ORIGINAL PAPER

Enhanced wound vascularization using a dsASCs seeded FPEGscaffold

David O. Zamora • Shanmugasundaram Natesan •

Sandra Becerra • Nicole Wrice • Eunna Chung •

Laura J. Suggs • Robert J. Christy

Received: 31 January 2013 / Accepted: 29 April 2013 / Published online: 26 May 2013

� Springer Science+Business Media Dordrecht (outside the USA) 2013

Abstract The bioengineering of autologous vascular

networks is of great importance in wound healing. Adi-

pose-derived stem cells (ASCs) are of interest due to their

ability to differentiate toward various cell types, including

vascular. We hypothesized that adult human ASCs

embedded in a three-dimensional PEG-fibrin (FPEG) gel

have the ability to modulate vascularization of a healing

wound. Initial in vitro characterization of ASCs isolated

from discarded burn skin samples (dsASCs) and embedded

in FPEG gels indicated they could express such pericyte/

smooth muscle cell markers as a-smooth muscle actin,

platelet-derived growth factor receptor-b, NG2 proteogly-

can, and angiopoietin-1, suggesting that these cells could

potentially be involved in a supportive cell role (i.e., per-

icyte/mural cell) for blood vessels. Using a rat skin exci-

sion model, wounds treated with dsASCs-FPEG gels

showed earlier collagen deposition and wound remodeling

compared to vehicle FPEG treated wounds. Furthermore,

the dsASCs-seeded gels increased the number of vessels in

the wound per square millimeter by day 16 (*66.7 vs.

*36.9/mm2) in these same studies. dsASCs may support

this increase in vascularization through their trophic con-

tribution of vascular endothelial growth factor, as deter-

mined by in vitro analysis of mRNA and the protein levels.

Immunohistochemistry showed that dsASCs were localized

to the surrounding regions of large blood-perfused vessels.

Human dsASCs may play a supportive role in the forma-

tion of vascular structures in the healing wound through

direct mechanisms as well as indirect trophic effects. The

merging of autologous grafts or bioengineered composites

with the host’s vasculature is critical, and the use of

autologous dsASCs in these procedures may prove to be

therapeutic.

Keywords Angiogenesis � ASCs � Wound healing �Fibrin � Collagen � PEG

Introduction

Extremity soft-tissue loss, resulting from direct impact or

thermal injury, is common to most military conflicts. His-

torically these wounds constitute the majority of injuries,

cost burden, and tend to increase the overall length of

hospitalization [1]. During Operation Iraqi Freedom and

Operation Enduring Freedom the frequent use of impro-

vised explosive devices resulted in an increase in complex

burn injuries involving a high percentage of total body

surface area (TBSA) [2, 3]. Current treatment strategies for

such burn injuries consists of immediately covering the

debrided wound areas with autografts or acellular allografts

[4]. Autologous skin grafts have better prognosis for

Disclaimer: The opinions and assertions contained herein are the

private views of the authors and are not to be construed as official or

reflecting the views of the Department of Defense or Department of

Army. The authors are employees of the U.S. Government, and this

work was prepared as part of their official duties. This research was

funded by the U.S. Army Medical Research and Materiel Command.

This study has been conducted in compliance with the Animal

Welfare Act, the implementing Animal Welfare Regulations, and the

principles of the Guide for the Care and Use of Laboratory Animals.

D. O. Zamora � S. Natesan � S. Becerra � N. Wrice �R. J. Christy (&)

Regenerative Medicine Research Program, United States Army

Institute of Surgical Research, 3698 Chambers Pass, BHT 1:

Bldg 3611, Fort Sam Houston, TX 78234-6315, USA

e-mail: robert.j.christy12.civ@mail.mil

E. Chung � L. J. Suggs

Department of Biomedical Engineering, The University of Texas

at Austin, 1 University Station, Austin, TX 78712-0238, USA

123

Angiogenesis (2013) 16:745–757

DOI 10.1007/s10456-013-9352-y

incorporation and wound healing, but autografts create a

new wound on the patient. Furthermore, undamaged skin

tissue that can be used for grafting is limited in patients

with large TBSA burns. Therefore, treatment of large

TBSA burn patients necessitates the use of allografts skin

substitutes. Currently there are a number of commercially

available products on the market (e.g., Integra, Alloderm)

however, the function of these grafts may be compromised

due to lack of cell infiltration and tissue revascularization.

Oxygen and nutrient perfusion typically occurs within a

100- to 200-lm area of a blood vessel; any cellular com-

ponents with a distance greater than this will be exposed to

ischemic conditions and eventually undergo apoptosis.

Unfortunately, acellular grafts are limited by the length of

time it takes for cell infiltration of the tissue, inosculation

and revascularization [4–6]. Therefore, developing alter-

native strategies to deliver cells and quickly revascularize

these wounds is essential.

New bioengineered skin substitutes that deliver cells

within the graft, promote revascularization and therefore

increase long-term survival and remodeling of an allograft

skin substitute are being developed. Initial skin substitutes

used such strategies as co-culturing of vascular endothelial

cells and fibroblast cells in a collagen based matrix [7], self

assembled co-cultured human umbilical vein endothelial

cells (HUVECs) and dermal microvascular cells [8],

sequentially seeding the apical and basal surfaces of acel-

lular dermis with cultured human keratinocytes and HU-

VECs [9], and pre-seeding endothelial progenitor cells

(EC) [10]. These methodologies have had some success.

Unfortunately, obtaining vascular and/or keratinocyte cells

from a patient with large TBSA wounds, where there is a

limited source of cells to develop these substitutes is

problematic. Current strategies to overcome this problem

focus on optimizing the two main components of tissue

engineered grafts: biomaterial scaffolds and cells. The use

of stem cells, such as epidermal stem cells or mesenchymal

stem cells from bone marrow or adipose tissue, which have

the potential to differentiate into various phenotypes, rather

than using specific cell types (e.g. keratinocytes) may

provide a strategy for functional wound repair and com-

plete regeneration of skin [11]. Many different biomaterials

have been developed and shown to have beneficial effects

on wound healing and provide a microenvironment that

allows cells to proliferate and differentiate. Natural poly-

mers including the extracellular matrix proteins, collagen

and hyaluronic acid have been used extensively in wound

repair and for acellular skin grafts because of their in vitro

and in vivo cell biocompatibility [4, 12]. Fibrin, another

natural biopolymer, has been used clinically as a hemo-

static agent and as a sealant for soft tissue wounds. Still the

major concern of using fibrin-based hydrogels as wound

repair scaffolds is their relatively quick contraction, low

mechanical stiffness (which limits durability), and their

rapid degradation once placed at the wound site. These

aspects alone can hinder proper tissue formation and

reconstruction. To overcome these problems, fibrin has

been modified to increase its durability, ease of use, and

longevity in a wound. One such modification involves the

copolymerization of fibrin with polyethylene glycol (PEG)

[13, 14]. The addition of extra cross-linking between fibrin

and polyethylene glycol (FPEG) during thrombin mediated

fibrin polymerization produces a highly hydrated gel

microenvironment allowing cell seeding within the matrix

for direct delivery of cells to a wound. This combining

stem cells with the natural biological activity of fibrin has

been shown to encourage tissue and blood vessel in-growth

in the healing wound [15, 16]. Based on these criteria,

FPEG is an excellent candidate for combinatorial therapies

involving both biomaterials and stem cells.

