Journal of Plastic, Reconstructive & Aesthetic Surgery (2018) 71, 101e111

Implant for autologous soft tissue

reconstruction using an adipose-derived
stem cell-colonized alginate scaffold
Tobias Hirsch a, Christine Laemmle a, Bjoern Behr a,
Marcus Lehnhardt a, Frank Jacobsen a, Dirk Hoefer b,*,
Maximilian Kueckelhaus a

a
Department of Plastic Surgery, BG University Hospital Bergmannsheil, Ruhr University Bochum,
Bochum, Germany
b
Department of Hygiene, Environment and Medicine, Hohenstein Institutes, Boennigheim, Germany

Received 31 January 2017; accepted 6 August 2017

KEYWORDS Summary Background: Adipose-derived stem cells represent an interesting option for soft
Adipose-derived stem tissue replacement as they are easy to procure and can generate their own blood supply
cells; through the production of angiogenic factors. We seeded adipose-derived stem cells on a bio-
ASCs; resorbable, biocompatible polymer alginate scaffold to generate autologous soft tissue con-
Scaffold; structs for repair.
Implant; Materials and methods: We built and optimized an alginate scaffold and tested its biocompat-
Alginate ibility using the MTT assay and its hydration capacity. We then isolated, characterized, and
differentiated murine, porcine, and human adipose-derived stem cells. We characterized their
angiogenic potential in vitro by VEGF ELISA and HUVEC tube formation assay in traditional cell
culture substrate and in the actual three-dimensional scaffold. We assessed the angiogenic po-
tential of adipose-derived stem cell-colonized scaffolds in ovo by chorion allantois membrane
angiogenesis assay.
Results: Adipose-derived stem cells differentiated into adipocytes within the alginate scaf-
folds and demonstrated angiogenic activity. VEGF secretion by adipose-derived stem cells
decreased significantly over the 21-day course of adipocyte differentiation in traditional cell
culture substrate, but not in scaffolds. Adipose-derived stem cells differentiated for 21 days
in scaffolds led to the longest HUVEC tube formation. Scaffolds colonized with adipose-
derived stem cells resulted in significantly improved vascularization in ovo.
Conclusions: We demonstrate the feasibility of implant production based on adipose-derived
stem cell-colonized alginate scaffolds. The implants demonstrate biocompatibility and

* Corresponding author. Department of Hygiene, Environment and Medicine, Hohenstein Institutes, Schlosssteige 1, 74357 Boennigheim,
Germany.
E-mail address: d.hoefer@hohenstein.de (D. Hoefer).

