Synergistic Angiogenesis Promoting Effects of Extracellular Matrix Scaffolds and Adipose-Derived Stem Cells During Wound Repair

Abstract

Slow vascularization rate is considered one of the main drawbacks of scaffolds used in wound healing. Several efforts, including cellular and acellular technologies, have been made to induce vascular growth in scaffolds. However, thus far, there is no established technology for inducing vascular growth. The aim of this study was to promote the vascularization capacities of scaffolds by seeding adipose-derived stem cells (ADSCs) on them and to compare the vascularization capacities of different scaffolds seeded with ADSCs. Two kinds of extracellular matrix scaffolds (small intestinal submucosa [SIS] and acellular dermal matrix [ADM]) and a kind of composite scaffold (collagen–chondroitin sulfate–hyaluronic acid [Co–CS–HA]) were selected. Subcutaneous implantation analysis showed that the vascularization capacity of SIS and ADM was greater than that of Co–CS–HA. ADSCs seeded in SIS and ADM secreted greater amounts of vascular endothelial growth factor than those seeded in Co–CS–HA. In a murine skin injury model, ADSC-seeded scaffolds enhanced the angiogenesis and wound healing rate compared with the nonseeded scaffolds. Moreover, ADSC-SIS and ADSC-ADM had greater vascularization capacity than that of ADSC-Co–CS–HA. Taken together, these results suggest that ADSCs could be used as a cell source to promote the vascularization capacities of scaffolds. The vascularization capacities of ADSC-seeded scaffolds were influenced by both the vascularization capacities of the scaffolds themselves and their effects on the angiogenic potential of ADSCs; the combination of extracellular matrix scaffolds and ADSCs exhibited synergistic angiogenesis promoting effects.

Introduction

Three-dimensional scaffolds are presently used as dermal regeneration templates for the treatment of full-thickness skin defects resulting from injuries and severe burns. Unfortunately, the clinical benefit of these scaffolds has been poorer than expected due to their low regenerative capacity and high infection risk for the patients.¹ The low vascularization, which limits the supply of immune cells, nutrients, and oxygen to the wound area, is believed to be the main reason for these problems.²

Many strategies have been employed to enhance the vascularization capacities of scaffolds. The simplest method is to incorporate angiogenic growth factors such as vascular endothelial growth factor (VEGF),³ basic fibroblast growth factor (bFGF),⁴ and angiogenin⁵; however, this method has several drawbacks, such as the sensitivity of growth factors to thermal processing and exposure to chemical solvents,⁶ and the short half-life of growth factors in vivo.⁷ Moreover, most studies have used single growth factors, which may not be sufficient to generate a strong effect on angiogenesis.⁸ Endothelial cells have also been used to deal with this problem^9,10; however, this procedure has not been included in clinical trials due to the difficulties in culture expansion techniques and the limited number of cells for implantation.⁸

Seeding stem cells, which have the ability to enhance angiogenesis in scaffolds, is being considered as a promising strategy to address this problem.^2,11,12 ADSCs have been shown to differentiate into endothelial cells,^13,14 enhance neovascularization in ischemic hindlimb model,^15,16 and secrete angiogenic growth factors,^17,18 suggesting the potential use for these cells in therapeutic vascularization and tissue engineering of vascularized constructs. Moreover, ADSCs can be isolated from a small volume of adipose tissue and expanded in vitro using standard cell culture technologies.^19,20 These findings indicate that ADSCs could be used as a cell source to promote the vascularization capacities of scaffolds.

Extracellular matrix (ECM) scaffolds have been successfully used for wound repair in both preclinical animal studies and human clinical applications.²¹ The ECM is secreted by the resident cells of each tissue and organ from which it is prepared. Therefore, the composition and distribution of the ECM scaffold constituents vary depending on the tissue source. Moreover, the preparation of ECM scaffolds requires several processing steps, including decellularization, hydration, dehydration, and sterilization; these steps can affect both the structure and type of host response these scaffolds elicit when used for tissue engineering.²²