When considering stem cells as a potential cell source

to be combined with FPEG; adipose-derived stem cells

(ASCs) are an attractive cell source due to their ease in

isolation, relative abundance, ability to differentiate

toward various cell types in vitro, and their beneficial

function in wound healing in vivo [17, 18]. For example,

ASCs have been shown to successfully enhance revascu-

larization of ischemic hind limbs in mice over such other

cells types as human mesenchymal stem cells (MSCs)

from bone marrow [19]. ASCs derived from humans

appear to assist in reestablishing blood flow in ischemic

tissue salvage experiments [20]. Intramyocardial injections

of human ASCs have been found to assist in local angio-

genesis following myocardial infarction in animals [21]. In

addition, recently it has been shown that ASCs embedded

in a dermal substitute help to improve the regeneration of

skin by increasing blood vessel formation and collagen

synthesis [22]. ASCs also have great implications in their

ability to enhance superficial wound healing. Local

implantation of ASCs has proven effective in supporting

epidermal healing in full-thickness skin wounds of pigs

and rats [23, 24].

Recent studies have shown that rat and human ASCs,

isolated from epididymal fat or from discarded human burn

tissue (dsASCs), also have plasticity to differentiate

towards vascular cell phenotypes when cultured in three-

dimensional matrices in vitro [25, 26]. We hypothesize that

the dsASCs embedded in three-dimensional gels have the

ability to differentiate into vascular cell types in vivo and

therapeutically modulate wound healing and vasculariza-

tion. In this study, we have systematically combined

human dsASCs, human fibrinogen, and polyethylene glycol

to create a unique matrix to be investigated using an

established wound healing model.

746 Angiogenesis (2013) 16:745–757

123

Materials and methods

dsASCs isolation

Discarded human burn tissues were isolated according to

our previously published protocol [25]. In brief, ASCs were

isolated from discarded skin samples from patients under-

going burn wound debridement at the U.S. Army Institute

of Surgical Research (USAISR) Burn Center, Fort Sam

Houston, Texas. The skin samples were brought to the

laboratory immediately after debridement and processed.

This study was conducted under the protocol reviewed and

approved by the U.S. Army Medical Research and Materiel

Command Institutional Review Board. Discarded burn

tissue samples were collected in accordance with the

approved protocol, HSC20080290 N. The authors were

blinded from some patient information including burn

depth, percentage of TBSA, and anatomical location of the

burn. The skin samples were washed 3–4 times with

Hank’s buffered salt solution (HBSS) to remove adherent

blood clots. The hypodermal layer was dissected away

from the dermal region, transferred to a Petri dish, and

finely minced with scissors. The minced tissue was sus-

pended in HBSS and centrifuged for 10 min at 5009g at

16 �C. The floating tissue was carefully collected; and to

every 1 ml of the floating fraction tissue, 3,500 units of

collagenase type II (Sigma-Aldrich, St. Louis, USA) was

added and incubated for 45–60 min at 37 �C in an orbital

shaker incubator at 125 rpm. The undigested tissue was

removed by sequential passage through 100- and 70-lm

nylon mesh filters. The filtrate was then centrifuged at

5009g for 10 min at 16 �C, treated with BD Pharm

LyseTM lysing buffer (BD Bioscience, San Jose, CA, USA)

to remove any remaining red blood cells, and washed twice

with HBSS. The final cell pellets were resuspended in

growth media (MesenPRO RSTM basal medium), supple-

mented with MesenPRO RSTM growth supplement, anti-

biotic–antimycotic (100 U/ml of penicillin G, 100 lg/ml of

streptomycin sulfate, and 0.25 lg/ml of amphotericin B),

and 2 mM of L-glutamine (Life Technologies, Carlsbad,

CA, USA). This medium preparation is represented as

‘‘complete medium.’’ The resulting cell number, typically

*1.7 9 106, were cultured in T75 flasks (BD Falcon, NJ,

USA) and maintained in a 5 % carbon dioxide (CO2)

humidified incubator at 37 �C. After 4 h in culture, the

growth medium was replaced to remove any floating deb-

ris. The remaining attached cells are regarded as dsASCs.

Bioscaffold preparation

Fibrin and polyethylene glycol hydrogels were prepared as

previously described by Zhang et al. [13]. Succinimidyl

glutarate modified polyethylene glycol (SG-PEG-SG;

3,400 Da; NOF America Corporation, White Plains, NY,

USA) was dissolved using 4 ml of tris-buffered saline

(TBS, pH 7.8, Sigma-Aldrich), and then sterilized using a

0.22-lm filter just prior to starting the experiment to obtain

a stock solution of 8 mg/ml. Next, 500 ll of fibrinogen

stock (40 mg/ml in TBS) and 250 ll of PEG stock (8 mg/

ml) were mixed in a 12-well cell culture plate and incu-

bated for 20 min in a 5 % CO2 humidified incubator at

37 �C. This mixture constitutes a molar concentration ratio

of 10:1, SG-PEG-SG: fibrinogen. After incubation, 250 ll

of dsASCs (100,000 cells in MesenPRO complete medium)

were mixed with the PEGylated fibrinogen solution, and

immediately 1 ml of thrombin stock (25 U/ml; Sigma-

Aldrich) in 40 mM of calcium chloride (CaCl2) at a final

concentration of 10 U/ml was added. The 2-ml solution

was then quickly triturated once with the pipettor, and

immediately 1-ml aliquots were placed individually in a

12-well format cell culture insert of 8-lm pore size. Once

gel-cell mixtures were aliquoted into their respective wells,

the mixtures were incubated in a 5 % CO2 humidified

incubator at 37 �C for 10 min to allow for complete gela-

tion. The resulting FPEG gels were then washed twice with

HBSS and incubated with alpha minimal essential media

(a-MEM) supplemented with 10 % FBS in a 5 % CO2

humidified incubator at 37 �C. The formation of tube-like

networks by dsASCs migration was then observed over an

11-day period using standard light microscopy techniques.

