S.S.; Sofla, A.; Reis, L.A.; Pahnke, A.; Xiao, Y.; Laschinger, C.; Radisic, M. Bioreactor for modulation of cardiac
microtissue phenotype by combined static stretch and electrical stim-ulation. Biofabrication 2014,6, 024113. [CrossRef]
19.
Nunes, S.S.; Miklas, J.W.; Liu, J.; Aschar-Sobbi, R.; Xiao, Y.; Zhang, B.; Jiang, J.; Massé, S.; Gagliardi, M.; Hsieh, A.; et al. Biowire:
A platform for maturation of human pluripotent stem cell–derived cardiomyocytes. Nat. Methods 2013,10, 781–787. [CrossRef]
20.
Kensah, G.; Lara, A.R.; Dahlmann, J.; Zweigerdt, R.; Schwanke, K.; Hegermann, J.; Skvorc, D.; Gawol, A.; Azizian, A.;
Wagner, S.; et al. Murine and human pluripotent stem cell-derived cardiac bodies form contractile myocardial tissue
in vitro
.Eur.
Heart J. 2013,34, 1134–1146. [CrossRef]
21.
Keijdener, H.; Konrad, J.; Hoffmann, B.; Gerardo-Nava, J.; Rütten, S.; Merkel, R.; Vázquez-Jiménez, J.; Brook, G.A.; Jockenhoevel,
S.; Mela, P. A bench-top molding method for the production of cell-laden fibrin micro-fibers with longitudinal topog-raphy.
J. Biomed. Mater. Res. Part B Appl. Biomater. 2020,108, 1198–1212. [CrossRef] [PubMed]
22.
Jockenhoevel, S.; Zund, G.; Hoerstrup, S.P.; Chalabi, K.; Sachweh, J.S.; Demircan, L.; Messmer, B.J.; Turina, M. Fibrin gel—
Advantages of a new scaffold in cardiovascular tissue engineering. Eur. J. Cardio-Thorac. Surg. 2001,19, 424–430. [CrossRef]
23.
de Melo, B.A.; Jodat, Y.A.; Cruz, E.M.; Benincasa, J.C.; Shin, S.R.; Porcionatto, M.A. Strategies to use fibrinogen as bioink for 3D
bioprinting fibrin-based soft and hard tissues. Acta Biomater. 2020,117, 60–76. [CrossRef] [PubMed]
24.
Lim, S.Y.; Hsiao, S.T.; Lokmic, Z.; Sivakumaran, P.; Dusting, G.J.; Dilley, R.J. Ischemic Preconditioning Promotes Intrinsic
Vascularization and Enhances Survival of Implanted Cells in an In Vivo Tissue Engineering Model. Tissue Eng. Part A
2012
,18,
2210–2219. [CrossRef] [PubMed]
25.
Kim, J.; Wu, B.; Niedzielski, S.M.; Hill, M.T.; Coleman, R.M.; Ono, A.; Shikanov, A. Characterizing natural hydrogel for
reconstruction of three-dimensional lymphoid stromal network to model T-cell interactions. J. Biomed. Mater. Res. Part A
2015
,
103, 2701–2710. [CrossRef]
26.
Weisel, J.; Cederholm-Williams, S. Fibrinogen and fibrin: Characterization, processing and medical applications. In Handbook of
Biodegradable Polymers; Harwood: Amsterdam, The Netherlands, 1997; pp. 347–365.
27.
Rubalskii, E.; Ruemke, S.; Salmoukas, C.; Aleshkin, A.; Bochkareva, S.; Modin, E.; Mashaqi, B.; Boyle, E.C.; Boethig, D.; Rubalsky,
M.; et al. Fibrin glue as a local drug-delivery system for bacteriophage PA5. Sci. Rep. 2019,9, 2091. [CrossRef]
28.
Wu, X.; Ren, J.; Li, J. Fibrin glue as the cell-delivery vehicle for mesenchymal stromal cells in regenerative medicine. Cytotherapy
2012,14, 555–562. [CrossRef]
29.
Luo, L.; Foster, N.C.; Man, K.L.; Brunet, M.; Hoey, D.A.; Cox, S.C.; Kimber, S.J.; El Haj, A.J. Hydrostatic pressure promotes
chondrogenic differentiation and microvesicle release from human embryonic and bone marrow stem cells. Biotechnol. J.