http://dx.doi.org/10.1016/j.bjps.2017.08.009
1748-6815/ª 2017 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
102
promote angiogenesis in vitro and in ovo. Therefore, they provide a combination of essential
properties for an implant intended for soft tissue replacement.
ª 2017 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Else-
vier Ltd. All rights reserved.
Introduction
The management of acute and chronic soft tissue defects
represents a complex challenge for the plastic surgeon,
especially in context of high prevalence of diseases such as
diabetes mellitus, chronic venous insufficiency, and pe-
ripheral artery occlusive disease or after radiation therapy.
These soft tissue defects potentially led to prolonged hos-
pitalization with protracted individual and economic bur-
den.1e4 Allogenic materials for soft tissue reconstruction
introduce the problem of insufficient vascularization,
leading to consecutive implant loss. Transfer of large vol-
umes of autologous adipose tissue results in central necrosis
and consecutive volume loss. In consequence, there is a
high demand for alternatives to conventional defect
coverage.5,6 The ideal material for soft tissue replacement
should not only replace the defect volume but also retain it
by promoting its successful integration with the surrounding
vascular system. Tissue engineering deals with the con-
struction of tissue-like biologic materials that can fulfill
these requirements.7,8
The aim of this study was to engineer an implant using a
three-dimensional biocompatible alginate scaffold colo-
nized with angiogenic adipose-derived stem cells and test
its potential for soft tissue replacement both in vitro and in
ovo. The biopolymer alginate is derived from algae or
produced by bacteria.9 Since its discovery in 1881, it has
been widely applied in the food and pharmaceutical in-
dustry.10 Hydrogels built from alginate are hydrophilic,
biocompatible, and non-immunogenic.11 Its physicochem-
ical properties such as viscosity and stiffness are easily
controllable and therefore make it an attractive candidate
for tissue engineering.9
Materials and methods
Establishing a three-dimensional alginate scaffold
for stem cell colonization
An aqueous solution of sodium alginate powder (3%) in
double-distilled water (ddH2O) was homogenized and sub-
sequently autoclaved. Then 15 grams (g) of the solution was
mixed with 2 milliliters (ml) of a 0.5 molar (M) calcium
carbonate (CaCO3) suspension using a magnetic stirrer, and
1 ml of a 1 M glucono-d-lakton solution was added. The
mixture was stood under ambient temperature, pressure,
and saturation until complete gelation was accomplished
(Supplemental Figure 1). Then the supernatant was dis-
carded, and the cured hydrogel was lyophilized at 0.05
millibar (mbar) and
55  C for 24  2 h to produce the
protoscaffold in cylindrical shape.
T. Hirsch et al.
Cytotoxicity of protoscaffolds
L929 (murine fibroblast cell line) cells were seeded in a 96-
well microtiter plate (Omnilab, Bremen, Germany). After
72 h, the Dulbecco’s Modified Eagle Media (DMEM) used for
incubation was removed. New DMEM, which had been
incubated for 24 h at 37  C with a protoscaffold (scaffold
prior to washing cycle) (n Z 8), was added at different
dilutions with regular DMEM (33%, 22%, 14.8%, and 9.9%).
Cells were cultured for an additional 72 h as described
above, and 0.5 milligram (mg)/ml 3-[4,5-dimethylthiazol-2-
yl-2,5-diphenyltetrazolium bromide (MTT) (Sigma, Tauf-
kirchen, Germany) was added. After 4 h, 0.01 M hydro-
chloric acid (HCl) containing 10% sodium dodecyl sulfate
was added to stop the reaction. Cells were lysed with
dimethyl sulfoxide, and absorbance at 562 nanometers
(nm) was measured using an enzyme-linked immunosorbent
assay (ELISA) plate reader (BioTek Instruments, Bad Frie-
drichshall, Germany).
Cell culture
All cells were cultured at 37  C, with 5% carbon dioxide
(CO2), and 90% humidified atmosphere. Different cell types
were procured as shown in Table 1.
The adipogenic differentiation media consisted of
DMEM þ 10% FCS HyClone, 2% L-glutamine, 1% insulin, 0.4%
indomethacin, 0.2% 3-isobutyl-1-methylxanthin, and 0.1%
dexamethasone.
Procurement of ASCs
Following approval from the local research ethics commit-
tee, the human ASCs were isolated from a single female
donor from subcutaneous adipose tissue as part of an
esthetic plastic surgery. Prior to the surgery, the patient
signed a written informed consent, conforming to the World
Medical Association Declaration of Helsinki (June 1964) and
subsequent amendments. Murine cells were isolated at the
University of Stuttgart after approval by German author-
ities. The porcine ASCs were procured from subcutaneous
fat tissue from a local slaughterhouse.
Isolation of ASCs and cell culture were performed ac-
cording to the description by Yang et al.12 with some
modifications. The subcutaneous tissue was washed
extensively with sterile PBS containing 1000 U/ml peni-
cillin and 1000 mg/ml streptomycin to remove contami-
nating blood cells. The specimen was then cut carefully.
Connective tissue and blood vessels were removed, and
