Content uploaded by Sylvain Catros
Author content
All content in this area was uploaded by Sylvain Catros on Apr 16, 2021
Content may be subject to copyright.
Journal of Materials Science: Materials in Medicine (2019) 30:28
https://doi.org/10.1007/s10856-019-6230-1
S.I.: BIOFABRICATION AND BIOINKS FOR TISSUE ENGINEERING
Original Research
Micropatterning of endothelial cells to create a capillary-like
network with defined architecture by laser-assisted bioprinting
Olivia Kérourédan1,2,3 ●Jean-Michel Bourget1,6 ●Murielle Rémy1,2 ●Sylvie Crauste-Manciet4,5 ●Jérôme Kalisky1,2 ●
Sylvain Catros1,2,3 ●Noëlie B. Thébaud1,2,3 ●Raphaël Devillard1,2,3
Received: 12 December 2017 / Accepted: 1 February 2019
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
Development of a microvasculature into tissue-engineered bone substitutes represents a current challenge. Seeding of
endothelial cells in an appropriate environment can give rise to a capillary-like network to enhance prevascularization of
bone substitutes. Advances in biofabrication techniques, such as bioprinting, could allow to precisely define a pattern of
endothelial cells onto a biomaterial suitable for in vivo applications. The aim of this study was to produce a microvascular
network following a defined pattern and preserve it while preparing the surface to print another layer of endothelial cells. We
first optimise the bioink cell concentration and laser printing parameters and then develop a method to allow endothelial cells
to survive between two collagen layers. Laser-assisted bioprinting (LAB) was used to pattern lines of tdTomato-labeled
endothelial cells cocultured with mesenchymal stem cells seeded onto a collagen hydrogel. Formation of capillary-like
structures was dependent on a sufficient local density of endothelial cells. Overlay of the pattern with collagen I hydrogel
containing vascular endothelial growth factor (VEGF) allowed capillary-like structures formation and preservation of the
printed pattern over time. Results indicate that laser-assisted bioprinting is a valuable technique to pre-organize endothelial
cells into high cell density pattern in order to create a vascular network with defined architecture in tissue-engineered
constructs based on collagen hydrogel.
Graphical Abstract
These authors contributed equally: Olivia Kérourédan, Jean-Michel
Bourget
*Olivia Kérourédan
olivia.kerouredan@u-bordeaux.fr
1INSERM, Bioingénierie Tissulaire, U1026, 146 rue Léo Saignat,
F-33076 Bordeaux, France
2Université de Bordeaux, Bioingénierie Tissulaire, U1026, 146 rue
Léo Saignat, F-33076 Bordeaux, France
3CHU de Bordeaux, Services d’Odontologie et de Santé Buccale,
Place Amélie Raba Léon, F-33076 Bordeaux, France
4Université de Bordeaux, ARNA Laboratory, team
ChemBioPharm, U1212 INSERM –UMR 5320 CNRS, 146 rue
Léo Saignat, F-33076 Bordeaux, France
5CHU de Bordeaux, Pharmacie du Groupe Hospitalier Sud, Avenue
de Magellan, F-33604 Pessac, France
6Present address: Energie, matériaux et télécommunication, Institut
National de Recherche Scientifique, Varenne, QC, Canada
Supplementary information The online version of this article (https://
doi.org/10.1007/s10856-019-6230-1) contains supplementary
material, which is available to authorized users.
1234567890();,:
1234567890();,:
1 Introduction
Osteosarcoma and osteoporosis are diseases causing bone
loss that affect a growing number of people. Population
aging in industrialized countries is a key factor behind this
increase. Worldwide, more than two million bone grafts are
performed annually to fulfill bone defects in orthopedic
surgery, neurosurgery and dental surgery [1].
In clinical cases of bone loss, current therapeutic solu-
tions include the use of grafts from human, animal or
synthetic origins. In this context, autograft is the current
gold standard. However, it is available in limited quantities
and necessitates the creation of a second wound (the donor
site) subject to infections, pain and morbidity. The use of
allograft and xenograft materials presents major risks of
transplant rejection and transmission of pathogens. Dis-
advantages of synthetic materials, such as polymers or
calcium phosphate ceramics, are related to their limited
osteoinductivity, mechanical strength and unpredictable
dissolution rates [2]. Moreover, these solutions do not
replicate local tissue architecture, nor at the macroscopic
and microscopic levels. The next generation of implantable
bone substitutes should be biocompatible, autologous,
mimic the shape of the implantation site, be rapidly inos-
culated to the host vasculature and replicate the bone
architecture [3]. Tissue engineering strategies, based on a
combination of biomaterials, cells and growth factors, have
been proposed as promising alternatives in order to solve
issues associated with current therapies and produce vas-
cularized bone substitutes that mimic both structure and
function of native bone [4].
Bone is a complex tissue with specific organization at
each scale. At the cellular scale, bone is considered as a
multicellular unit and a dynamic structure based on a close
collaboration between vascular, osteoblastic, osteoclastic
and neural cells [5]. On tissue scale, the macroscopic
architecture of bone is different from the periphery (cortical/
compact bone) to the center (cancellous/trabecular bone)
[6]. Although some differences exist between cortical and
cancellous layers, they both contain a highly vascularized
network.
Despite significant progress in tissue engineering,
reconstruction of large tissues remains limited by the lack of
a proper vascularization of the construct. Indeed, nutrients
and oxygen diffusion is increasingly difficult as the con-
struct becomes thicker. This is particularly true in recon-
struction of bone tissue since the extracellular matrix is
intended to be very dense. Moreover, development of
microvasculature and microcirculation is a critical step for
bone regeneration after surgery [7]. Implantation of large
grafts without microvascularization has proven to be
insufficiently perfused and resulted in graft failure [8].
Development of in vitro cellular models, combining
progenitor cells and biomaterials, could allow to overcome
this limitation and promote bone regeneration [9].