RT-PCR analysis

Total ribonucleic acid (RNA) from dsASCs in FPEG gels

(3, 5, 7, 9, and 11 days) were isolated using TRIzol� LS

Reagent (Invitrogen), with slight modifications. Before

processing, the gels were rinsed with HBSS once and

carefully removed from the culture well. Four gels from

each time point were pooled together and minced, 16 ml of

TRIzol� LS Reagent was added, and then samples were

incubated for 15 min on ice. After incubation, 8 ml of

chloroform was added, samples were mixed, and the

aqueous phase was separated by centrifugation

(13,0009g). Total RNA was then purified using Qiagen’s

mini spin columns (Qiagen, Valencia, CA, USA). The

concentration and quality of the purified RNA was deter-

mined at 260/280 optical density ratio using a NanoDrop

spectrometer (NanoDrop Technologies, Inc., Wilmington,

DE, USA). Complementary deoxyribonucleic acid (cDNA)

was synthesized from 150 ng of total RNA, using Super-

Script III first-strand synthesis supermix with oligo-dT

primers (Invitrogen). A negative control sample (H2O) was

used to assess random production of cDNA through con-

taminants. Oligonucleotide primer sequences (Acta2,

CD140b, NG2, Angpt-2, Angpt-1, and VEGF165) were

purchased from SA Biosciences (A Qiagen Company;

Angiogenesis (2013) 16:745–757 747

123

Frederick, MD). Master mixes were made containing 200

nM of forward and reverse primers containing SYBR

Green-ER, and qPCR supermix (Invitrogen); and the syn-

thesized cDNA was added to the appropriate wells. Real-

time polymerase chain reaction (RT-PCR) was carried out

using a Bio-Rad CFX96 thermal cycler system (Bio-Rad,

Hercules, CA, USA). Message RNA expression levels were

normalized to glyceraldehye-3-phosphate dehydrogenase

(GAPDH). Fold increase in expression levels for each

specific gene was normalized to the expression levels of

control passage 2 dsASCs. Fold increase in expression

levels for each gene was determined by the 2-DDCt method.

Immunochemistry

To analyze the differentiation of dsASCs, cells were seeded

within the FPEG gels and allowed to grow for 11 days

before being harvested. On the day of harvest, the medium

was aspirated from the wells; the FPEG-dsASCs gels were

gently removed from their cell culture inserts using a

spatula and immediately processed for immunocytochem-

istry. Briefly, the gels were washed with HBSS (twice,

5 min), fixed with 4 % paraformaldehyde (EMS, Hatfield,

PA) for 20 min, treated serially with increasing concen-

trations of sucrose (from 5 to 20 %, 30 min each incuba-

tion), and then incubated overnight with 20 % sucrose at

4 �C. The sucrose-treated gels were then embedded in a

20 % sucrose-HistoPrep (Fisher Scientific, Pittsburgh, PA,

USA) mixture (2:1) and flash-frozen using liquid nitrogen.

Sections, 10–12 lm thick, were then cut using a cryostat

(Leica Microsystems, Nussloch, Germany), washed with

sterile HBSS, and fixed with 4 % paraformaldehyde for

20 min. Nonspecific Fc receptor-mediated sites were

blocked by incubating the sections for 1 h with 1 % bovine

serum albumin in HBSS containing 0.01 % Triton X-100

and then washed twice (5 min) with HBSS. To assess the

endothelial immunophenotype, sections were stained with

anti-human CD31 (PECAM-1, 8 lg/ml; R&D Systems)

and von Willebrand factor (vWF, 10 lg/ml; Millipore,

Billerica, MA, USA) specific monoclonal primary anti-

bodies. For identifying pericyte immunophenotype,

human-specific monoclonal antibodies specific to chron-

droitin sulfate proteoglycan (NG2, 20 lg/ml; Millipore),

and platelet-derived growth factor receptor beta (PDGFRb/

CD140b, 10 lg/ml; R&D Systems) antibodies were used.

After incubation of unconjugated primary labeled anti-

bodies, sections were washed twice (5 min) with HBSS

and incubated with 5 lg/ml host species-specific Alexa

fluor 488 and/or Alexa fluor 594 secondary antibodies

(Invitrogen) for 45 min at 4 �C. Finally, the sections were

washed twice with HBSS (5 min) and nuclei stained with

Hoechst 33342 (Invitrogen). Nonspecific fluorescence was

evaluated using sections incubated with respective

fluorophore-labeled secondary antibodies. Epifluorescence

of cells and gel sections were observed using Olympus

IX71 inverted microscope equipped with reflected fluo-

rescence system (Olympus America Inc.). Photomicro-

graphs were taken using a DP71 digital camera, and image

overlay was carried out using DP controller application

software.

Animal model

Male Rowett nude (RNU) rats (athymic rats), deficient in T

cell function, weighing 175–250 g, were obtained from

Harlan Laboratories (Indianapolis, IN) and housed in the

animal care facility at the USAISR with access to water

and rat chow ad libitum. This study was conducted in

compliance with the Animal Welfare Act, the implement-

ing Animal Welfare Regulations, and the principles of the

Guide for the Care and Use of Laboratory Animals. The

rats underwent general anesthesia for the surgery after

receiving a preemptive dose of analgesic (buprenorphine

0.1 mg/kg, subcutaneous) 30 min prior to induction.

Anesthesia was induced by placing rats into a plexiglass

chamber filled with 1–3 % isoflurane and oxygen. Anes-

thetic was maintained using a vaporizer setting of 1–3 %

isoflurane delivered with a nose cone on a Bain circuit

hooked to the rodent gas anesthesia machine (VetEquip,

Inc., Pleasanton, CA, USA). A full-thickness skin excision

wound 1.5 cm in diameter was created on the dorsum of

the rat down to the panniculus. The rats were randomly

divided into three groups with a total of two rats per group

and treated as follows: saline control group (250 ll of

saline), FPEG gel, and FPEG-dsASCs gel treatments. Upon

placement of treatments, the wounds were covered with

DuoDERM� dressing (ConvaTec, Skillman, NJ, USA) and

evaluated at 4, 8, 12, and 16 days. When the study was

completed, the rats were euthanized; and the wound beds,

including the healthy skin margin of the healed area sur-

rounding the wound, were harvested and fixed with 10 %

neutral-buffered formalin for histological analysis.