the tissue was cut into 1 mm3 pieces. The extracellular
matrix was digested with 0.1% collagenase type I
Implant for autologous soft tissue reconstruction
Table 1 Different cell types used for cell culture.
Name Cell type
L929 Murine fibroblast cell line
hASC Human adipose-derived stem cells
HF AST Human fibroblasts from adult skin tissue
HUVEC Human umbilical vein endothelial cells
M2 Murine bone marrow-derived stem cells
pASCs Porcine adipose-derived stem cells
(Invitrogen, USA) at 37  C and shaken vigorously for 60 min
to separate the stromal cells from primary adipocytes.
The collagenase type I activity was then neutralized by
adding an equal volume of low-glucose DMEM (L-DMEM,
Hyclone, USA) containing 10% fetal bovine serum (FBS,
Invitrogen, USA). Dissociated tissue was filtered to remove
debris and centrifuged at 1500 rpm for 10 min. The sus-
pended portion containing lipid droplets was discarded,
and the cell pellet was resuspended and washed twice.
Contaminating erythrocytes were lysed using an osmotic
buffer, and the remaining cells were plated onto a 6-well
plate at a density of 1  106/ml. Plating and expansion
medium consisted of L-DMEM with 10% FBS, 100 U/ml
penicillin, and 100 mg/l streptomycin. Cultures were
maintained at 37  C with 5% CO2. The medium was
replaced after 48 h and every 72 h thereafter. Once the
adherent cells reached more than 80% confluence, they
were detached using 0.25% trypsine0.02% EDTA and
replated at a dilution of 1:3.
Adipocytes were induced by the StemPro adipogenesis
differentiation kit (Gibco, Darmstadt, Germany). After
differentiation, adipocytes were characterized by Oil Red O
and Nile red staining solutions to show lipid droplets in
induced cells.
Viability testing of hASCs in alginate scaffolds
Cubes of 0.5  0.5  0.5 cm were cut from protoscaffolds
(n Z 10). A cell suspension of 5  104 human adipose-
derived stem cells (hASCs) in DMEM þ 10% fetal calf
serum (FCS) þ 2% L-glutamine þ 1% penicillin/strepto-
mycin (p/s) was dripped onto each scaffold, and the cell-
scaffold constructs were incubated for 30 min. Regular
media was then added, and this was followed by a 72-h
incubation period. After incubation, media was dis-
carded and scaffolds were washed with phosphate-
buffered saline (PBS), incubated with 5 mg/ml 40 ,6-
diamidino-2-phenylindol (DAPI) solution, and washed
again. Moreover, similarly produced cell-scaffold con-
structs were observed by fluorescence microscopy and
then incubated with MTT-solution for 2 h, after which they
were observed under a light microscope. Additional cell-
scaffold constructs were incubated for 7 days, after
which acetoxymethyl calcein and propidium iodide live-
dead stainings were performed.
103
Origin
DSMZ: ATCC nbr. CCL!, NCTC clone 929 L
Isolated by the Hohenstein Institute from
human pendulous abdominal tissue
Isolated by the Hohenstein Institute from
human donor skin
Isolated by the Hohenstein Institute from
donated human cord blood
Isolation be the University of Stuttgart
Isolation at University Hospital Bergmannsheil
Bochum
Adipogenic differentiation of mesenchymal stem
cells (MSCs) in alginate scaffolds
Alginate scaffolds were colonized as described above with
hASCs (n Z 5), murine bone marrow-derived stem cells
(M2s) (n Z 5), and porcine adipose-derived stem cells
(pASCs) (n Z 15) and then incubated with adipogenic dif-
ferentiation media. Negative controls were human fibro-
blasts from adult skin tissue (HF ASTs) that are unable to
differentiate in adipogenic differentiation media and non-
differentiated hASCs, pASCs, M2s, and HF ASTs that were
cultured in reduced media with less FCS compared to reg-
ular media (n Z 1 each). Cell viability was analyzed with
calcein staining at several time points, and adipocytes were
stained with Nile red after 21 days of incubation. Moreover,
a collagen-containing cell suspension was used as described
above (n Z 15 for pASCs, n Z 5 for hASCs, M2s and HF ASTs)
and tested against HF ASTs that are unable to differentiate
(n Z 5).
Vascular endothelial growth factor (VEGF)
secretion assessment during adipogenic
differentiation of pASCs in cell culture and in
alginate scaffolds
At various time points during the 21-day period of pASC
adipogenic differentiation in cell culture or alginate scaf-
folds (n Z 5), VEGF levels were measured using a porcine
vascular endothelial cell growth factor ELISA kit (Cusabio
Biotech, Wuhan, China). Photometric evaluation was per-
formed with a GENios multi-well plate reader (Tecan
Group, Maennedorf, Switzerland). Cells were seeded at an
initial cell count of 5  104 in both groups.
Chorion allantois membrane (CAM) assay in the in
ovo system for in ovo testing of pASC angiogenic
potential of colonized alginate scaffolds
Fertilized chicken eggs (White Leghorn Lohmann LSL)
were incubated at 37  C and 60% humidity and turned in 8-
h intervals. Material for the experiment was injected by
the in ovo method. On day 7 of incubation, egg shells were
cut open and the CAM displayed after removal of the inner
amniotic membrane. Scaffolds were either colonized with