Endothelial cells seeded into a biodegradable biomaterial
such as collagen or fibrin, in presence of mesenchymal
cells, are able to generate a microvascular network. This
in vitro formed network is able to haste the inosculation
process and the integration to host tissue. Coculture of
endothelial cells with osteoblasts can generate capillary-like
structures in large tissue-engineered constructs [10,11].
Additionally, interaction between osteoblastic and endo-
thelial cells is beneficial for bone formation [12]. This
cooperation promoting vascularization in tissue-engineered
bone constructs was demonstrated in vivo by highlighting
the presence of perfused microvessels connected to the
vascular supply of the host [13]. Coculture-based pre-
vascularization involving endothelial and osteoblast cells
has shown to stimulate anastomosis between pre-formed
and host vasculatures, and to promote the growth of host
capillaries into the implanted construct [14]. Moreover, it
has been suggested that coculture of endothelial cells and
osteoblasts could promote prevascularization and osteoin-
tegration of the substitute [15].
A recent technology that boosted the growing field of
tissue engineering has been the introduction of cell bioprint-
ing as a relevant method to meet the need for precise orga-
nization of biological components [16]. Combined with other
additive manufacturing approaches, cell bioprinting will allow
to reproduce the architecture of tissues at both micro and
macro scales. Therefore, as each component of a tissue pre-
sents a distinct organization, bioprinting could be a powerful
tool to engineer each part relative to one another. Especially,
one of the most important components of bone, namely a
microvascular network, could be arranged using laser-assisted
bioprinting technology leading to a tissue-specificvascular-
ization. Indeed, in parallel with extrusion-based and ink-jet
printing, LAB has emerged as a promising approach and an
alternative method in the assembly and micropatterning of
biomaterials and cells [17]. LAB setup is based on a laser
beam focalized on a donor substrate (comprising a laser-
absorbing layer) and a receiving substrate. During the process,
the absorbing layer is vaporized by the laser pulse, resulting in
a jet of bioink that is propelled on the receiving substrate.
Based on laser-induced forward transfer (LIFT) [18], this non-
contact method does not suffer from some of the drawbacks
of other bioprinting technologies such as head clogging, high
shear stress, limited cell density and low resolution. More-
over, LAB advantages include high cell density, high cell
viability and high resolution [19]. LAB enables printing of
biomaterials such as hydrogels [20] as well as biological
components such as peptides, DNA and cells [21–23]. It has
been used to guide bone regeneration by in situ bioprinting
[24] and to print endothelial cells in order to create a micro-
vascular network in Matrigel®[25].
28 Page 2 of 12 Journal of Materials Science: Materials in Medicine (2019) 30:28
In the present study, endothelial cells were printed by
LAB following a specific pattern onto a collagen hydrogel
previously seeded with mesenchymal stem cells, in order
to generate a microvascular network with a defined
architecture. First, the concentration of endothelial cells in
the bioink was optimized to favor formation of a micro-
vascular network in the desired pattern. Second, a strategy
consisting of overlaying the printed biopaper with col-
lagen containing VEGF was developed in order to opti-
mize network formation and stability. Finally three
patterns of this optimized cell bioink, adapted to mouse
calvaria defect model, were tested in order to be used for
future in vivo applications.
2 Materials and methods
2.1 TdTomato-labeled endothelial progenitor cells
Human endothelial progenitor cells (EPCs) were isolated
from cord blood of healthy newborns from the Etablisse-
ment Français du Sang Aquitaine Limousin (EFS AL), as
previously described [26]. All protocols follow the tenets of
the declaration of Helsinki and mothers provided written
informed consent to have their samples be used for research.
Ministerial approval and authorizations to collect samples
from hospital were obtained for this study (DC-2008-412;
Convention EFSAL-CHU de Bordeaux). Samples were
treated anonymously. EPCs were transduced with a lenti-
viral vector containing the tdTomato protein gene (red
fluorescent protein) under the control of the phosphogly-
cerate kinase (PGK) promoter as previously described [27].
TdTomato-labeled EPCs were cultured in endothelial cell
growth medium-2 (EGM-2 MV, Lonza, Walkersville, MD,
USA), in a controlled atmosphere (5% CO2, 95% relative
humidity (RH), 37 °C–Incubator (Binder, Germany)). Pas-
sage 6 to 9 cells were used for the bioprinting experiments.
2.2 Stem cells from the apical papilla
Stem cells from the apical papilla (SCAPs) were isolated
from germs of third molars, obtained from young patients at
the Service de Chirurgie Buccale du Centre Hospitalier
Universitaire de Bordeaux (DC-2008-412; Convention
INSERM-CHU de Bordeaux). Patient’s oral informed
consent to have their samples be used for research purposes
was obtained. Samples were treated anonymously.
CD105+-selected cells were characterized by
fluorescent-activated cell sorting for the expression of
mesenchymal stem cell markers (CD90, CD73, CD45,
CD146), as described previously [28]. Green Fluorescent
Protein-labeled SCAPs were used for visualization of spa-
tial distributions of cells into the constructs.
SCAPs were cultured in polystyrene tissue culture flasks
in Minimum Essential Medium alpha (α-MEM, Gibco,
Paisley, Scotland, UK) supplemented with 10% fetal bovine
serum (FBS, GE Healthcare, Pasching, Austria), in a con-
trolled atmosphere (5% CO2, 95% RH, 37 °C).
2.3 Laser-assisted bioprinting set-up
2.3.1 Laser-assisted bioprinting workstation and printing
parameters
The LAB workstation used in this study was previously
described [17]. Briefly, the laser beam was focused on a
quartz ribbon that was coated with a thin absorbing layer of
gold (60 nm) and a 30 µm layer of cell bioink (donor slide)
(Fig. 1a). The transfer process was performed in air, at room
temperature, with a distance of 1000 µm between the ribbon
and the substrate. All experiments were performed using a
repetition rate of 1 kHz. Various printing parameters (bioink
concentration, laser energy) were tested in order to optimize
the formation of organized vascular network.