Histology and vessel quantification

Histological analysis was performed on *5 lm sections of

formalin-fixed paraffin embedded granulation tissue and

normal skin tissue surrounding the wound collected from

the wound bed of the athymic nude rats. The sections were

stained with either hematoxylin and eosin or Masson’s

trichrome and examined under light microscopy to assess

the overall wound healing pattern in the rat tissue. For

blood vessel identification, paraffin embedded tissues were

immunolabeled with anti-vWF antibody (Cat#250A-1,

rabbit polyclonal) using Cell Marque (Rocklin, CA, USA)

and the Ventana automated tissue immunostaining unit

748 Angiogenesis (2013) 16:745–757

123

located at the Brooke Army Medical Center Histology

Laboratory, Fort Sam Houston, Texas. Prepared slides

were then analyzed using standard light microscopy and

blood vessels [*10 lm in luminal diameter were quan-

titated in the center of the wound. The parameters that

qualified a vessel to be counted are as follows: (1) positive

vWF staining, (2) vessels were counted only once even if

multiple sections of the same vessel appeared in the field of

view, as assessed by trajectory of vessel and diameter size

and (3) vessel had to be located within a 1.5-cm window

centrally located within the wound, as delineated by the

Olympus microscope software. Vessels quantified were

located within the newly forming tissue-gel treatments,

taking care to avoid vessels in the host tissue below and

surrounding the regenerating wound. For any given slide,

five 1-mm2 regions located within 1.5-cm wound center

were counted in a blinded fashion. The five regions counted

were averaged and are represented in Fig. 6a as raw data

per animal. To analyze the diameter of vessels that the

donated human cells associated with, dual labeled vessels

for human specific mitochondrial antigen (h-MT) and

CD31 were digitally measured using the Olympus micro-

scope software in vessels of these same regions.

Enzyme-linked immunosorbent assay (ELISA)

To assess VEGF protein production by dsASCs, the cells

(P2) were seeded within the FPEG gels similarly to the gels

prepared for the animal studies (50,000 cells/ml of gel).

However, in these experiments, the gels were cast within a

6-well plate filter insert (8.0 lm pore size; BD Falcon,

Franklin Lakes, NJ, USA). FPEG gels without dsASCs

were processed simultaneously as controls. Three millili-

ters of complete medium was added to the bottom chamber

and 2 ml of complete medium to the top chamber of the

filter insert, and then plates were placed in a 5 % CO2

humidified incubator at 37 �C. A complete medium change

was performed daily on both the inner and the outer wells.

Prior to removing the medium, 1 ml of conditioned med-

ium was collected from each sample and stored at -80 �C

until ready for analysis. Conditioned medium from the

corresponding time points (1, 3, 5, 7, 9, 11, and 15 days) were

quantified for the levels of human VEGF165 using com-

mercial ELISA kits (Quantikine�, R&D Systems) per

provided protocols and standard curves.

In vivo dsASCs localization

Cryosections (*10 lm) were cut from frozen tissue sam-

ples using a Leica CM1850 (Leica Microsystems, Buffalo

Grove, IL, USA) at a set temperature of -20 �C. Sections

were applied to poly-prep slides (Sigma-Aldrich, St. Louis,

MO, USA) and stored at -80 �C until further processed.

Tissue sections were acclimated to room temperature for

20 min and then rehydrated with 19 wash buffer (Dako,

Carpinteria, CA, USA) for 10 min. The endogenous

enzymes were blocked according to EnVision G/2 Dou-

blestain System (Dako), rabbit/mouse (diaminobenzi-

dine ? permanent red) kit instructions. To localize human-

specific cells within the tissue, sections were incubated with

10 lg/ml of human-specific anti-mitochondrial (h-MT),

mouse-purified monoclonal primary antibody (10 lg/ml,

Millipore, Billerica, MA, USA) for 1 h at room temperature

followed by an alkaline phosphatase/horseradish peroxi-

dase-labeled secondary antibody polymer system. Primary

labeled tissue was further processed for dual labeling fol-

lowing instructions from the labeling kit (Dako). Sections

were then incubated with 5 lg/ml of second primary anti-

body for 2 h using anti-platelet endothelial cell adhesion

molecule (PECAM-1; Millipore) to identify blood vessel

formation within the tissue. Concentration requirements for

the second primary antibody, anti-PECAM, was incubated

for 2 h at room temperature. Close attention was paid when

developing the tissue with diaminobenzidine and permanent

red chromagen because the rate varies with tissue type.

Lastly, some tissues were counterstained with 0.2 % methyl

green (Vector Laboratories, Burlingame, CA, USA) for

10 min to stain nuclei, then air-dried and mounted using an

aqueous mounting medium.

Results

Differentiation of dsASCs

Previous studies by our lab fully characterized the pheno-

type of the isolated dsASCs used in this study and con-

firmed the stem cell nature of these cells [25]. To investigate

the potential of the dsASCs (Fig. 1a) to differentiate toward

vascular phenotype lineages, the cells were cultured in a

three-dimensional FPEG gel (Fig. 1b) and maintained over

an 11-day period and analyzed at different time points (3, 5,

7, 9, and 11). Within 3 days of culture (Fig. 1c), the cells

had began sprouting tube-like structures; and by days 7 and

11 they became denser (Fig. 1d, e, respectively). Previous

reports by our lab [26] and others [27, 28] have demon-

strated that rat ASCs have an ability to express an endo-

thelial cell phenotype. Since the tube-like structures that the

human dsASCs formed within the FPEG were reminiscent

of endothelial cell tube structures, we analyzed these cul-

tures for their expression of CD31 and vWF. Interestingly,

CD31 and vWF were undetectable in these cultures by our

immunohistochemistry and PCR assays at the time points

indicated (data not shown). Current dogma indicates that

ASCs may exist in vivo as a pericytic niche [29]; therefore,

we investigated their potential for pericyte marker

Angiogenesis (2013) 16:745–757 749

123

expression (Fig. 1f). In these three-dimensional culture

conditions, dsASCs were induced to express pericyte-spe-

cific markers when cultured in two-dimensional tissue

culture conditions, as determined by 2-DDCt. A slight

increase in expression (*1–5-fold) of smooth muscle a-

actin (Acta2) and CD140b was observed, with little mod-

ulation of their expression over an 11 day period. A modest

increase (*15 fold by day 11) in NG2 expression was

observed over this same timeframe. Angpt-1 transcript,

which is expressed mainly by perivascular cells and mural

cells, increased in these cultures between days 3 and 7 and

returned to baseline levels by day 11, perhaps indicating the

termination of network formation. Angpt-2, a negative

regulator of Angpt-1 signaling during angiogenesis is gen-

erally known to be expressed by endothelial cells and

mesenchymal stem cells. In these cultures, the cells

exhibited a substantial fold increase (*5- to 15-fold

increase) of Angpt-2 between days 3 and 9 but drastically

Fig. 1 Molecular and

morphological characterization

of human dsASCs. a Phase-

contrast photomicrograph

depicting the morphology of

dsASCs (P1), b image depicts

the *1.5-cm-diameter FPEG

gel after removal from the filter

insert. The orange rectangle

depicts the gel at *0.3- to 0.5-

cm thickness. Once seeded in

the FPEG gel, dsASCs exhibited

a tubular network formation

over time c day 3, d day 7, and

e day 11. f Total RNA was

isolated from dsASCs and

FPEG-dsASCs gels (days 3–11)

and analyzed by real-time PCR

for transcript expression of

endothelial (CD31, vWF) and

pericytic (ACTA2, CD140b,

NG2, Angpt-1 and Angpt-2)

markers. While dsASCs

demonstrated mRNA expression

of pericytic markers over the

time frames examined (shown),

endothelial markers remained

undetectable in our assays (data

not shown). Gene expression

levels are represented as mean

fold changes (±standard

deviation). (g-l)

Immunostaining of dsASCs

cultured in FPEG gels for

11 days confirmed RT-PCR

results demonstrating the

expression of such pericytic

markers by dsASCs. Original

magnifications: a, c, d, e 9200;

g–l: 9400. (Color figure online)

750 Angiogenesis (2013) 16:745–757

123

down-regulated by day 11. Initial expression of Angpt-2

may be indicative of the inherent ‘‘stemness’’ of these cells.