Here, we selected two kinds of ECM scaffolds (small intestinal submucosa [SIS] and acellular dermal matrix [ADM]) derived from porcine, along with a composite scaffold, namely, collagen–chondroitin sulfate–hyaluronic acid (Co–CS–HA),²³ consisting of Co, CS, and HA; this scaffold was developed to imitate the ingredients and their ratios in the natural dermal matrix. These scaffolds may have different vascularization capacities in vivo. Moreover, when ADSCs are seeded in these scaffolds, they not only serve as cell carriers^24,25 providing mechanical support but also facilitate cell-scaffold interactions, which actively influence the cellular responses, including proliferation, differentiation,²⁶ apoptosis,²⁷ and angiogenic growth factor secretion by stem cells.²⁸ Therefore, we hypothesized that the combination of ADSCs and different scaffolds would produce different effects on angiogenesis, owing to both the vascularization capacities of the different scaffolds and their influence on the angiogenic potential of ADSCs.

The aim of this study was to promote the vascularization capacities of SIS, ADM, and Co–CS–HA by seeding ADSCs on them and to compare the vascularization capacities of these ADSC-seeded scaffolds by using a murine skin injury model.

Materials and Methods

Preparation of scaffolds

Small intestinal submucosa

SIS was developed as previously described.²⁹ Briefly, a segment of fresh porcine jejunum was obtained from a local slaughterhouse. After carefully washing in water, the tunica mucosa, the serosa, and tunica muscularis of the segment were mechanically removed. The remaining intestinal submucosa tube was slit longitudinally and sectioned in a length of approximately 10 cm. The obtained SIS was thoroughly rinsed in a saline solution to remove the resident cells. The SIS sheets were sterilized in 0.1% peracetic acid. Finally, the resultant SIS was vacuum-sealed into hermetic packaging and stored at 4°C for future use.

Acellular dermal matrix

ADM was developed from porcine skin as previously described.³⁰ Fresh porcine skin was obtained from a local slaughterhouse. After complete cleaning, excision of the subdermal fat tissue and the epidermis, the resulting skin was treated with a 0.25% trypsin solution at 37°C for 1.5 h and then extensively washed with distilled water. Subsequently, the dermal matrix was incubated in 1 M sodium hydroxide solution at room temperature for 16 h and thoroughly rinsed in phosphate-buffered saline (PBS) at room temperature with continuous shaking until the pH value of the entire tissue sample became neutral. ADM was lyophilized and vacuum-sealed into hermetic packaging and stored at 4°C for future use.

Collagen-chondroitin sulfate-hyaluronic acid

Co–CS–HA was developed as previously described.²³ Bovine tendon Co I, CS, HA, 2-(N-morpholino)ethane-sulfonicacid, N-hydroxysuccinimide, and 1-ethyl-3-3-dimethylaminopropylcarbodiimide hydrochloride (EDC) were all purchased from Sigma Chemical (Badlapur). Co I was dissolved at 4°C at a concentration of 12.5 mg/ml in a solution of 0.05 M acetic acid. CS and HA were dissolved at 4°C at a concentration of 12.5 mg/ml in a solution of double-distilled water, respectively. The pH of Co I was adjusted to 7.4 at 4°C. CS was added in Co I solution before HA was added to it. The ratio of the three elements (V/V/V) was 9:1:1. The elements were added at a speed of 0.5 ml/min. After being well mixed in culture dish with a glass rod, the slurry was poured into a 6-well plate and was frozen at –80°C for 3 h and then they were lyophilized. These meshes were subsequently cross-linked for 24 h at room temperature using 40% ethanol-water (pH 5.5) solution supplemented with 50 mM 2-(N-morpholino)ethane-sulfonicacid, 5 mM EDC, and 5 mM N-hydroxysuccinimide. Then, the cross-linked membranes were rinsed twice for 1 h with 0.1 M disodium phosphate, twice for 2 h with 1 M sodium chloride, six times for 24 h with 2 M sodium chloride, and ten times with double-distilled water to remove residual EDC. After being frozen again at –80°C for 3 h, membranes were lyophilized and vacuum-sealed into hermetic packaging and stored at 4°C for future use.