2022
,
17, e2100401. [CrossRef]
30.
Graceffa, V. Development of a fibrin-mediated gene delivery system for the treatment of cystinosis via design of experiment. Sci.
Rep. 2022,12, 3752. [CrossRef]
31.
Zhou, Y.; Fan, Y.; Chen, Z.; Yue, Z.; Wallace, G. Catechol functionalized ink system and thrombin-free fibrin gel for fabricating
cellular constructs with me-chanical support and inner micro channels. Biofabrication 2022,14, 015004. [CrossRef]
32.
Uehlin, A.F.; Vines, J.B.; Feldman, D.S.; Nyairo, E.; Dean, D.R.; Thomas, V. Uni-Directionally Oriented Fibro-Porous PLLA/Fibrin
Bio-Hybrid Scaffold: Mechano-Morphological and Cell Studies. Pharmaceutics 2022,14, 277. [CrossRef]
33.
de Torre, I.G.; Weber, M.; Quintanilla, L.; Alonso, M.; Jockenhoevel, S.; Cabello, J.C.R.; Mela, P. Hybrid elastin-like recombinamer-
fibrin gels: Physical characterization and
in vitro
evaluation for cardio-vascular tissue engineering applications. Biomater. Sci.
2016,4, 1361–1370. [CrossRef] [PubMed]
34.
Sawadkar, P.; Mandakhbayar, N.; Patel, K.D.; Buitrago, J.O.; Kim, T.H.; Rajasekar, P.; Lali, F.; Kyriakidis, C.; Rahmani, B.;
Mohanakrishnan, J. Three dimensional porous scaffolds derived from collagen, elastin and fibrin proteins orchestrate adipose
tissue regeneration. J. Tissue Eng. 2021,12. [CrossRef] [PubMed]
35.
Moreira, R.; Velz, T.; Alves, N.; Gesche, V.N.; Malischewski, A.; Schmitz-Rode, T.; Frese, J.; Jockenhoevel, S.; Mela, P. Tissue-
Engineered Heart Valve with a Tubular Leaflet Design for Minimally Invasive Transcatheter Implantation. Tissue Eng. Part C
Methods 2015,21, 530–540. [CrossRef] [PubMed]
36.
Lu, J.; Lee, Y.-K.; Ran, X.; Lai, W.-H.; Li, R.A.; Keung, W.; Tse, K.; Tse, H.-F.; Yao, X. An abnormal TRPV4-related cytosolic Ca2+ rise
in response to uniaxial stretch in induced pluripotent stem cells-derived cardiomyocytes from dilated cardiomyopathy patients.
Biochim. Et Biophys. Acta (BBA)—Mol. Basis Dis. 2017,1863, 2964–2972. [CrossRef] [PubMed]
37.
Song, M.; Jang, Y.; Kim, S.-J.; Park, Y. Cyclic Stretching Induces Maturation of Human-Induced Pluripotent Stem Cell-Derived
Cardiomyocytes through Nuclear-Mechanotransduction. Tissue Eng. Regen. Med. 2022,19, 781–792. [CrossRef]
38.
Yu, T.; Miyagawa, S.; Miki, K.; Saito, A.; Fukushima, S.; Higuchi, T.; Kawamura, M.; Kawamura, T.; Ito, E.; Kawaguchi, N.; et al.
In Vivo Differentiation of Induced Pluripotent Stem Cell-Derived Cardiomyocytes. Circ. J. 2013,77, 1297–1306. [CrossRef]
39.
Lux, M.; Andrée, B.; Horvath, T.; Nosko, A.; Manikowski, D.; Hilfiker-Kleiner, D.; Haverich, A.; Hilfiker, A.
In vitro
maturation of
large-scale cardiac patches based on a perfusable starter matrix by cyclic mechanical stimulation. Acta Biomater.
2016
,30, 177–187.
[CrossRef]
40.
Kreutzer, J.; Viehrig, M.; Pölönen, R.-P.; Zhao, F.; Ojala, M.; Aalto-Setälä, K.; Kallio, P. Pneumatic unidirectional cell stretching
device for mechanobiological studies of cardiomyocytes. Biomech. Model. Mechanobiol. 2020,19, 291–303. [CrossRef]
41.