2.3.2 Substrate preparation and collagen overlay
Rat tail collagen type I (Collagen High concentration;
Corning, Bedford, MA, USA) was diluted in DMEM to a
final concentration of 2 mg/mL. 141 µL of this collagen
mixture was spread at 4 °C onto the quartz substrate and
was allowed to solidify for 1 h in controlled atmosphere
(5% CO2, 95% RH, 37 °C) prior to perform printing
experiments. For the preparation of cell-seeded biopapers,
SCAPs were seeded on solidified collagen with a density of
35.000 cells/cm21 h post gelation and 24 h before printing.
Overlay of a second layer of collagen hydrogel over the
printed patterns was performed using 141 µL of 2 mg/ml
collagen, with or without addition of VEGF at 20 ng/mL
(VEGF-165, human recombinant, Promokine, Heidelberg,
Germany). SCAPs were seeded on this second layer of col-
lagen 30 min post gelation with a density of 35,000 cells/cm2
(Fig. 1b). The same procedure was repeated two times to
obtain multilayered constructs.
2.3.3 Bioink preparation and printing parameters
TdTomato-labeled EPCs were detached from the poly-
styrene tissue culture surface with a solution of trypsin.
Cells were suspended in EGM-2 MV. Depending on the
experiment, this cell suspension was concentrated from 20
to 70 million cells/mL. A 3 cm diameter quartz slide
(Société VM, Epinal, France) was coated with a thin
absorbing layer of gold (60 nm) using a sputter coater
(EMSCOPE SC500, Elexience). Then, 30 µm-thick film of
cell bioink (33 µL) was manually spread on the ribbon
Journal of Materials Science: Materials in Medicine (2019) 30:28 Page 3 of 12 28
surface. The laser focused on the gold layer induces the
transfer of droplets of the endothelial cell bioink to the
cellularized biopaper according to the selected pattern
(Fig. 1c). In the present study, cell transfer follows a spe-
cific pattern of droplets spaced from 60 µm between spots,
with 1250 µm between each line of spots.
2.4 Coculture tdTomato-labeled EPCs/SCAPs
The printed samples were cultured in a coculture med-
ium, consisting of equal proportions of the culture media
of each cell type (α-MEM, EGM2-MV). Antibiotics
(1:100; Pen/Strep, penicillin/streptomycin, Gibco, Grand
Island, New York, USA) were used for cell culture post-
printing.
2.5 Post-printing characterization and image
analysis
Accuracy of the patterns, their preservation and their evo-
lution at different times post-printing were analyzed by
fluorescence microscopy (Leica MZ10 F, Leica Micro-
systems Ltd, Heerbrugg, Switzerland). Cell migration was
visualized using the LumaScope 600TM (Etaluma Inc.,
Carlsbad, CA, USA). Cell Counter plugin of image pro-
cessing software ImageJ (Open source, Public domain) was
used to calculate cell density in patterns post-printing within
defined areas. The microvascular network was characterised
at day 6 using Angiogenesis Analyser ImageJ’s plugin [29].
The extend of the network was estimated based on the
number of cell Segments.
2.6 Immunofluorescence
Network organization was assessed by in situ immuno-
fluorescence of cluster of differentiation (CD) 31. After
fixation in 4% paraformaldehyde for 20 min at 4 °C, the
mouse anti-human CD31 antibody (1:300; eBioscience,
San Diego, CA, USA) was added for 1 h at RT. Secondary
antibody Alexa 488-conjugated goat anti-mouse (1:400;
Invitrogen-Life Technologies, Carlsbad, CA, USA) was
then added for 1 h at RT. Endothelial cells were also
stained using the lectin Ulex Europeaeus agglutinin I FITC
(UEA-I, 20 µg/ml; Vector Laboratories, Eurobio ABCys,
France). UEA-I was added overnight at 4 °C. Nuclei were
counterstained using DAPI (1:5000; Sigma-Aldrich, St-
Louis, MO, USA). After mounting, slides were observed
by confocal microscopy (Leica TCS SPE, DMI 4000B,
Mannheim, Germany). Three-dimensional configuration of
constructs and spatial distributions of cells were analyzed
using image analysis software (IMARIS, Bitplane,
Switzerland).
Fig. 1 Experimental set-up and
microscopy images of SCAPs
and printed endothelial cells: a
Schematic representation of the
material used for LAB
procedure, bSchematic
representation of the
experimental set-up for in vitro
creation of vascular network by
LAB, cPhase contrast (up) and
fluorescence (down) images of
tdTomato-labeled endothelial
cells printed in continuous line
on collagen/SCAPs biopaper 15
minutes after printing
28 Page 4 of 12 Journal of Materials Science: Materials in Medicine (2019) 30:28
2.7 Statistical analysis
Statistical analyses were performed using GraphPad Prism
software (GraphPad, San Diego, CA, USA) by two-way
ANOVA followed by Bonferroni’s multiple comparison
test. Differences were considered to be significant with p<
0.05.
3 Results
3.1 Cell density of the bioink influences endothelial
cell density of patterns after printing
Correlation between bioink cell concentration and printed
cell density was first investigated. For that purpose, the
endothelial cell bioink was printed at different concentra-
tions (20, 35, 50 and 70 million cells/mL) and energies
(respectively 22, 23, 24 and 26 µJ). When observed
immediately after bioprinting, the cell density of printed
patterns increases with the initial concentration of bioink
(Fig. 2). Bioink concentration of 70 million cells/mL and
laser energy of 26 µJ were chosen as the optimal printing
parameters since they lead to the significantly highest cell
density (2.2 ± 0.6 × 103cells/mm2) (Fig. 2b).