Morphology examination of scaffolds

Scaffolds were dehydrated by treatment with a series of grade ethanol solution (50% for 12 h, 75%, 85%, and 95% each for 2 h), and then they were placed overnight in a vacuum oven at room temperature before being coated with gold for scanning electron microscope (SEM; Hitachi S-3400N). Paraffin sections of the scaffolds were stained with hematoxylin and eosin (H&E) and visualized in a bright field using a microscope (BX-51; Olympus).

Detection of nuclear remnants in scaffolds

Cellular SIS (CSIS), cellular dermal matrix (CDM), and rehydrated scaffolds were embedded in tissue freezing medium optimum cutting temperature (OCT) (Leica) and immediately frozen in freezing microtome (CM1900, Leica) at −20°C. Frozen sections were stained with propidium iodide (PI) (500 nM) (Sigma), and the image collection was processed by a fluorescence microscope (IX71; Olympus).

Vascularization and biocompatibility of the scaffolds

All the animal experiments in this study were conducted according to the committee guidelines of the Fourth Military Medical University for animal experiments, which met the NIH guidelines for the care and use of laboratory animals. Thirty-five 8-week-old C57BL/6 female mice were divided into five groups: CSIS, CDM, SIS, ADM, and Co–CS–HA. For induction of anesthesia, the mice were subsequently placed prone on the operating table and connected to a circuit delivering 3% inhalational methoxyflurane (mixed with oxygen). A constant directed flow of 1% inhalational methoxyflurane was used for maintenance anesthesia. SIS, ADM, and Co–CS–HA were sterilized by ⁶⁰CO irradiation. CSIS, CDM, and these scaffolds were placed in dorsal midline subcutaneous pocked. At 7, 21, and 28 days postsurgery, the transplanted scaffolds were taken out and were immediately fixed with 4% phosphate-buffered formalin. Scaffolds transplanted for 7 and 21 days were prepared for histological analysis using H&E staining, and scaffolds transplanted for 28 days were used for immunohistochemical analysis. The primary antibodies included polyclonal rabbit anti-CD4, monoclonal rabbit anti-CD8, and monoclonal rat anti-CD68 (all from Abcam). Biotinylated secondary antibodies (Dako) were used. All the samples were examined under a microscope (BX-51; Olympus).

Isolation and culture of ADSCs

Three-week-old C57BL/6-green fluorescent protein (GFP) transgenic mice were gifted from the Institute of Neuroscience of the Fourth Military Medical University. Adipose tissue was obtained from the inguinal region of the mice and extensively washed with 10 ml PBS. The ECM was then digested with 0.075% (w/v) collagenase type I (Sigma-Aldrich) at 37°C for 1 h. After centrifugation, the supernatant was discarded, and the pellet was resuspended and filtered through a 100-μm cell strainer to remove undigested tissue fragments. The suspension was then centrifuged again and resuspended in 10 ml of α-minimum essential medium (MEM) (Gibco) containing 10% fetal bovine serum (FBS) (Gibco), 0.292 mg/ml glutamine, 100 units/ml penicillin, and 100 μg/ml streptomycin (all from Sigma). This suspension was placed into T75 flasks and allowed to incubate at 37°C in a humidified chamber containing 5% CO₂ for 24 h, after which the nonadherent cells and debris were removed by aspiration. Adherent cells were then culture expanded with media changes at 2-day intervals. Cells at passage 3 were used for all experiments.

Characterization of ADSCs

Flow cytometry analysis

The phenotype of cultured ADSCs was evaluated by flow cytometry. Approximately 5 × 10⁵ cells were harvested by trypsin, washed twice with PBS, and incubated with phycoerythrin (PE)-conjugated rat anti-CD44, CD90, CD45 (all from Biolegend), unconjugated rat anti-CD29, CD105, and CD34 (all from Abcam), respectively. fluorescein isothiocyanate-conjugated goat anti-rat IgG secondary antibodies (Abcam) were used. Phycoerythrin (PE)-conjugated and unconjugated isotype-matching IgGs were used as controls. Cells were then analyzed on Elite ESP flow cytometry (Beckman Coulter).