Dou, W.; Malhi, M.; Cui, T.; Wang, M.; Wang, T.; Shan, G.; Law, J.; Gong, Z.; Plakhotnik, J.; Filleter, T. A Carbon-Based Biosensing
Platform for Simultaneously Measuring the Contraction and Electrophysiology of iPSC-Cardiomyocyte Monolayers. ACS Nano
2022,16, 11278–11290. [CrossRef]
Gels 2023,9, 677 17 of 17
42.
Kroll, K.; Chabria, M.; Wang, K.; Häusermann, F.; Schuler, F.; Polonchuk, L. Electro-mechanical conditioning of human iPSC-
derived cardiomyocytes for translational research. Prog. Biophys. Mol. Biol. 2017,130, 212–222. [CrossRef] [PubMed]
43.
Burnham, M.P.; Harvey, R.; Sargeant, R.; Fertig, N.; Haddrick, M. A Scalable Approach Reveals Functional Responses of iPSC
Cardiomyocyte 3D Spheroids. SLAS Discov. Adv. Sci. Drug Discov. 2021,26, 352–363. [CrossRef] [PubMed]
44.
LaBarge, W.; Mattappally, S.; Kannappan, R.; Fast, V.G.; Pretorius, D.; Berry, J.L.; Zhang, J. Maturation of three-dimensional,
hiPSC-derived cardiomyocyte spheroids utilizing cyclic, uniaxial stretch and electrical stimulation. PLoS ONE
2019
,14, e0219442.
45.
Yoshida, S.; Sumomozawa, K.; Nagamine, K.; Nishizawa, M. Hydrogel microchambers integrated with organic electrodes for
efficient electrical stimulation of human iPSC-derived cardiomyocytes. Macromol. Biosci.
2019
,19, 1900060. [CrossRef] [PubMed]
46.
Tan, A.; Fujisawa, K.; Yukawa, Y.; Matsunaga, Y.T. Bottom-up fabrication of artery-mimicking tubular co-cultures in collagen-based
microchannel scaffolds. Biomater. Sci. 2016,4, 1503–1514. [CrossRef]
47.
Schimek, K.; Busek, M.; Brincker, S.; Groth, B.; Hoffmann, S.; Lauster, R.; Lindner, G.; Lorenz, A.; Menzel, U.; Sonntag, F.; et al.
Integrating biological vasculature into a multi-organ-chip microsystem. Lab Chip 2013,13, 3588–3598. [CrossRef]
48.
Heywood, H.K.; Bader, D.L.; Lee, D.A. Rate of oxygen consumption by isolated articular chondrocytes is sensitive to medium
glucose concentration. J. Cell. Physiol. 2006,206, 402–410. [CrossRef]
49.
Z ˛abczyk, M.; Natorska, J.; Janion-Sadowska, A.; Malinowski, K.P.; Janion, M.; Undas, A. Elevated Lactate Levels in Acute
Pulmonary Embolism Are Associated with Prothrombotic Fibrin Clot Properties: Contribution of NETs Formation. J. Clin. Med.
2020,9, 953. [CrossRef]
50.
Zheng, D.-W.; Hong, S.; Zhang, Q.-L.; Dong, X.; Pan, P.; Song, W.-F.; Cheng, S.-X.; Zhang, X.-Z. Controllable gelation of artificial
extracellular matrix for altering mass transport and improving cancer therapies. Nat. Commun. 2020,11, 4907. [CrossRef]
51.
Flanagan, T.C.; Cornelissen, C.; Koch, S.; Tschoeke, B.; Sachweh, J.S.; Schmitzrode, T.; Jockenhoevel, S. The
in vitro
development of
autologous fibrin-based tissue-engineered heart valves through optimised dynamic conditioning. Biomaterials
2007
,28, 3388–3397.
[CrossRef]
52.
Necela, B.M.; Velz, T.; Alves, N.; Gesche, V.N.; Malischewski, A.; Schmitz-Rode, T.; Frese, J.; Jockenhoevel, S.; Mela, P. The
antineoplastic drug, trastuzumab, dysregulates metabolism in iPSC-derived cardiomyocytes. Clin. Transl. Med.
2017
,6, 5.
[CrossRef] [PubMed]
Disclaimer/Publisher’s Note:
The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.
Available via license: CC BY 4.0
Content may be subject to copyright.