3.2 High cell density patterns favor network
formation
Correlation between the local density of the endothelial
cells observed 1 h post-printing and the formation of a
capillary-like network revealed a direct relationship
Fig. 2 Determination of optimal printing parameters of endothelial
cells for endothelial cell patterns creation: aComparison of patterns of
tdTomato-labeled EPCs after printing with different initial concentra-
tions of cell bioink and different laser energies, bCell density of
printed patterns depending on initial cell bioink concentration (*p<
0.05,**p< 0.01,***p< 0.001); Images are representative of n=10
experiments
Journal of Materials Science: Materials in Medicine (2019) 30:28 Page 5 of 12 28
between the two parameters (Fig. 3a). Patterns with an
initial cell density of 440 ± 200 (45.5%) cells/mm2did not
allow formation of network at day 6. Patterns with initial
cell density of 1447 ± 321 (22.2%) cells/mm2led to poorly
interconnected and unstable networks at D6. Patterns with
initial cell density of 2176 ± 556 (25.6%) cells/mm2led to
highly interconnected and stable networks at D6 (Fig. 3b).
In order to define more precisely the relationship between
initial cell density of patterns and the extend of the formed
network, the length of the cell network was quantified using
Angiogenesis Analyser ImageJ’s plugin. The equation of
this relation was y =0.4523 × –12.87 and the coefficient of
determination (r2) was 0.80 (Fig. 3c).
3.3 Overlay with VEGF/SCAPs-containing collagen
allows to maintain a network of endothelial
cells
In order to increase network formation and stability, overlay
conditions of the printed patterns with a second layer of
hydrogel were optimized. Without overlay, patterns stay
organized over the time of the experiment. However, the
formed network disappeared at D6 (Fig. 4a). Patterns cov-
ered with collagen alone (without VEGF and SCAPs)
immediately or 24 h post-printing maintained pattern orga-
nization. Cells stay mostly round in both conditions and
fluorescence was gradually loss through day 6, indicative of
cell death (Fig. 4b, c). Endothelial cells covered with col-
lagen containing VEGF and SCAPs immediately or 24 h
post-printing maintained their organization over time and
led to well-organized networks consistent with the initial
pattern. However, endothelial cells still tried to elongate
laterally toward the adjacent lines (Fig. 4d, e). Immediately
after addition of collagen-VEGF and SCAPs, changes in
endothelial cell organization and formation of network were
assessed by time-lapse, as shown in Online Resource 1. The
condition consisting in overlaying patterns with collagen/
VEGF and SCAPS immediately post printing was selected
for further experiments to fit in vivo procedures.
Confocal microscopy on printed endothelial cells on
SCAPs at day 6 stained for the cell junction protein CD31,
the lectin UEA-I and nuclear counterstaining demonstrated
that multicellular network of endothelial cells is forming
(Fig. 5a, b). The organization of cell constructs in three
Fig. 3 Determination of optimal cell densities promoting the formation
of networks at day 6: aObservation of network formation depending
on the initial density of patterns of tdTomato-labeled EPCs after
printing, bCell density of printed patterns in correlation with the
generation and stability of network (***p< 0.001), cRelation between
cell density of printed patterns and the total length of cell segments at
day 6; Images are representative of n=6 experiments
28 Page 6 of 12 Journal of Materials Science: Materials in Medicine (2019) 30:28
Fig. 4 Pattern follow-up and formation of networks depending on
overlay conditions: aNo overlay, bOverlay immediately post-printing
with collagen hydrogel, cOverlay 24 h post-printing with collagen
hydrogel, dOverlay immediately post-printing with collagen/VEGF
hydrogel and SCAPs, eOverlay 24 h post-printing with collagen/
VEGF hydrogel and SCAPs. Endothelial patterns were followed at day
0, day 3 and day 6; Images are representative of n=4 experiments
Fig. 5 Characterization of
endothelial structures in
coculture by confocal
microscopy at day 6: aBranched
network of endothelial cells
observed at day 6 after CD31
(green) and DAPI (blue)
staining, bOrthogonal views of
endothelial structures after
UEA-I FITC and DAPI staining
at day 6, cIdentification of the
three cellular layers of the
constructs (Z-stack), delimited
by white dashed lines (in green,
CD31 immunolabeling; in blue,
DAPI), dSpatial distribution of
cells in a three-layered construct
at day 6 (in green, GFP-labeled
SCAPs; in red, tdTomato-
labeled EPCs), ePattern follow-
up and formation of networks in
afive-layered construct at day 1,
day 3 and day 6, fIdentification
of the five cellular layers of the
constructs (Z-stack) (in green,
GFP-labeled SCAPs; in red,
tdTomato-labeled EPCs), g
Spatial distribution of cells in a
five-layered construct at day 6
(in green, GFP-labeled SCAPs;
in red, tdTomato-labeled EPCs);
Images are representative of n=
6 experiments
Journal of Materials Science: Materials in Medicine (2019) 30:28 Page 7 of 12 28
cellular layers was assessed by confocal microscopy
(Fig. 5c). Three-dimensional reconstruction images showed
spatial distribution of cells at day 6 (Fig. 5d).
The generated networks were highly interconnected and
stable in vitro. They were preserved until at least D20.
In order to assess the validity of our model into multi-
layered constructs, endothelial cells were printed perpen-
dicular to the first printed pattern onto the second layer of
biopaper (VEGF/SCAPs-containing collagen). Immediately
after bioprinting, the new printed pattern was recovered by a
third layer of biopaper. The kinetics of network formation
remained similar to that seen previously (Fig. 5e). The
organization of cell constructs in five layers was assessed by
confocal microscopy (Fig. 5f) and three-dimensional
reconstruction images (Fig. 5g).