Adipogenic differentiation assays

A total of 2 × 10⁵ ADSCs were seeded into each well of a six-well plate. ADSCs were cultured in a-MEM containing 10% FBS, 0.292 mg/ml glutamine, 100 units/ml penicillin, 100 μg/ml streptomycin, 2 μM insulin (Sigma-Aldrich), 0.5 mM isobutyl-methylxanthine (Sigma-Aldrich), and 10 nM dexamethasone (Sigma-Aldrich). The medium was changed every two days. After 14 days of culture, the cells were washed thrice in PBS after being fixed in 4% paraformaldehyde and then incubated in 0.3% Oil Red O (Sigma-Aldrich) solution for 15 min. After being washed thrice in PBS, cells were routinely observed and photographed under a phase-contrast inverted microscope.

Osteogenic differentiation assays

A total of 2 × 10⁵ ADSCs were seeded into each well of a six-well plate. ADSCs were cultured in a-MEM containing 10% FBS, 0.292 mg/ml glutamine, 100 units/ml penicillin, 100 μg/ml streptomycin, 5 mM _L-glycerophosphate (Sigma-Aldrich), 100 nM dexamethasone (Sigma-Aldrich), and 50 μg/ml ascorbic acid. The medium was changed every two days. After 21 days of culture, cells were washed twice in PBS after being fixed in 4% paraformaldehyde and then incubated in 0.1% alizarin red solution (Sigma-Aldrich) in Tris–HCl (pH 8.3) at 37°C for 30 min. After being washed twice in PBS, cells were routinely observed and photographed under an inverted microscope.

Preparation of stem cell-seeded scaffolds in vitro

Before cell seeding, the ADM, SIS, and Co–CS–HA were sterilized by ⁶⁰CO irradiation and cut into 10 mm in diameter and 0.4–0.6 mm in height. The scaffolds were individually rehydrated in culture media. Scaffolds were covered with 100 μl growth medium alone in the scaffold groups and with an equal volume of cell suspension containing 1 × 10⁵ ADSCs in the ADSC-scaffold groups. After 1 h of incubation, 1 ml of growth medium was added into each well. Scaffolds were incubated under standard culture conditions for 24 h, after which the overlying medium or cell suspension was aspirated. The scaffolds were flipped to place the opposite surface facing up, and the corresponding medium or cell suspension solution was placed on the other side. Scaffolds were then incubated for 24 h. The ADSC-seeded scaffolds were used for SEM, H&E staining, and transplantation.

Effect of scaffolds on the proliferation and apoptosis of ADSCs

Metabolic activity assay

Metabolic activity was quantified using a 3-(4,5-dimethylthizazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfonyl)-2H-tetrazolium (MTS, CellTiter 96^TM Aqueous; Promega) assay. ADSCs were seeded on scaffolds at a concentration of 5 × 10⁴ cells/cm² in a 96-well plate. At 1, 3, 5, and 7 days, 100 μl of FBS-free culture medium and 20 μl of MTS reagent were added into every specimen. After incubation for 2 h in a CO₂ incubator, all the reactant mixtures were extracted and added into 96-well plates. Then, the absorbance of each specimen in the 96-well plates was measured in a micro-plate reader at a wave length of 490 nm. Background absorbance was corrected by subtracting the absorbance index of culture medium from the specimen data.

Flow cytometry analysis

After being seeded onto SIS, ADM, and Co–CS–HA, respectively, at a concentration of 2 × 10⁴ cells/cm² and cultured for 72 h in a 24-well plate, cells were trypsinized as previously described; ³¹ and cell precipitates were washed twice with PBS and resuspended in 1 ml of physiological saline by repeated vibration to ensure a single-cell suspension. Then, 2 ml of cold dehydrated alcohol was quickly mixed with the cell suspension to fix cells at 4°C for 24 h. Finally, the cells were washed twice again with PBS, stained with 100 mg/ml propidium iodide at 4°C for 30 min, and subjected to cell-cycle analysis using Elite ESP flow cytometry (Beckman Coulter). For apoptosis analysis, single-cell suspension of ADSCs was washed twice with cold PBS, incubated for 15 min with fluorescein-conjugated annexin V and PI, and analyzed using the same flow cytometer.