To corroborate PCR data, protein expression of CD31,

vWF, and CD140b and NG2 markers by the cells was

examined by immunofluorescence staining. Similarly, cell

cultures did not stain positive for vascular endothelial

markers CD31 and vWF (data not shown) but did stain

strongly for NG2 and CD140b (Fig. 1h, k, respectively).

Wound healing and vascularization

To investigate the influence of FPEG-dsASCs gels on

wound healing, we implemented our established rat full-

thickness skin excision wound model and treated the

wounds with either dsASCs seeded FPEG gels, FPEG gels

alone, or saline, Fig. 2a (i–iv). The treatments were left on

the wounds over 4, 8, 12, and 16 days. At these time points,

the rats were euthanized; and wound beds, along with

normal skin adjacent to it, were examined by histological

technique. Overall, wounds appeared to close at similar

rates between all treatment groups. However, wound

remodeling and collagen deposition (blue stain in Figs. 2b,

4b) appeared to occur sooner in FPEG-dsASCs treated

wounds than in control treatments. Closer examination

showed that vessel-like structures were also present and

that these structures appeared sooner in the FPEG-dsASCs

treated wounds (day 8; Fig. 3) than in wounds lacking cell

treatment. The vessel-like structures in Figs. 3c, 4a, b did

contain red blood cells and were lined with a monolayer of

cells. To confirm that these vessel-like structures are per-

fused blood vessels, sections were immunostained for the

vascular endothelial cell specific marker vWF. The sections

showed positive vWF staining confining to the vessel-like

structures, thus confirming the vascularization of the

wound bed (Fig. 5). Upon quantitating blood vessel den-

sity, it became apparent that FPEG-dsASCs treatment

increased the amount of blood vessels in the healing

wound, compared to FPEG alone and saline treatments

(Fig. 6a). Furthermore, the blood vessels within the wound

bed of FPEG-dsASCs treatments appeared to be larger and

stained darker for vWF (Fig. 6b, c) than FPEG treatments

alone (Fig. 6d, e) and suggested that the presence of

dsASCs may enhance angiogenesis in these tissue gels.

Role of the dsASCs

To determine whether dsASCs play a direct role in blood

vessel formation in vivo, we needed to localize the dsASCs

within the wound bed with a h-MT specific antibody. With

this reagent, the human cells consistently localized to

medium sized vessels of *15.06 lm diameter, but not

smaller capillary vessels *5.27 lm, (Fig. 7a, d). It is

Fig. 2 Gross histology of treated excision wound. a (i) Circular

excision wound created on the dorsum of athymic nude rats. a (ii)

Clear FPEG gel, ±dsASCs, is immediately placed into the excised

wounds and securely bandaged into place. a (iii) Photomicrograph of

a day 4 excised wound bed that was cross-sectioned mid-sagitally.

Higher magnification of white rectangle is depicted in a (iv). The

horizontal dashed line in a (iii, iv) demarcates the host wound tissue

from the FPEG gel. Above the dashed line is the FPEG gel; below the

dashed line is the host. Gross examination of the wound bed indicated

that the groups treated with FPEG gels ± dsASCs had completely

integrated with the host tissue by day 12 (b). Masson’s trichrome-

stained tissue sections from all treatment groups on days 4, 8, 12, and

16. Original magnifications: a Gross photograph; b 940

Angiogenesis (2013) 16:745–757 751

123

important to note that FPEG gel treatments alone contained

little to no medium sized vessels, but did have smaller

capillary vessels that ranged in diameter from 4 to 9 lm.

This is consistent other analysis depicted in Figs. 3b, 5

(Days 8 & 16), and 6d, e. To analyze this observation,

higher magnification of the vessels indicated that the

human cells were incorporated into the vessel structure by

physically surrounding the blood vessel in a similar fashion

that pericytes do (Fig. 7b; brown stain) but did not appear

to localize to the inner CD31? endothelial lining (red

stain). Anti-CD31 antibodies used were cross-reactive to

both rat and human species alike.

To investigate a potential mechanism for dsASCs to

increase blood vessel density in a healing wound, we

examined their potential for VEGF expression in vitro

within a 3D FPEG gel culture system. FPEG-dsASCs gels

were harvested at different time points (3, 5, 7, 9, 11, and

15 days), and total RNA was isolated and analyzed for

VEGF expression. As early as 3 days after culture, an

increase in their VEGF expression was observed, as com-

pared to dsASCs cultured in a standard two-dimensional

culture system (Fig. 8a). A constant increase in the

amounts of VEGF mRNA expression by dsASCs was

observed over the period of culture. This observation also

translated into soluble VEGF protein by these cells, as was

determined by ELISAs of conditioned media during the

same time points (Fig. 8b).

Discussion

Extremity trauma from blunt force or penetrating objects

and chemical or thermal burns remains a leading cause of

morbidity for civilian and military personnel alike. More-

over, the healing and tissue regeneration of such large-

volume wounds remains a significant hurdle in treating

these patients and restoring functionality to their limbs. As

a part of the treatment regimen, necrotic or burned tissue is

surgically debrided from the wound site and discarded.

Often during this process, viable tissue is inevitably

removed as well. We previously reported on the ability to

isolate ASCs from the viable hypodermis region of deb-

rided skin (dsASCs), which is typically removed from burn

injured patients [25]. These cells are able to be isolated in

significant quantities for clinical use and maintain their

stem cell marker expression in vitro. In the present study,

we extend these findings by investigating their therapeutic

use in the angiogenic process of wound healing and tissue

regeneration when used in combination with a novel FPEG

hydrogel.

ASCs and vascularization

The notion that ASCs could be beneficial for the regener-

ation of vascular systems, either through endothelial or

mural cell differentiation or the secretion of paracrine fac-

tors, has been well documented in the literature over the

past decade [18, 23, 30–33]. In particular, attention has been

focused on the ability of ASCs to differentiate toward

vascular endothelium [26–28]. A number of reports indicate

that both human and animal ASCs can molecularly and

phenotypically resemble endothelium, but only up to a

certain extent. Generally, exogenous growth factors (i.e.,

VEGF, EGF) are supplemented into culture conditions to

induce ASCs toward EC differentiation. However, these

cells do not appear to fully express, simultaneously, all

characteristics of an end-differentiated endothelial cell (i.e.,

CD31, vWF, or eNOS expression, LDL uptake, single-layer

cobble-stone morphology, or tube formation on Matrigel).