3.4 Generation of models for in vivo applications
To validate that printing conditions determined previously
have enough precision to print endothelial cells into a small
calvaria bone defect in mice, three patterns of endothelial
cells were printed within 3 mm (Fig. 6a). When observed
under binocular epifluorescence microscope at day 1 and 3,
the three different patterns tested were partially preserved at
day 3. As expected for the filled circle (several dots) the
printed area was randomly covered by a network of endo-
thelial cells. For the two other patterns (double circles and
double circles with a line), the gap between the two lines
was filled out but the central section stays free of endothelial
cells (Fig. 6b).
4 Discussion
In the present study, LAB technology was used to precisely
position endothelial cells in controlled density in order to
form a microvascular network with a defined architecture
into a collagen hydrogel containing mesenchymal cells. We
demonstrated that a high concentration of EPCs is necessary
to promote the formation of a microvascular network and
that the printed EPCs must be immediately covered with
collagen containing VEGF and SCAPs to maintain this
organization. LAB accuracy allowed the creation of precise
patterns adapted to critical-size calvarial bone defects of 3
mm in murine model.
Multiple micro- and nano-patterning technologies have
been developed to engineer cell alignment in vitro [30].
Laser-assisted bioprinting is a relevant tool enabling precise
cell patterning with high cell densities, promoting cell
cooperation and vascularization. LAB is a non-contact
technology enabling rapid printing of patterns at micrometric
resolution, with low cell stress and mortality [31]. Giving
those advantages, this technology is interesting to control the
cell organization and to study cellular interactions [32].
Fig. 6 Patterns for in vivo printing into a mice calvaria bone defect: aSchematic representation of endothelial cell patterns for in vivo application,
bObservation of patterns of tdTomato-labeled endothelial cells at day 1 and day 3; Images are representative of n=5 experiments
28 Page 8 of 12 Journal of Materials Science: Materials in Medicine (2019) 30:28
The parameters of the laser, such as laser energy, need to
be adapted according to the rheology of the bioink that
varies depending on cell type and concentration. The
composition of the bioink modifies its properties. In parti-
cular, the addition of cells influences the viscoelastic
behavior of the bioink before and during the printing pro-
cess [33]. For this study, adjusting laser parameters and
culture conditions has been necessary to optimize the for-
mation of networks. The control of energy and bioink
concentration allowed to vary cell density within the pat-
terns for the promotion of cell-cell interactions. Initial cell
density of printed patterns and length of the network were
found to be correlated. An initial cell density of 2176 ± 556
cells/mm2in printed patterns led to the formation of highly
interconnected and stable networks. Also we determined
that this optimal cell density was obtained with a bioink
concentration of 70 million cells/mL and laser energy of 26
µJ. Contrary to other studies that address the concept of cell
culture ratio in coculture to favor the formation of vascular
networks, our work allowed to define precisely the local
density of endothelial cells needed for network formation
thanks to LAB reproducibility and accuracy.
Endothelial cells viability post-printing was not investi-
gated here, since formation of endothelial networks and
tdTomato fluorescence suggested that cell viability was
preserved over time. Moreover, many studies have already
demonstrated the excellent capacities of laser-based tech-
nologies to preserve cell viability, proliferation and differ-
entiation after printing [34,35].
Our model is based on a coculture of endothelial cells
with mesenchymal stem cells. Coculture is essential to give
rise to endothelial networks and preserve the printed pat-
tern. Previous data from our laboratory showed the influ-
ence of coculture on the migration of endothelial cells and
formation of self-assembled networks [36]. Human umbi-
lical vein endothelial cells (HUVECs) have been demon-
strated to migrate along osteoblastic progenitors inducing
the formation of tubular-like structures through a direct
contact between cells [37,38]. Additional experiments
showed that printed endothelial cells in monoculture did not
give rise to endothelial networks despite overlay with
collagen-VEGF (Online Resource 2). After 72 h, tdTomato
EPCs printed with an interline distance of 1250 μm had
spread on the entire surface of collagen hydrogel. At day 6,
no network was formed and the pattern was difficult to
discern. This disorganization of printed pattern in mono-
culture is consistent with previous data from our laboratory.
Bourget et al. showed that mesenchymal cells are critical to
preserve pattern over time and guide endothelial cells self-
organization [32]. In this study, SCAPs allowed to maintain
pattern and local density of endothelial cells, favoring the
formation of networks. Beyond its participation to network
architecture, our results showed that cocultures of SCAPs/
EPCs increase the secretion of basement membrane com-
ponents (collagen IV and laminin) at day 6 (Online
Resource 3).
In order to improve network formation and stability, we
proposed here to overlay the endothelial cell network with a
second layer of collagen hydrogel. Collagen type I is the
most commonly used material in cell and tissue culture [39].
This soft material preserves cell viability after printing by
minimizing injuries during landing. Addition of VEGF to
the second layer of hydrogel was found to be necessary to
allow the formation of a network, even with mesenchymal
cells. The culture medium used already contained some
VEGF, overlay with the second layer of collagen probably
limited the accessibility of this growth factor to endothelial
cells.
Wu and Ringeisen used a LAB setup to produce branch/
stem structures (similar to the vein structure on a leaf) of
HUVECs in Matrigel®, stabilized by adding human umbi-
lical vascular smooth muscle cells on HUVECs [25]. It is
consistent with the literature indicating that the assembly of
endothelial cells into capillaries necessitates smooth muscle
cells, pericytes or mesenchymal cells [10]. Our study was
necessary to develop a protocol without using Matrigel®,
which will be problematic in a clinical setting. Matrigel®is
a natural extracellular matrix derived from Englebreth-
Holm-Swarm tumors produced in mice [40]. This material
is not the best candidate for experiments with a clinical
perspective due to its unknown composition and growth
factor content [41,42]. Moreover, Matrigel®is not “U.S.