Effect of scaffolds on the angiogenic growth factors secretion by ADSCs

Real-time polymerase chain reaction analysis

After being seeded onto SIS, ADM, and Co–CS–HA, respectively, at a concentration of 2 × 10⁴ cells/cm² and cultured for 72 h in a 24-well plate, ADSCs were trypsinized and were used for RNA extraction. Total cellular RNA was exacted by using the TRIzol Reagent (Invitrogen Life Technology), and first-strand cDNA synthesis was performed according to the manufacturer's protocol. Real-time polymerase chain reaction products were detected with SYBR Green dye by using Light Cycler Instrument (Toyobo). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was amplified as internal control. Primer sequences were as follows: VEGF: 5′-primer (5′-AGAGCAACATCACCATGCAG-3′) and 3′-primer (5′-CAGTGAACGCTCCAGGATTT-3′); hepatocyte growth factor (HGF): 5′-primer (5′-GTGGATGCCAAGCCAAGCT-3′) and 3′-primer (5′-CAGTAGGGTGGATGGTTAGTTTGAA-3′); bFGF: 5′-primer (5′-CACCAGGCCACTTCAAGGA-3′) and 3′-primer (5′-GATGGATGCGCAGGAAGAA-3′); GAPDH: 5′-primer (5′-ATCATCCCTGCATCCACT-3′) and 3′-primer (5′-ATCCACGACGGACACATT-3′). We calculated expression levels by the comparative C_T method using GAPDH as an endogenous reference gene.

Enzyme-linked immunosorbent assay analysis

SIS, ADM, and Co–CS–HA were seeded with ADSCs, respectively, at a concentration of 2 × 10⁴ cells/cm² and cultured in a 24-well plate for 6 days. Scaffolds without ADSCs were also cultured as control. Every 48 h, the conditioned medium of ADSC-seeded scaffolds and empty scaffolds was collected, respectively, and assayed using commercially available sandwich enzyme-linked immunosorbent assays (R&D Systems) for mouse VEGF. Then, fresh medium was added. At the same time, ADSCs were trypsinized, and cell counts were performed with a hemacytometer. VEGF levels were normalized to the number of cells at the time of harvest, expressed as pg per 10⁶ cells.

Full-thickness cutaneous wound model

A total of forty-nine 8-week-old female C57BL/6 mice were randomized to seven treatment groups: control, only scaffolds implantation (3 groups), and scaffolds seeded with ADSCs (3 groups). Each mouse was anesthetized as previously described, and a 1-cm diameter punch biopsy instrument was placed with moderate force onto the dorsum of the mouse to create an impression of the circumference. Next, the middle of the outlined region of skin was grasped with forceps, and the 1 cm diameter region was sharply excised along the outline with a pair of scissors. The excised tissue was full-thickness skin in depth, leaving subcutaneous dorsal muscle exposed after excision. In the scaffold and ADSC-scaffold groups, the grafts were removed from PBS and were placed into the dorsal wound. The wound was then covered by two layers of vaseline gauze with discontinuous suture onto the marginal recipient mice skin of the defect area by 3-0 silk suture. In the control group, the wound was just covered by two layers of vaseline gauze and discontinuously sutured.

Wound healing rate

To determine the rate of wound healing in wound area, the wounds were imaged by digital camera after surgery at 0–3 weeks. Photographs were uploaded to an appropriate computer platform and were analyzed using the Image J image analysis software (NIH image software). All photographs were taken with the experimental mouse placed adjacent to a metric ruler that was used for distance calibration and standardization, allowing subsequent quantitative analysis. The percentage of wound closure was calculated as follows: (area of original wound–area of actual wound)/area of original wound × 100%.

Gross study of wound bed cutaneous vascular infiltration

Animals were euthanized at postoperative day 21, and approximately 3 × 3 cm, full-thickness cutaneous biopsies of the wound repair bed and surrounding tissue were obtained. Tissue specimens were carefully placed on the bottom of a polystyrene cell culture dish (Corning Enterprises) and spread with moderate tension along the plate bottom, with the superficial surface facing up. Next, standard incandescent illumination was immediately directed under the specimen into the polystyrene dish, illuminating the vascular infiltration of wound bed biopsy specimens. Photographs were taken with a digital camera. After visualization of the vessels, tissue specimens were divided into two parts. One part was placed in 4% phosphate-buffered formalin and prepared for staining with H&E. The second part was embedded in tissue freezing medium optimum cutting temperature (OCT) (Leica) and immediately frozen in freezing microtome (CM1900; Leica) at −20°C to investigate the immunofluorescent analysis.