Early studies by our lab and colleagues using rat ASCs

indicated that these cells were capable of expressing such

endothelial markers as CD31 and vWF, as well as form

lumen containing capillary-like tubes in vitro. Moreover,

this was accomplished without the use of growth factors and

was attributed to the unique composition of the FPEG 3D

hydrogel. Using similar methodologies, human dsASCs

Fig. 3 Histology of treated wounds. a, b, c Day 8 wounds were more

closely examined as both b FPEG and c FPEG-dsASCs wounds

appeared to possess functional blood vessels. Closer examination

indicated that wounds treated with FPEG-dsASCs had numerous

organized vessel structures (hashed oval region), whereas FPEG gel

(b) or saline (a) treatments alone exhibited fewer blood vessels.

Original magnifications: 9100

752 Angiogenesis (2013) 16:745–757

123

were isolated and their vasculogenic potential investigated

here. Initial studies indicated that these human dsASCs

were capable of forming networks in the three-dimensional

FPEG gel (Fig. 1). However, under these conditions,

dsASCs did not appear to express CD31 or vWF (data not

shown) but did express such pericytic markers as smooth

muscle actin (Acta2), platelet-derived growth factor

receptor beta (CD140b), chrondroitin-sulfate proteoglycan

(NG2), and angiopoietin (Angpt-1). Furthermore, dsASCs

appeared to target and surround newly formed blood vessels

of approximately 15.06 lm diameter, in a rat skin excision

wound model, but did not appear to localize at the inner

endothelial lining of the vessels. At no time were human

dsASCs detected as being associated with vessels of smaller

diameter (*8 lm or less). Whether this is reflective of

actual biology or beyond the limit of our detection still

remains to be determined. The immunolabeling method

used here is a unique technique that is specific for the

localization of human-specific mitochondrial antigen in

human, not rat, cells. Taken together, this data suggests that

the implanted human dsASCs appear to differentiate and

function in a supportive pericytic role and not as a lumenal

endothelial cell, as observed in our animal model. This

observation is in line with the emerging hypothesis that

ASCs, a subtype of mesenchymal stem cells, have a peri-

cytic niche and localize to perivascular compartments to

support and stabilize blood vessels. A current model pro-

posed by da Silva et al. [29] is that the ASCs-pericytic niche

stabilizes blood vessels and contributes to tissue and

immune system homeostasis under physiological conditions

but is able to assume more of an active tissue regeneration

role upon injury. In our animal model, it appears that the

dsASCs survive the transplant and home to the perivascular

region of the newly formed blood vessels.

ASCs and their paracrine effects

ASCs secrete a variety of growth factors that can exert their

effects on wound repair and tissue regeneration. Some of

these factors include basic fibroblast growth factor, hepa-

tocyte growth factor, insulin growth factor binding protein,

granulocyte colony-stimulating factor, and transforming

growth factor beta 1 [31, 32]. The potential to express such

a vast cytokine profile makes ASCs attractive not only for

their ability to differentiate toward other cells types but

also for their ability to therapeutically affect wound healing

by the secretion of such paracrine factors. It is noteworthy

to mention that such factors also have the ability to mod-

ulate immune and inflammatory responses at the wound

site, recruit local stem cells into the wound microenvi-

ronment, and reduce the amount of apoptotic cells, all

while promoting an angiogenic response [18]. All of these

effects are desirable for creating a microenvironment that is

conducive for tissue regeneration. VEGF is also secreted

by ASCs and is mainly recognized for its ability to promote

vascular endothelial cell proliferation and migration [33].

However, VEGF adds a multitude of benefits to a healing

wound, aside from endothelial cell recruitment, that should

be further considered in a wound healing scenario. Studies

have shown that VEGF is sufficient to induce fibroblast

migration, which is a critical component of the wound-

healing process. VEGF is also known to increase the epi-

thelialization of healing dermal wounds [34]. It is worth

noting that other isoforms of VEGF exist and, in particular,

that VEGF-E also induces keratinocyte migration and

proliferation [35]. It remains to be determined whether

ASCs express all forms of VEGF or just VEGF-A. In our

studies, ASCs expressed VEGF-A transcript when cultured

Fig. 4 Functional blood vessels in the regenerated tissue. a Closer

examination of FPEG-dsASCs treated tissues (day 8) indicated well

formed blood vessels within the regenerating tissue, complete with

endothelium (arrows) and red blood cells within the lumens.

Fibroblast-like cells (dark pink), oriented in perpendicular fashion

to the vessel, appeared throughout the regenerating tissue. b By day

12, collagen layers had been remodeled (blue) into mature bundles,

and function blood vessels were still present (arrow). Original

magnifications: 9400. (Color figure online)

Angiogenesis (2013) 16:745–757 753

123

Fig. 5 von Willebrand Factor (vWF) labeled vessels. To definitively

discern blood vessels in each of the treatment conditions, tissues were

immunolabeled for the vascular endothelial-specific marker vWF

(brown stain) and counterstained for contrast with hematoxylin (blue).

By day 4, both FPEG-dsASCs and FPEG gels alone appeared to have

an initial infiltration of polymorphonuclear cells (dark blue) located

within the upper portion of the gel (hashed oval). By day 8, vWF

expressing endothelium (brown stain) lined the newly formed blood

vessels in the remodeled FPEG-ASCs treated wound and, to a lesser

extent, in the FPEG-alone treated wound. By day 16, a clear difference

was observable in the amount of vessels staining positive for vWF in

the FPEG-ASCs treated group, compared to the FPEG gel alone. The

black hashed line indicates host-gel interface region, with the host

tissue below the line and remodeled gel above. The white hashed line

indicates the newly re-epithelialized region (above) with the treated

regions below. Original magnifications: 9100. (Color figure online)