Food and Drug Administration”approved. On the contrary,
VEGF has been widely used for clinical trials and its
association with collagen type I seems to represent a safer
alternative [43,44]. Regarding cell types used, since our
study aims at the formation of vascularized bone substitutes,
we used SCAPs, that are known to have a potential of
osteoblastic differentiation [45] and EPCs, that have been
previously reported to form tubular structures when cocul-
tured with osteoblasts [27,46]. SCAPs have the advantage
to be easily accessible since they can be isolated from
human wisdom teeth. These cells seem to be relevant can-
didates for bone regeneration, especially to treat defects in
the orofacial sphere which is among our future goals.
Confocal observations of the printed endothelial cells
show organisation of cells into a branched network. How-
ever, formation of lumen needs to be improved. Some key
regulators should be included in order to optimize our
model by enhancing lumen formation, tube maturation and
stabilization [47]. In particular, some studies demonstrated
that endothelial cell tubulogenesis requires five growth
factors: stem cell factor, interleukin-3, stromal-derived
factor-1α,fibroblast growth factor-2 and insulin. More-
over, mural cells such as pericytes and vascular smooth
muscle cells have a key role in vascular morphogenesis and
Journal of Materials Science: Materials in Medicine (2019) 30:28 Page 9 of 12 28
could be introduced in our model [48]. Nonetheless, the aim
of our article was to determine optimal parameters to obtain
organized vascular networks, in order to transpose this
model to in vivo experiments. In this context of in vivo
bioprinting, surrounding factors produced by the host
organism are privileged to favor lumen formation.
LAB allowed to elaborate multilayered constructs by
interposing collagen seeded with SCAPs between endo-
thelial printed patterns. SCAPs-seeded collagen was not
printed by LAB here since our study was centered around
precise organization of endothelial cells and vascular
structures only. In the long term, it is planned to automate
the process by combining the LAB with other bioprinting
technologies, such as extrusion or inkjet. These technolo-
gies are more suitable for printing hydrogels and creation of
three-dimensional architectures inside which different cell
types can be precisely printed by LAB according to their
respective location in native tissues. In this context, opti-
mization of biopaper properties is under consideration in
order to reduce contraction issues related to collagen scaf-
fold. Previous results showed that thickness of our con-
structs critically decreased over time (Online Resource 4).
Several alternatives are currently explored such as self-
assembled osseous sheets based on cell-derived bone
extracellular matrix [49].
Future work will concern in vivo bioprinting of EPCs,
using the bioink concentration, culture conditions and laser
parameters determined in the present study, to assess the
effect of prevascularization with an organized vascular
network on bone healing in vivo. A major advantage of
LAB is the short printing time (only few seconds) allowing
easy transposition to in vivo experiments. Previous work
from our group has demonstrated the ability to use the LAB
setup in vivo. Nano-hydroxyapatite (n-HA) was deposited
into mouse calvaria defect of critical size [24,50]. Bio-
printing technologies are relevant methods that allow to
precisely pattern biological components and to study effects
of patterning on tissue regeneration. Cooper et al. used
inkjet bioprinting technology to spatially control osteoblast
differentiation in vitro and guide bone formation in vivo by
creating patterns of bone morphogenetic protein-2 (BMP-2)
on a human dermal allograft scaffold [51]. The aim of our
future in vivo experiments will be to investigate if pat-
terning endothelial cells, leading to vascular network with
defined architecture, can promote and spatially control
calvarial bone formation. For that purpose, three different
geometries of endothelial patterns were designed: i/“filled
circle”to evaluate the osteoinductive potential of printed
cells; ii/“double circles”to evaluate the osteoconductive
potential; iii/“double circles with a line”to assess the
optimal distances between two patterns to have an effect on
the regenerative process. These organized microvascular
structures could guide angiogenesis for bone healing by
connecting to surrounding vasculature at the periphery of
the defect. Bone formation could occur in a particular
geometry, consistent with the initial printed pattern.
5 Conclusion
Laser-assisted bioprinting technology allows to preorganize
endothelial cells in order to generate microvascular net-
works with defined architecture. A high local EPCs density,
obtained by controlling the initial cell bioink concentration,
is necessary to generate a stable microvascular network and
addition of VEGF to the collagen used to cover the EPCs is
necessary to preserve the network, even in the presence of
mesenchymal cells. The printing parameters determined
during these in vitro experiments will be applied for future
in situ bioprinting applications. This study is a step in the
improvement of vascularization of bone substitutes as well
as in the promotion of bone regeneration and has important
prospects in the field of biofabrication for regenerative
medicine. This technology will allow generation of specific
vasculature organization for larger volume constructs.
Acknowledgements This work was supported by the Institut français
pour la recherche odontologique (IFRO) and Bordeaux Consortium for
Regenerative Medicine (BxCRM). The authors acknowledge «Fon-
dation des Gueules Cassées, Paris-France » (n°54-2017) and « Fon-
dation de l’Avenir, Paris-France» (N°AP-RM-17-038) for their
financial support. The authors would also like to thank Sophia Ziane
(INSERM U1026, Bordeaux, France), Nathalie Dusserre and Davit
Hakobyan (ART Bioprint, INSERM U1026, Bordeaux, France) and
Sébastien Marais (Bordeaux Imaging Center, Bordeaux, France) for
their excellent technical support.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
Publisher’s note: Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
References
1. Van der Stok J, Van Lieshout EMM, El-Massoudi Y, Van Kra-
lingen GH, Patka P. Bone substitutes in the Netherlands-a sys-
tematic literature review. Acta Biomater. 2011;7:739–50.
2. García-Gareta E, Coathup MJ, Blunn GW. Osteoinduction of bone
grafting materials for bone repair and regeneration. Bone.
2015;81:112–21.
3. Drosse I, Volkmer E, Capanna R, De Biase P, Mutschler W,
Schieker M. Tissue engineering for bone defect healing: an update
on a multi-component approach. Injury. 2008;39(Suppl 2):S9–20.
4. Black CRM, Goriainov V, Gibbs D, Kanczler J, Tare RS, Oreffo
ROC. Bone tissue engineering. CurrMol Biol Rep. 2015;1:132–40.