a collagen-containing pASC solution (n Z 21) or with pASC
104
solution without collagen (n Z 6). Negative controls were
cell-free scaffolds with (n Z 16) or without collagen
(n Z 5). Three days prior to CAM application, scaffolds
were cultured with reduced media and 1 day prior with
additive-free DMEM without potentially angiogenic fac-
tors. Scaffolds were colonized with 2  105 pASCs. One
scaffold sample of 0.5 cm length per egg was placed on
the CAM, after which the egg was sealed and incubated
for another 72 h, sitting in an upright position at the same
ambient conditions as before. Eggs were then opened
again and covered in 4% formaldehyde. The CAM and
scaffold sample were removed and assessed by stereo-
microscopy. Afferent vessels migrating toward the scaf-
fold on the CAM were counted using Image J (freeware
NIH).
VEGF secretion assessment of pASCs in alginate
scaffolds with and without collagen
Alginate cubes with an edge length of 0.5 cm were colo-
nized with 9  104 pASCs in cell suspension (n Z 4) or in
collagen-containing cell suspension (n Z 2) and incubated
for 2 days in reduced media. The media was then replaced
by DMEM without additives and assessed for VEGF content
after another 24 h of incubation.
Human umbilical vein endothelial cell tube
formation assay
BD Matrigel matrix (50 ml/well) was added to a cooled 96-
well plate, which then was centrifuged at 0  C and 1200
rounds per minute (rpm) for 10 min and then incubated for
1 h at 37  1  C, with 5% CO2, and 90  2% relative hu-
midity. Human umbilical vein endothelial cells (HUVECs)
were trypsinized, centrifuged for 5 min at 1200 rpm, and
suspended in M200 media. Then 50 ml of the HUVEC sus-
pension containing 2  103 HUVECs was added to each well
of the matrix-coated plate, and this was subsequently
incubated for 2 h to achieve complete cell adherence. The
media was then replaced with 50 ml of test media, and in-
cubation proceeded for another 24 h. Calcein-staining and
fluorescence microscopy assessment of HUVEC tubes was
then performed by measuring their length using Image J
(Freeware NIH). Test medias either contained collagen or
no collagen. Each test media (n Z 5 each) was assessed in
triplicate. Negative control was M200 without supplements,
while positive control was HUVEC culture media supple-
mented with growth factors.
Histology of pASC-colonized alginate scaffolds on
CAM
Samples (n Z 4 each) were stained with hematoxylin and
eosin stain for light microscopy. For fluorescence micro-
scopic display of the CAM vessels and cell nuclei within the
alginate scaffolds, a rhodamine-linked agglutinin antibody
and DAPI were utilized, and samples were assessed at
550 nm and 575 nm wavelength.
T. Hirsch et al.
Statistical analysis
Data are presented as mean  SD. Analysis of variance
(ANOVA) followed by Turkey’s multiple comparison test was
used to compare the differences in proliferation, CAM
angiogenesis, and HUVEC tube formation assays and in
VEGF secretion studies. p < 0.05 was considered significant
(a Z 0.05). Statistical analysis was performed using MS
Excel (Microsoft Corporation).
Image analyses were performed using Image J (Freeware
NIH).
Results
Establishing a three-dimensional alginate scaffold
for stem cell colonization
The produced hydrogel demonstrated macroscopically ho-
mogenous porosity and evenly distributed sodium gluconate
crystals that formed during gelation (Figure 1). The porous
structure was also observed under a light microscope.
Electron microscopy showed a very wavy inner structure
leading to a bigger surface area of the scaffold. The desired
scaffold form was either achieved by transferring the gel
into preformed vehicles shortly after start of the gelling
process or cutting of the lyophilized protoscaffold. The
porous size after lyophilization was 1 mm.
Protoscaffold hydration properties
The protoscaffold’s weight decreased continuously over the
course of five alternating hydration and lyophilization cy-
cles (Supplemental Figure 2). The mean weight of the
freshly gelled hydrogels (n Z 8) was 17.06  0.32 g,
whereas after five cycles, the rehydrated hydrogel weight
was 9.74  1.02 g. In the lyophilized condition, the proto-
scaffold’s weight was 0.71  0.01 g before and
0.46  0.02 g after five lyophilization cycles. Dry scaffolds
demonstrated a hydration capacity of 20.83  2.44 times
their weight throughout the five cycles, and this did not
significantly decrease over time.
Cytotoxicity of protoscaffolds
The DMEM incubated with protoscaffolds led to a growth
inhibition of 30% when subsequently incubated with L929.
Scaffold washing with ddH2O for 2 h and 4 h significantly
reduced cytotoxicity (Figure 2). The duration of the
washing process did not have any significant effect on
scaffold cytotoxicity.
Viability of hASCs in alginate scaffolds
The scaffolds (n Z 10) demonstrated colonization with
hASCs after 72 h (Figure 3). The vital cells formed clusters
within the scaffold. Viability of the cells was confirmed
using a MTT assay. Similar observations were made after 7
days of incubation.
Implant for autologous soft tissue reconstruction 105