Fig. 6 Vascular quantification. a Blood vessels were quantitated in

each of the treatment groups by counting the vWF? blood vessels in

sections of the healing tissues. Raw data is presented as the average

number of blood vessels counted per square millimeter in each subject

per treatment (±standard deviation). The number of vWF? vascular

structures increased over a 16-day treatment period. However, FPEG-

dsASCs treatment groups demonstrated a higher density of blood

vessels per square millimeter than FPEG treatments alone. Upon

histological examination, vessels in the FPEG-dsASCs treated groups

(b) also appeared to be larger in size than those in FPEG alone

treatment groups (d) at day 8; this trend continued through day 16 (c,

e, respectively). Special considerations were made to ensure the

accuracy of our blood vessel counts. For example, the center of the

healing wound was chosen for histological analysis to quantify blood

vessels in our model. Peripheral wound blood vessels infiltrate too

quickly to discern a difference in vessel formation, and the center of

the wound is typically the most difficult and final region of a wound to

heal. Therefore, vessel formation in this central area is more relevant

for tissue regeneration purposes than the peripheral vascular-rich

region. Original magnifications: 9400

754 Angiogenesis (2013) 16:745–757

123

in our three-dimensional FPEG gel system, and this

expression increased over time. This observation was also

true for VEGF-A protein levels being secreted into the

culture medium as well, demonstrating that the dsASCs

embedded within the gel have the ability to express VEGF-

A and that this growth factor is able to diffuse from the gel

to increase its concentration in the microenvironment of the

gel system. It remains to be determined whether this

Fig. 7 Localization of human dsASCs in wound model. a To

determine the fate of the transplanted dsASCs as the wound tissue

is remodeled, an antibody specific for human mitochondrial protein

was used to immunolabel tissue sections of FPEG-dsASCs treated

wound beds (Day 16). Human dsASCs were localized to the

surrounding regions of larger blood vessels (brown stain, solid

arrow). Interestingly, human dsASCs did not colocalize to the smaller

capillary-like vessels (hashed oval). To confirm the structures that

dsASCs associated with are indeed blood vessels, tissues were

double-labeled for PECAM-1/CD31 (red stain, hashed arrow).

b Higher magnification clearly demonstrates the red staining of

PECAM-1 along the inner luminal endothelial layer (hashed arrow),

whereas the outer brown (solid arrow) demonstrates the location of

the human dsASCs. c Photomicrograph depicting antibody isotype

controls for the immunostaining procedure. d Bar graph depicting

representative diameter sizes of dsASCs-FPEG gel treatments.

Original magnifications: a, c 9100; b 9400. (Color figure online)

Fig. 8 In vitro VEGF Expression by dsASCs. a Using real-time PCR,

VEGF transcript expression was assessed in dsASCs (P1–P4) seeded

within FPEG gels. Fold increases in transcript expression levels for

VEGF were determined by 2-DDCt method (±standard deviation), and

comparison RNA was obtained from dsASCs that were cultured side

by side under simultaneous conditions in a T25 flask. Interestingly,

when embedded within the FPEG gel, the dsASCs upregulate their

expression levels of VEGF mRNA; this observation translates into

b VEGF protein expression of VEGF as well. Each condition was

performed in triplicate, with a total three separate experiments being

performed (±standard deviation)

Angiogenesis (2013) 16:745–757 755

123

increase in VEGF protein is reflective of increasing num-

bers of ASCs due to proliferation or whether the existing

ASCs are simply expressing a higher level of the gene

encoding VEGF. Either phenomenon is favorable in this

wound-healing scenario. Equally important to VEGF

expression is the angiopoietin system. Indeed, Angpt-1 is

important in vessel stability and has been shown to con-

tribute to the formation of patent vessels of wider diameter

in vivo than control conditions [36, 37]. The synergistic

coordinated expression of VEGF and Angpt-1 by donated

dsASCs in our gel system provides a favorable pro-

angiogenic milieu of growth factors necessary for thera-

peutic angiogenesis to occur in this gel system.

FPEG and wound healing

In the clinic, fibrin has been used as an FDA-approved

hemostatic and sealant agent in a variety of applications.

Additionally, fibrin based hydrogels produced from com-

mercially available purified fibrinogen and thrombin have

been used widely in novel tissue engineering applications,

and include the engineering of such tissues as adipose,

cardiovascular, ocular, muscle, and skin tissues [38]. Such

fibrin hydrogels have also been used to promote angio-

genesis when needed [39]. Our preliminary work has

demonstrated several unique features of FPEG hydrogels

that make it attractive in wound healing over other types of

hydrogel dressings. FPEG exhibits unique features of both

synthetic hydrogels and natural materials. Studies by Ah-

mann et al. [15] showed that the degradation products of

fibrin are bioactive and can enhance vascular smooth

muscle cell proliferation and increase collagen matrix

deposition, both desirable effects in wound healing. In our

studies, the FPEG gels (±dsASCs) appeared to support a

denser and more organized collagen wound bed than saline

controls alone (Fig. 2b, day 16). Finally, fibrin possesses an

inherent biologic capability to encourage wound healing by

stimulating tissue and blood vessel in-growth [16, 40]. In

our studies described here, we observed an increase in the

presence of blood vessels in FPEG-treated wounds over

saline controls (Fig. 3), indicating that the FPEG gels alone

had beneficial effects to the healing wound. Moreover,

blood vessel formation was enhanced in FPEG-dsASCs-

seeded gels over FPEG gels alone, indicating a desirable

effect that is presumably attributed to the presence of the

dsASCs in the gel. It is possible that the enhanced blood

vessel formation in the dsASCs-FPEG treated wounds may

be a direct result of the increased VEGF production by the

dsASCs in the gel. However, since VEGF expression was

assayed in vitro, it is difficult to say with certainty that the

FPEG-dsASCs gel would perform similarly in vivo.

Nonetheless, patent red blood cell-containing vessels were

present in higher numbers than FPEG or saline-treated

wounds. It is likely that in addition to VEGF, the dsASCs

are secreting other biologic factors that influence this

enhanced vascularization. It is also likely that fibrin deg-

radation products may be playing a role in the wound-

healing process, and whether dsASCs are contributing to

the FPEG degradation remains to be determined.

Conclusion

We have successfully isolated adipose-derived stem cells

from discarded human tissue; these cells maintain their

expression profile of key stem cell markers when cultured

in vitro. Our lab has previously differentiated these cells

toward osteocytes and adipocytes, and now pericyte-like

cells. In a rat excision wound model, human dsASCs appear

to integrate with newly formed host tissue and assist to

increase the presence of functional vascular networks in this

region. Future studies will investigate the combinatorial

effect of these cells with the initial stromal vascular fraction

isolated to investigate its potential use in autologous tissue

transplants, wound repair, and skin regeneration.

Acknowledgments Dr. Natesan is supported by a Postdoctoral

Fellowship Grant from the Pittsburgh Tissue Engineering Initiative

(PTEI). Funding for this work was provided by the TATRC Foun-

dation and the Deployment Related Medical Research Program.

Conflict of interest The opinions or assertions contained herein are

the private views of RJC and are not to be construed as official or as

reflecting the views of the Department of the Army or the Department

of Defense.