Immunofluorescent analysis and measurement of microvessel density

GFP-positive ADSCs were identified with a polyclonal rabbit-anti-GFP primary antibody (Abcam); endothelial cells were identified with a monoclonal rat-anti-CD31 primary antibody (Abcam). Alexa Fluor 488 goat anti-rabbit IgG secondary antibody (Invitrogen) and Fluor 594 goat anti-rat IgG secondary antibody (Invitrogen) were used. Hoechst 33342 dye was used to stain nuclei. The signals were examined under a laser scanning confocal microscope (FV-1000; Olympus). The number of microvessels was counted. In each section, three high-power fields were randomly selected and photographed, and microvessels were counted in each field. Vessels with a diameter of <50 μm were counted.²⁵ The number of microvessels was averaged, and vascular density was expressed as microvessel density (microvessels/high-power field).

Statistical analysis

Data are expressed as mean ± standard deviation. Analysis was performed using the Statistical Program for Social Science for Windows. Comparisons of multiple groups were done with ANOVA with corrections for multiple comparisons. The effect of interaction between scaffold and ADSCs was done with factorial design ANOVA. p-value <0.05 was considered statistically significant.

Results

Characteristics of the scaffolds

SEM micrographs showed that the ADM, SIS, and Co–CS–HA scaffolds had uniform and widely interconnected pores (Fig. 1a–c). H&E staining showed that there was no cellular component, cellular debris, skin accessory, or blood vessel in SIS, ADM, or Co–CS–HA scaffolds (Fig. 1d–f). This was confirmed by PI staining, which showed that compared with positive control (Fig. 1g, h), there were no nuclear remnants in SIS, ADM, or Co–CS–HA scaffolds (Fig. 1i–k). These indicated that the cells were completely removed from the tissue.

FIG. 1.

Characteristics of small intestinal submucosa (SIS), acellular dermal matrix (ADM), and collagen–chondroitin sulfate–hyaluronic acid (Co–CS–HA). (A): Scanning electron microscope (SEM) photographs (a–c) and histological photographs (hematoxylin and eosin [H&E] staining) (d–f) of the scaffolds. Scale bars represent 50 μm. (B): propidium iodide staining of the scaffolds. Compared with positive control (g, h), there were no nuclear remnants in SIS, ADM, or Co–CS–HA scaffolds (i–k). Scale bars represent 20 μm. Color images available online at www.liebertonline.com/ten.

Vascularization and biocompatibility of the scaffolds

H&E staining of the excised tissue revealed that the scaffolds had been repopulated with mouse cells with no evidence of severe acute inflammatory response by the host animal at day 7 and day 21 after subcutaneous transplantation (Fig. 2A). There was evidence of vascularization of the tissue. At day 21, there were more new blood vessels in seeded SIS and ADM than in Co–CS–HA. Inflammatory cell infiltration into the grafts was evaluated using anti-CD4 (T cells), anti-CD8 (T cells), and anti-CD68 (macrophages) staining of the sections from 28-day explants (Fig. 2B). Compared with the staining in the positive control (CSIS and CDM), the transplanted scaffolds contained minimal positive staining for CD4+ T cells, CD8+ T cells, and CD68+ macrophages.

FIG. 2.

Vascularization and biocompatibility of SIS, ADM, and Co–CS–HA. (A): Vascularization of SIS, ADM, and Co–CS–HA (a–f). At day 21, there were more new blood vessels in seeded SIS (d) and ADM (e) than in Co–CS–HA (f). The black arrows represent location of new blood vessels. (B): Inflammatory cell infiltration into CSIS, CDM, SIS, ADM, and Co–CS–HA at 28 days (g–u). Immunostaining for CD4 (g–k), CD8 (l–p), and CD68 (q–u) was performed. Scale bars represent 50 μm. Color images available online at www.liebertonline.com/ten.