5. Florencio-Silva R, Sasso GR, da S, Sasso-Cerri E, Simões MJ,
Cerri PS. Biology of bone tissue: structure, function, and factors
that influence bone cells. Biomed Res Int. 2015;2015:421746
28 Page 10 of 12 Journal of Materials Science: Materials in Medicine (2019) 30:28
6. Samuel SP, Baran GR, Wei Y, Davis BL. Biomechanics-Part II.
In: Khurana JS, editor. Bone Pathology. Totowa: Humana Press;
2009. p. 69–77.
7. Santos MI, Reis RL. Vascularization in bone tissue engineering:
physiology, current strategies, major hurdles and future chal-
lenges. Macromol Biosci. 2009;10:12–27.
8. Mercado-Pagán ÁE, Stahl AM, Shanjani Y, Yang Y. Ann Biomed
Eng.2015;43:718–29.
9. Kirkpatrick CJ. Developing cellular systems in vitro to simulate
regeneration. Tissue Eng Part A. 2014;20:1355–7.
10. Liu Y, Lu J, Li H, Wei J, Li X. Engineering blood vessels through
micropatterned co-culture of vascular endothelial and smooth
muscle cells on bilayered electrospun fibrous mats with pDNA
inoculation. Acta Biomater. 2015;11:114–25.
11. Liu Y, Teoh S-H, Chong MSK, Lee ESM, Mattar CNZ,
Randhawa NK, et al. Vasculogenic and osteogenesis-enhancing
potential of human umbilical cord blood endothelial colony-
forming cells. Stem Cells. 2012;30:1911–24.
12. Pirraco RP, Iwata T, Yoshida T, Marques AP, Yamato M, Reis
RL, et al. Endothelial cells enhance the in vivo bone-forming
ability of osteogenic cell sheets. Lab Investig J Tech Methods
Pathol. 2014;94:663–73.
13. Fuchs S, Ghanaati S, Orth C, Barbeck M, Kolbe M, Hofmann A,
et al. Contribution of outgrowth endothelial cells from human
peripheral blood on in vivo vascularization of bone tissue engi-
neered constructs based on starch polycaprolactone scaffolds.
Biomaterials. 2009;30:526–34.
14. Unger RE, Ghanaati S, Orth C, Sartoris A, Barbeck M, Halstenberg
S, et al. The rapid anastomosis between prevascularized networks on
silk fibroin scaffolds generated in vitro with cocultures of human
microvascular endothelial and osteoblast cells and the host vascu-
lature. Biomaterials. 2010;31:6959–67.
15. Unger RE, Dohle E, Kirkpatrick CJ. Improving vascularization of
engineered bone through the generation of pro-angiogenic effects
in co-culture systems. Adv Drug Deliv Rev. 2015;94:116–25.
16. Mironov V, Trusk T, Kasyanov V, Little S, Swaja R, Markwald
R. Biofabrication: a 21st century manufacturing paradigm. Bio-
fabrication. 2009;1:022001.
17. Guillemot F, Souquet A, Catros S, Guillotin B, Lopez J, Faucon
M, et al. High-throughput laser printing of cells and biomaterials
for tissue engineering. Acta Biomater. 2010;6:2494–500.
18. Bohandy J, Kim BF, Adrian FJ. Metal deposition from a sup-
ported metal film using an excimer laser. J Appl Phys.
1986;60:1538–9.
19. Schiele NR, Corr DT, Huang Y, Raof NA, Xie Y, Chrisey DB.
Laser-based direct-write techniques for cell printing. Biofabrica-
tion. 2010;2:032001.
20. Unger C, Gruene M, Koch L, Koch J, Chichkov BN. Time-
resolved imaging of hydrogel printing via laser-induced forward
transfer. Appl Phys A. 2011;103:271–7.
21. Colina M, Serra P, Fernández-Pradas JM, Sevilla L, Morenza JL.
DNA deposition through laser induced forward transfer. Biosens
Bioelectron. 2005;20:1638–42.
22. Dinca V, Kasotakis E, Catherine J, Mourka A, Mitraki A, Popescu
A, et al. Development of peptide-based patterns by laser transfer.
Appl Surf Sci. 2007;254:1160–3.
23. Hopp B, Smausz T, Papdi B, Bor Z, Szabó A, Kolozsvári L, et al.
Laser-based techniques for living cell pattern formation. Appl
Phys Mater Sci Process. 2008;93:45–9.
24. Keriquel V, Guillemot F, Arnault I, Guillotin B, Miraux S,
Amédée J, et al. In vivo bioprinting for computer- and robotic-
assisted medical intervention: preliminary study in mice. Biofab-
rication. 2010;2:014101.
25. Wu PK, Ringeisen BR. Development of human umbilical vein
endothelial cell (HUVEC) and human umbilical vein smooth
muscle cell (HUVSMC) branch/stem structures on hydrogel layers
via biological laser printing (BioLP). Biofabrication.
2010;2:014111.
26. Thébaud NB, Bareille R, Daculsi R, Bourget C, Rémy M, Ker-
djoudj H, et al. Polyelectrolyte multilayer films allow seeded
human progenitor-derived endothelial cells to remain functional
under shear stress in vitro. Acta Biomater. 2010;6:1437–45.
27. Thébaud NB, Aussel A, Siadous R, Toutain J, Bareille R, Mon-
tembault A, et al. Labeling and qualification of endothelial pro-
genitor cells for tracking in tissue engineering: An in vitro study.
Int J Artif Organs. 2015;38:224–32.
28. Devillard R, Rémy M, Kalisky J, Bourget J-M, Kérourédan O,
Siadous R, et al. In vitro assessment of a collagen/alginate com-
posite scaffold for regenerative endodontics. Int Endod J.
2017;50:48–57.
29. Carpentier G. ImageJ contribution: Angiogenesis Analyzer. 2012.
Disponible sur: https://imagej.nih.gov/ij/macros/toolsets/
Angiogenesis%20Analyzer.txt.