References

1. Chan RK et al (2012) Development of a vascularized skin con-

struct using adipose-derived stem cells from debrided burned

skin. Stem Cells Int 2012:841203

2. Owens BD et al (2008) Combat wounds in operation Iraqi Free-

dom and operation Enduring Freedom. J Trauma 64(2):295–299

3. Wolf SE et al (2006) Comparison between civilian burns and

combat burns from Operation Iraqi Freedom and Operation

Enduring Freedom. Ann Surg 243(6):786–792; discussion 792-5

4. Brusselaers N et al (2010) Skin replacement in burn wounds.

J Trauma 68(2):490–501

5. Carmeliet P, Jain RK (2000) Angiogenesis in cancer and other

diseases. Nature 407(6801):249–257

6. Rouwkema J, Rivron NC, van Blitterswijk CA (2008) Vascu-

larization in tissue engineering. Trends Biotechnol 26(8):434–441

7. Sahota PS et al (2003) Development of a reconstructed human

skin model for angiogenesis. Wound Repair Regen 11(4):275–284

8. Sorrell JM, Baber MA, Caplan AI (2007) A self-assembled

fibroblast-endothelial cell co-culture system that supports in vitro

vasculogenesis by both human umbilical vein endothelial cells

and human dermal microvascular endothelial cells. Cells Tissues

Organs 186(3):157–168

9. Schechner JS et al (2003) Engraftment of a vascularized human

skin equivalent. FASEB J 17(15):2250–2256

756 Angiogenesis (2013) 16:745–757

123

10. Shepherd BR et al (2006) Vascularization and engraftment of a

human skin substitute using circulating progenitor cell-derived

endothelial cells. FASEB J 20(10):1739–1741

11. Zhang CP, Fu XB (2008) Therapeutic potential of stem cells in

skin repair and regeneration. Chin J Traumatol 11(4):209–221

12. Girandon L et al (2011) In vitro models for adipose tissue engi-

neering with adipose-derived stem cells using different scaffolds

of natural origin. Folia Biol (Praha) 57(2):47–56

13. Zhang G et al (2006) A PEGylated fibrin patch for mesenchymal

stem cell delivery. Tissue Eng 12(1):9–19

14. Liu H, Collins SF, Suggs LJ (2006) Three-dimensional culture for

expansion and differentiation of mouse embryonic stem cells.

Biomaterials 27(36):6004–6014

15. Ahmann KA et al (2010) Fibrin degradation enhances vascular

smooth muscle cell proliferation and matrix deposition in fibrin-

based tissue constructs fabricated in vitro. Tissue Eng Part A

16(10):3261–3270

16. Chalupowicz DG et al (1995) Fibrin II induces endothelial cell

capillary tube formation. J Cell Biol 130(1):207–215

17. Zuk PA et al (2002) Human adipose tissue is a source of multi-

potent stem cells. Mol Biol Cell 13(12):4279–4295

18. Hong SJ, Traktuev DO, March KL (2010) Therapeutic potential

of adipose-derived stem cells in vascular growth and tissue repair.

Curr Opin Organ Transplant 15(1):86–91

19. Kim Y et al (2007) Direct comparison of human mesenchymal

stem cells derived from adipose tissues and bone marrow in

mediating neovascularization in response to vascular ischemia.

Cell Physiol Biochem 20(6):867–876

20. Moon MH et al (2006) Human adipose tissue-derived mesen-

chymal stem cells improve postnatal neovascularization in a

mouse model of hindlimb ischemia. Cell Physiol Biochem

17(5–6):279–290

21. Kondo K et al (2009) Implantation of adipose-derived regener-

ative cells enhances ischemia-induced angiogenesis. Arterioscler

Thromb Vasc Biol 29(1):61–66

22. Meruane MA, Rojas M, Marcelain K (2012) The use of adipose

tissue-derived stem cells within a dermal substitute improves skin

regeneration by increasing neoangiogenesis and collagen syn-

thesis. Plast Reconstr Surg 130(1):53–63

23. Blanton MW et al (2009) Adipose stromal cells and platelet-rich

plasma therapies synergistically increase revascularization during

wound healing. Plast Reconstr Surg 123(2 Suppl):56S–64S

24. Lu F et al (2008) Improved viability of random pattern skin flaps

through the use of adipose-derived stem cells. Plast Reconstr

Surg 121(1):50–58

25. Natesan S et al (2011) Debrided skin as a source of autologous

stem cells for wound repair. Stem Cells 29(8):1219–1230

26. Natesan S et al (2011) A bilayer construct controls adipose-

derived stem cell differentiation into endothelial cells and peri-

cytes without growth factor stimulation. Tissue Eng Part A

17(7–8):941–953

27. Cao Y et al (2005) Human adipose tissue-derived stem cells

differentiate into endothelial cells in vitro and improve postnatal

neovascularization in vivo. Biochem Biophys Res Commun

332(2):370–379

28. Fischer LJ et al (2009) Endothelial differentiation of adipose-

derived stem cells: effects of endothelial cell growth supplement

and shear force. J Surg Res 152(1):157–166

29. da Silva Meirelles L, Caplan AI, Nardi NB (2008) In search of

the in vivo identity of mesenchymal stem cells. Stem Cells

26(9):2287–2299

30. Gimble JM, Katz AJ, Bunnell BA (2007) Adipose-derived stem

cells for regenerative medicine. Circ Res 100(9):1249–1260

31. Rehman J et al (2004) Secretion of angiogenic and antiapoptotic

factors by human adipose stromal cells. Circulation 109(10):

1292–1298

32. Rubina K et al (2009) Adipose stromal cells stimulate angio-

genesis via promoting progenitor cell differentiation, secretion of

angiogenic factors, and enhancing vessel maturation. Tissue Eng

Part A 15(8):2039–2050

33. Song SY, Chung HM, Sung JH (2010) The pivotal role of VEGF

in adipose-derived-stem-cell-mediated regeneration. Expert Opin

Biol Ther 10(11):1529–1537

34. Brem H et al (2009) Mechanism of sustained release of vascular

endothelial growth factor in accelerating experimental diabetic

healing. J Invest Dermatol 129(9):2275–2287

35. Wise LM et al (2012) The vascular endothelial growth factor

(VEGF)-E encoded by orf virus regulates keratinocyte prolifer-

ation and migration and promotes epidermal regeneration. Cell

Microbiol 14(9):1376–1390

36. Kidoya H et al (2008) Spatial and temporal role of the apelin/APJ

system in the caliber size regulation of blood vessels during

angiogenesis. EMBO J 27(3):522–534

37. Chiu LL, Radisic M (2010) Scaffolds with covalently immobi-

lized VEGF and Angiopoietin-1 for vascularization of engineered

tissues. Biomaterials 31(2):226–241

38. Janmey PA, Winer JP, Weisel JW (2009) Fibrin gels and their

clinical and bioengineering applications. J R Soc Interface

6(30):1–10

39. van Hinsbergh VW, Collen A, Koolwijk P (2001) Role of fibrin

matrix in angiogenesis. Ann N Y Acad Sci 936:426–437

40. Lesman A et al (2011) Engineering vessel-like networks within

multicellular fibrin-based constructs. Biomaterials 32(31):7856–

7869

Angiogenesis (2013) 16:745–757 757

123