Figure 1 Protoscaffold structure. aec: Prior to lyophilization; def: after lyophilization. a: Freshly gelled hydrogel; b: light
microscopy of hydrogel; c: hydrogel surface captured by scanning electron microscopy; d: cut ready-to-use protoscaffold after
lyophilization; e: variation of structures after lyophilization; f: macroscopic image of porous formation.

Figure 2 Cytotoxicity of alginate scaffolds. The ddH2O-washed scaffolds demonstrated significantly less cytotoxic effects
compared to untreated scaffolds throughout all media dilutions (33%, 22%, 14.8%, and 9.9%) (*). Dilutions were prepared by the
addition of regular DMEM. There was no significant difference between 2 h and 4 h washing cycle (#) (p < 0.001).
106 T. Hirsch et al.

Figure 3 Viability of hASCs in alginate scaffolds. a þ b: Grouped and round-shaped hASCs within the scaffold structure after 72 h
of incubation; c þ d: within the scaffold structure after 7 days of incubation; a: DAPI-fluorescence staining of cell nuclei; b: light
microscopy of MTT-stained viable cells; c: calceinefluorescence stain of vital cells; d: live (calcein)/dead (PI)efluorescence stain.

Adipogenic differentiation of human, murine, and
porcine MSCs
The MSCs within the alginate scaffold successfully differ-
entiated into adipocytes in cell culture flasks after 21 days
of incubation (Figure 4). Differentiation of hASCs, pASCs,
and M2 within the alginate scaffold was successful, and high
cell density was observed after only 7 days of incubation
(Supplemental Figure 3). Well-sized lipid vacuoles indicated
a high degree of adipocyte differentiation. However, cell
density decreased toward the end of the 21-day culture
period. The addition of collagen did not seem to influence
stem cell functionality but instead led to a stretched,
fibroblast-like morphology within the alginate scaffold.
VEGF secretion during adipogenic differentiation
alginate scaffolds did not alternate significantly over the
course differentiation (Figure 5). Significant differences in
VEGF secretion by the addition of collagen to the scaffolds
were not observed (Supplemental Figure 4).
CAM angiogenesis assay
Alginate scaffolds with 2  105 pASCs with and without
collagen were tested for the quantity of afferent embryonic
CAM vessels (Figure 6a). Mean vessel count for alginate
scaffolds with collagen was 56.00  2.74, for alginate
scaffolds without collagen 52.83  3.21, for cell free
scaffolds with collagen 39.38  2.30, and for cell free
scaffolds without collagen 32.60  4.08 (Figure 6b). There
was no significant difference between the collagen and
non-collagen groups.

pASCs cultured for 21 days and differentiated into adipo-
cytes were assessed for VEGF secretion and compared with
undifferentiated pASCs in cell culture. The undifferenti-
ated multi-potent stem cells secreted significantly more
VEGF than the differentiated cells. Moreover, the highest
VEGF release occurred during the first day of the differ-
entiation process. However, VEGF secretion by pASCs in
HUVEC tube formation assay
The media of collagen-containing samples did not lead to
significantly longer HUVEC tubes, whereas positive and
negative controls caused significantly longer and shorter
tube formation, respectively, than media from cell-
colonized alginate scaffolds (Figure 7).

Figure 4 Adipogenic differentiation of human, murine, and porcine MSCs. Cells were differentiated in cell culture for 21 days. a:
hASCs; b: M2; c: pASCs. Intracellular lipid vacuoles stained with Oil Red O at 21 days.
Implant for autologous soft tissue reconstruction 107

Figure 5 VEGF secretion by pASCs in cell culture and in alginate scaffolds during adipogenic differentiation. In cell culture,
undifferentiated pASCs released significantly more VEGF than differentiated pASCs from day 7 onward (#). Although there was no
significant difference in VEGF release between pASCs on day 1 of differentiation and undifferentiated pASCs, VEGF content was
significantly higher on day 1 of differentiation (*). pASCs in alginate scaffolds did not generally release more VEGF than pASCs in cell
culture. During differentiation in the scaffold, there also was no significant difference compared to the undifferentiated control,
but the VEGF content of native pASCs on day 1 was significantly higher than that on day 21 (þ) (p < 0.05).