30. Li Y, Huang G, Zhang X, Wang L, Du Y, Lu TJ, et al. Engi-
neering cell alignment in vitro. Biotechnol Adv. 2014;32:347–65.
31. Devillard R, Pagès E, Correa MM, Kériquel V, Rémy M, Kalisky
J, et al. Cell patterning by laser-assisted bioprinting. Methods Cell
Biol. 2014;119:159–74.
32. Bourget J-M, Kérourédan O, Medina M, Rémy M, Thébaud NB,
Bareille R, et al. Patterning of endothelial cells and mesenchymal
stem cells by laser-assisted bioprinting to study cell migration.
Biomed Res Int. 2016;2016:3569843.
33. Zhang Z, Xu C, Xiong R, Chrisey DB, Huang Y. Effects of living
cells on the bioink printability during laser printing. Biomicro-
fluidics. 2017;11:034120.
34. Gruene M, Pflaum M, Deiwick A, Koch L, Schlie S, Unger C,
et al. Adipogenic differentiation of laser-printed 3D tissue grafts
consisting of human adipose-derived stem cells. Biofabrication.
2011;3:015005.
35. Catros S, Fricain J-C, Guillotin B, Pippenger B, Bareille R, Remy
M, et al. Laser-assisted bioprinting for creating on-demand pat-
terns of human osteoprogenitor cells and nano-hydroxyapatite.
Biofabrication. 2011;3:025001.
36. Li H, Daculsi R, Grellier M, Bareille R, Bourget C, Remy M, et al.
The role of vascular actors in two dimensional dialogue of human
bone marrow stromal cell and endothelial cell for inducing self-
assembled network. PLoS ONE. 2011;6:e16767.
37. Grellier M, Ferreira-Tojais N, Bourget C, Bareille R, Guillemot F,
Amédée J. Role of vascular endothelial growth factor in the
communication between human osteoprogenitors and endothelial
cells. J Cell Biochem. 2009;106:390–8.
38. Grellier M, Bordenave L, Amédée J. Cell-to-cell communication
between osteogenic and endothelial lineages: implications for
tissue engineering. Trends Biotechnol. 2009;27:562–71.
39. Skardal A, Atala A. Biomaterials for integration with 3-d bio-
printing. Ann Biomed Eng. 2015;43:730–46.
40. Nguyen EH, Daly WT, Le NNT, Farnoodian M, Belair DG,
Schwartz MP, et al. Versatile synthetic alternatives to Matrigel for
vascular toxicity screening and stem cell expansion. Nat Biomed
Eng. 2017;1.
41. Hughes CS, Postovit LM, Lajoie GA. Matrigel: a complex protein
mixture required for optimal growth of cell culture. Proteomics.
2010;10:1886–90.
42. Vukicevic S, Kleinman HK, Luyten FP, Roberts AB, Roche NS,
Reddi AH. Identification of multiple active growth factors in
basement membrane Matrigel suggests caution in interpretation of
cellular activity related to extracellular matrix components. Exp
Cell Res. 1992;202:1–8.
43. Hendel RC, Henry TD, Rocha-Singh K, Isner JM, Kereiakes DJ,
Giordano FJ, et al. Effect of intracoronary recombinant human
vascular endothelial growth factor on myocardial perfusion: evi-
dence for a dose-dependent effect. Circulation. 2000;101:118–21.
Journal of Materials Science: Materials in Medicine (2019) 30:28 Page 11 of 12 28
44. Henry null, Abraham null. Review of preclinical and clinical
results with vascular endothelial growth factors for therapeutic
angiogenesis. Curr Interv Cardiol Rep. 2000;2:228–41.
45. Wang L, Yan M, Wang Y, Lei G, Yu Y, Zhao C, et al. Pro-
liferation and osteo/odontoblastic differentiation of stem cells
from dental apical papilla in mineralization-inducing medium
containing additional KH(2)PO(4). Cell Prolif. 2013;46:214–22.
46. Thébaud NB, Siadous R, Bareille R, Remy M, Daculsi R, Amédée
J, et al. Whatever their differentiation status, human progenitor
derived-or mature-endothelial cells induce osteoblastic differ-
entiation of bone marrow stromal cells. J Tissue Eng Regen Med.
2012;6:e51–60.
47. Koh W, Stratman AN, Sacharidou A, Davis GE. In vitro three
dimensional collagen matrix models of endothelial lumen forma-
tion during vasculogenesis and angiogenesis. Methods Enzymol.
2008;443:83–101.
48. Bowers SLK, Norden PR, Davis GE. Molecular signaling path-
ways controlling vascular tube morphogenesis and pericyte-
induced tube maturation in 3D extracellular matrices. Adv Phar-
macol San Diego Calif. 2016;77:241–80.
49. Kawecki F, Clafshenkel WP, Auger FA, Bourget J-M, Fradette J,
Devillard R. Self-assembled human osseous cell sheets as living
biopapers for the laser-assisted bioprinting of human endothelial
cells. Biofabrication. 2018;10:035006.
50. Keriquel V, Oliveira H, Rémy M, Ziane S, Delmond S, Rousseau
B, et al. In situ printing of mesenchymal stromal cells, by laser-
assisted bioprinting, for in vivo bone regeneration applications.
Sci Rep. 2017;7:1778.
51. Cooper GM, Miller ED, Decesare GE, Usas A, Lensie EL,
Bykowski MR, et al. Inkjet-based biopatterning of bone mor-
phogenetic protein-2 to spatially control calvarial bone formation.
Tissue Eng Part A. 2010;16:1749–59.
28 Page 12 of 12 Journal of Materials Science: Materials in Medicine (2019) 30:28
A preview of this full-text is provided by Springer Nature.
Content available from Journal of Materials Science: Materials in Medicine
This content is subject to copyright. Terms and conditions apply.