During adipogenic differentiation of pASCs in cell cul-
ture, the conditioned culture media from day 21 led to the
longest HUVEC tubes, which were even significantly longer
than the ones from day 1 undifferentiated pASCs media.
Media from day 7 of differentiation onward caused signifi-
cantly longer tube formation than 21-day undifferentiated
pASC media. In the alginate scaffolds, progressing differ-
entiation to adipocytes resulted in progression of HUVEC
tube length (Figure 8).
Histology of pASC-colonized alginate scaffolds on
CAM
Histology showed cells in both the pASC-colonized and non-
colonized alginate scaffolds. Their morphology with a
nucleieplasma relation indicated that they were vital cells.
However, immunohistochemistry with agglutinin and DAPI
staining were not successful because of the unspecific up-
take of the stains and the resulting strong background
fluorescence, preventing further specification and quanti-
fication of the cells within the scaffold (Supplemental
Figure 5).
Discussion
The aim of this study was to engineer an implant using a
biocompatible and angiogenic matrix for soft tissue
replacement. An alginate scaffold was fashioned, and after
in vitro testing and process optimization, its cytotoxicity
was minimized. Adipose-derived stem cells were isolated,
characterized, and integrated into the scaffold with sub-
sequent differentiation into adipocytes. The superior
angiogenic properties of the construct regarding sustained
VEGF release, HUVEC tube formation, and in ovo vessel
formation were demonstrated by VEGF ELISA, HUVEC tube
formation assay, and a CAM angiogenesis assay.
The construction of the scaffold by internal gelling in our
study proved to be simple, fast, reproducible, and cost-
effective. The scaffolds were versatile in that they could
either be cut into shape as freshly gelled hydrogels,
enabling intraoperative fitting to a soft tissue defect, or
formed by transfer prior to the completion of the gelling
process into a preformed container. The scaffold texture
before and after cell colonization was firm, which would be
of great importance for surgical handling. The most com-
mon production form of alginate scaffolds is external gel-
ling for fabrication of microstructural alginate beads, which
are used for the controlled release of drugs or other macro-
molecules.13e16 However, internal gelling is considered su-
perior when it comes to bigger three-dimensional con-
structs because of the more homogenous gel structure that
results from a slower gelling process.17 Scaffolds showed a
constant hydration capacity of around 21 times their own
weight, indicating high inner porosity. Although sponge-like
swelling during hydration was observed, scaffolds retained
their original shape, which would enable intraoperative
fitting into the required shape in either hydrated or dried
condition.
The good biocompatibility of the biopolymer alginate is
well known.18 Therefore, a cytostatic influence rather than a
direct cytotoxic effect of the scaffold was expected. Inter-
estingly, washing the protoscaffolds with ddH2O significantly
decreased these cytostatic effects, possibly because of the
elimination of unbound polymer chains, sodium gluconate
crystals, and calcium. A certain level of cytostatic mecha-
nisms is believed to be beneficial for alginate in the context
of tissue engineering because of the prevention of cell
108 T. Hirsch et al.

Figure 6 Angiogenic potential in ovo: after CAM removal from the egg and storage in formalin, images were produced via
stereomicroscopy. The number of vessels were then counted using Image J software. Group sizes were as follows: pASC-colonized
scaffolds without collagen, n Z 6; pASC-colonized scaffolds with collagen, n Z 21; cell-free alginate scaffolds, n Z 5; cell-free
alginate scaffolds with collagen, n Z 16. Cell-free alginate scaffolds with and without collagen served as negative controls. a:
pASC-colonized alginate scaffold in ovo immediately prior to explantation of the CAM; b: explanted CAM with collagen-containing
pASC-colonized alginate scaffold; c: the negative control of a cell-free collagen-containing alginate scaffold demonstrates a lower
density of embryonic vascular density than stem cell-colonized alginate scaffolds on the macroscopic level; d: pASC-colonized
scaffolds with and without collagen demonstrated significantly more afferent vessels on the CAM than their respective negative
controls (NC, #, x). The comparison between pASC-colonized scaffolds with and without collagen (*), and direct comparison with
the respective negative controls (þ) did not demonstrate significant differences. # Z p < 0.05; x Z p < 0.001; * Z p > 0.05;
þ Z p > 0.05.

aggregation with the lack of central nutrient supply and density in our experiments decreased over time, which has
consecutive necrosis within the scaffold.19 also been described by others.21 However, after 7 days of
Alginate itself does not offer binding structures for incubation, the remaining cells were vital and arranged in
cellular surface proteins for cell adhesion.20 The cell clusters, suggesting that cells without contact diffused out
Implant for autologous soft tissue reconstruction 109

Figure 7 HUVEC tube sizes after conditioning with media from pASC-colonized scaffolds. Incubation of HUVECs with media from
pASC-colonized alginate scaffolds and collagen-containing pASC-colonized alginate scaffolds led to significantly longer tubes than
with their respective negative controls. There was no significant difference between pASC-colonized alginate scaffolds and
collagen-containing pASC-colonized alginate scaffolds, as well as no significant difference with the negative controls. Tubes in
positive control in combination with growth factor-enhanced media were significantly larger and tubes in negative controls in
combination with growth factor-enhanced media were significantly smaller than those in pASC-colonized scaffolds (p1 Z 1.
quartile; p3 Z 3. quartile; p < 0.001).

of the scaffold. A further implication is that incubation of
the scaffold after colonization does not help achieve higher
cell density within the scaffold, underscoring the impor-
tance of fast and effective isolation of large amounts of
ASCs from autologous fat. This would enable a procedure
where stem cells are procured, colonized in the scaffold,
and implanted into the soft tissue defect in a single oper-
ation, decreasing patient morbidity and enhancing eco-
nomic effectiveness.
Adipogenic differentiation of ASCs in alginate has been
described previously.14,21,22 However, this study is the first
to use a scaffold fabricated with alginate. Human, porcine
and murine MSCs were successfully differentiated into ad-
ipocytes within the alginate scaffolds. For differentiation
within the soft tissue defect, adipogenic induction media
can be primarily integrated into the scaffold.14
VEGF secretion by pASCs decreased significantly during
adipogenic differentiation in cell culture. Interestingly, this
effect did not occur within the alginate scaffold, where a
constant VEGF release was observed. This behavior may be
caused at physiologic pH by the electrostatic interactions
between the positive VEGF molecules and negative alginate
polymer chains.23 Slow release of VEGF from alginate
microbeads over a time span of 14e20 days has previously
been described.23,24 Constant VEGF release may provide
the basis for a sustainable long-term vascularization trigger
in ovo.
The colonization of alginate scaffolds with pASCs, which
release a variety of growth factors besides VEGF,25 caused a
significantly stronger in ovo vascularization than that with
the scaffold alone.
In the HUVEC tube formation assay, incubation with
media from colonized scaffolds led to significantly longer
tubes than incubation with media from the scaffolds alone.
Interestingly and despite lower VEGF levels, differentiated
pASCs caused longer tube formation compared to undif-
ferentiated stem cells. These findings again may be caused
by a number of other growth factors produced by
adipocytes.26
In all in vitro and in ovo experiments, collagen did not
demonstrate any significant effect on the level of VEGF,
tube formation, or vascularization.
The benefits of a tissue-engineered implant for autolo-
gous soft tissue reconstruction using a derived stem cell-
colonized biocompatible scaffold are diverse: the autolo-
gous stem cells will not cause an immune response. Instead,
they will release various growth factors promoting vascu-
larization and tissue integrity. The biocompatible scaffold
does not have cytotoxic properties and will resolve by
resorption. It can be pre-shaped prior to surgery or
customized for a perfect fit intraoperatively. In the case of
ASCs, the stem cells can be procured by liposuction and
subsequently colonized on the scaffold during the same
one-step surgical procedure prior to scaffold implantation
(Figure 8). This possibility may make such an implant a cost-
effective alternative with low morbidity within the recon-
structive ladder.
Autologous transplantation of fatty tissue in small vol-
umes has shown good results. However, adipose tissue
exceeding a critical size is at risk to suffer from central
necrosis with consecutive resorption and volume loss. The
use of a scaffold in combination with ASCs may address this
issue in three ways: (i) it may sustain its original volume
despite loss of cellular content until vascularization enables
repopulation of necrotic zones, (ii) the ASCs may maintain
an accelerated vascularization of the implant, (iii) the
alginate matrix may lead to a sustained release of growth
factors produced by ASCs.
110 T. Hirsch et al.

Figure 8 HUVEC tube sizes after conditioning with media from adipogenic pASCs differentiated in cell culture or in alginate
scaffolds. Media from pASCs differentiated in alginate scaffolds (blue stripes) lead to significantly longer tube formation than
media from pASCs differentiated in cell culture (yellow stripes) on days 1 and 14 of differentiation (#). In the scaffold group, the 21-
day differentiated pASCs resulted in the significantly longest tubes (*). The shortest tubes were observed in the 21-day undif-
ferentiated negative control (x). In the cell culture group, 21-day differentiated pASCs led to the longest tubes, which were
significantly longer than tubes from negative control from day 1 (þ) (PC Z positive control; NC Z negative control; d Z day;
p1 Z 1. quartile; p3 Z 3. quartile) (* Z p < 0.05) (# Z p < 0.001) (þ Z p < 0.001) (x Z p < 0.05).

In this study, a general concept using a biocompatible
scaffold in combination with ASC colonization has demon-
strated potential on a small-scale level. However, the 2.
greatest challenge in the past has been to successfully
3.
translate this potential to a larger scale. Therefore, the
next steps toward clinical application will be implant
testing in a large animal for defining applicable volumes 4.
and further explore the potential for clinical application.
Conflict of interest 5.
The authors report no proprietary or commercial interest in 6.
any product mentioned or concept discussed in this article.
Appendix A. Supplementary data 7.
Supplementary data related to this article can be found at 8.
http://dx.doi.org/10.1016/j.bjps.2017.08.009.
9.
References
10.
1. Naidoo P, Liu VJ, Mautone M, Bergin S. Lower limb complica-
tions of diabetes mellitus: a comprehensive review with clini-
copathological insights from a dedicated high-risk diabetic foot
multidisciplinary team. Br J Radiol 2015;88(1053). http:
//dx.doi.org/10.1259/bjr.20150135.
Onida S, Davies AH. Predicted burden of venous disease.
Phlebology 2016 Mar;31(Suppl. 1):74e9.
Sampson UKA, Fowkes FGR, McDermott MM, et al. Global and
regional burden of death and disability from peripheral artery
disease: 21 world regions, 1990 to 2010. Glob Heart 2014:145e58.
Zielins ER, Luan A, Brett EA, Longaker MT, Wan DC. Thera-
peutic applications of human adipose-derived stromal cells for
soft tissue reconstruction. Discov Med 2015;19(105):245e53.
Volz A-C, Huber B, Kluger PJ. Adipose-derived stem cell dif-
ferentiation as a basic tool for vascularized adipose tissue
engineering. Differentiation 2016 Mar 11;92(1-2):52e64.
Mizuno H, Tobita M, Uysal AC. Concise review: adipose-derived
stem cells as a novel tool for future regenerative medicine.
Stem Cells 2012:804e10.
Chan BP, Leong KW. Scaffolding in tissue engineering: general
approaches and tissue-specific considerations. Eur Spine J
2008;17(Suppl. 4):476e9.
Sterodimas A, De Faria J, Correa WE, Pitanguy I. Tissue engi-
neering in plastic surgery: an up-to-date review of the current
literature. Ann Plast Surg 2009;62(1):97e103.
Lee KY, Mooney DJ. Alginate: properties and biomedical ap-
plications. Prog Polym Sci 2012;37(1):106e26.
Liew CV, Chan LW, Ching AL, Heng PWS. Evaluation of sodium
alginate as drug release modifier in matrix tablets. Int J Pharm
2006;309(1e2):25e37.
Implant for autologous soft tissue reconstruction
11. Shapiro L, Cohen S. Novel alginate sponges for cell culture and
transplantation. Biomaterials 1997 Apr;18(8):583e90.
12. Yang XF, He X, He J, et al. High efficient isolation and sys-
tematic identification of human adipose-derived mesenchymal
stem cells. J Biomed Sci 2011;18. ISSN: 1423-0127:59
[Electronic].
13. Leslie SK, Cohen DJ, Sedlaczek J, Pinsker EJ, Boyan BD,
Schwartz Z. Controlled release of rat adipose-derived stem
cells from alginate microbeads. Biomaterials 2013;34(33):
8172e84.
14. Handel M, Hammer TR, Hoefer D. Adipogenic differentiation of
scaffold-bound human adipose tissue-derived stem cells (hASC)
for soft tissue engineering. Biomed Mater 2012;7(5):1e10.
15. Puguan JMC, Yu X, Kim H. Characterization of structure,
physico-chemical properties and diffusion behavior of Ca-
Alginate gel beads prepared by different gelation methods. J
Colloid Interface Sci 2014;432:109e16.
16. Komatsu M, Konagaya S, Egawa EY, Iwata H. Maturation of human
iPS cells-derived dopamine neuron precursors in alginate-
Ca(2þ) hydrogel. Biochim Biophys Acta 2015;1850(9):1669e75.
17. Kuo CK, Ma PX. Ionically crosslinked alginate hydrogels as
scaffolds for tissue engineering: part 1. Structure, gelation
rate and mechanical properties. Biomaterials 2001;22(6):
511e21.
18. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem Rev
2001:1869e79.
111
19. Hunt NC, Shelton RM, Grover LM. Reversible mitotic and
metabolic inhibition following the encapsulation of fibroblasts
in alginate hydrogels. Biomaterials 2009;30(32):6435e43.
20. Rowley JA, Madlambayan G, Mooney DJ. Alginate hydrogels as
synthetic extracellular matrix materials. Biomaterials 1999;
20(1):45e53.
21. Kim D, Monaco E, Maki A, et al. Morphologic and transcriptomic
comparison of adipose- and bone-marrow-derived porcine
stem cells cultured in alginate hydrogels. Cell Tissue Res 2010;
341(3):359e70.
22. Jing W, Lin Y, Wu L, et al. Ectopic adipogenesis of precondi-
tioned adipose-derived stromal cells in an alginate system. Cell
Tissue Res 2007;330(3):567e72.
23. Peters MC, Isenberg BC, Rowley JA, Mooney DJ. Release from
alginate enhances the biological activity of vascular endothe-
lial growth factor. J Biomater Sci Polym Ed 1998;9(12):
1267e78.
24. Gu F, Amsden B, Neufeld R. Sustained delivery of vascular
endothelial growth factor with alginate beads. J Control
Release 2004;96(3):463e72.
25. Tsuji W, Rubin JP, Marra KG. Adipose-derived stem cells: im-
plications in tissue regeneration. World J Stem Cells 2014;6(3):
312e21.
26. Ailhaud G, Grimaldi P, Négrel R. Cellular and molecular aspects
of adipose tissue development. Annu Rev Nutr 1992;12(1):
207e33.