Characterization of ADSCs

Flow cytometry analysis of ADSCs showed that the third passage ADSCs were negative for hematopoietic lineage markers CD45 and CD34 and were positive for CD29, CD44, CD90, and CD105 (Fig. 3a–f). These results were consistent with those of a previous study.¹⁹ ADSCs differentiated into adipocytes when cultured in appropriate differentiation media; the adipocytes formed Oil red O-positive lipid clusters (Fig. 3g)—osteoblasts—as indicated by alizarin red staining (Fig. 3h).

FIG. 3.

Characterization of adipose-derived stem cells (ADSCs). (A): Flow cytometry analysis of ADSCs (a-f). (B): Differentiation of ADSCs. ADSCs differentiated into adipocytes and osteoblasts when cultured in appropriate differentiation media. Adipocytes and osteoblasts were indicated by Oil red O (g) and alizarin red (h) staining. Scale bars represent 100 μm. Color images available online at www.liebertonline.com/ten.

Effects of the scaffolds on the morphology and proliferation of ADSCs

SEM micrographs showed that ADSCs actively migrated along the scaffold surface in three-dimensional fashion (Fig. 4a–c). H&E staining showed that the ADSCs adhered well and were evenly distributed in the interior of the scaffolds (Fig. 4d–f). None of the scaffolds induced significant cell apoptosis (Fig. 4g–i). The percentage of ADSCs in the S + G2 M phases was higher when they were cultured on SIS and ADM than when they were cultured on Co–CS–HA (Fig. 4j–l). In order to compare the influence of different scaffolds on the proliferation of ADSCs, the growth curve of ADSCs on the three different scaffolds was investigated by the MTS assay (Fig. 4C). There was no statistically significant difference in the cell numbers across the scaffolds at day 1. From day 3 to day 7, the proliferation of ADSCs was faster in SIS and ADM than in Co–CS–HA; there was no statistically significant difference in the proliferation rate between SIS and ADM.

FIG. 4.

Effects of the scaffolds on the morphology, apoptosis, and proliferation of ADSCs. (A): SEM photographs of ADSC-seeded scaffolds (a–c); histological photographs of ADSC-seeded scaffolds (d–f). (B): Flow cytometry analysis. None of these scaffolds induced significant cell apoptosis (g–i). The percentage of ADSCs in the S + G2 M phases was higher when they were cultured on SIS (49.7%) and ADM (48.3%) than when they were cultured on Co–CS–HA (20.75%) (j–l). (C): Assessment of cellular viability via MTS assay at 1, 3, 5, and 7 days. *p < 0.05. Scale bars represent 50 μm. Color images available online at www.liebertonline.com/ten.

Effects of scaffolds on the angiogenic growth factor secretion by ADSCs

Real-time polymerase chain reaction analysis of the expression levels of growth factors in the different scaffolds revealed that after 72 h of culture, the ADSCs cultured on SIS expressed the greatest amount of VEGF, followed by those cultured on ADM and Co–CS–HA (Fig. 5A). However, the expression levels of bFGF and HGF were not statistically different across the three scaffolds. To examine the protein levels of VEGF secreted by ADSCs, we performed enzyme-linked immunosorbent assay (Fig. 5B), which revealed that at 2 and 4 days, the ADSCs cultured on SIS secreted the greatest amount of VEGF, followed by those cultured on ADM and Co–CS–HA. At 6 days, the amount of VEGF secreted by ADSCs cultured on SIS and ADM was not significantly different. The results of conditional media of empty scaffolds were negative, which indicated that there was no significant cross-reactivity between the antibodies and the growth factors released from scaffolds and excluded the possibility that the total amount of growth factor measured in the conditional media of the ADSC-seeded scaffolds represented the sum of the growth factors released from the scaffolds plus the VEGF secreted by the cells.

FIG. 5.

Angiogenic growth factors secretion of ADSCs cultured on different scaffolds. (A): Real-time polymerase chain reaction analysis of expression levels of growth factors. ADSCs cultured on SIS expressed the significantly greatest amount of vascular endothelial growth factor (VEGF), followed by those cultured on ADM and Co–CS–HA. (B): Enzyme-linked immunosorbent assay analysis. At 2 and 4 days, ADSCs cultured on SIS secreted the significantly greatest amount of VEGF, followed by those cultured on ADM and Co–CS–HA. At 6 days, ADSCs cultured on SIS and ADM secreted greater amounts of VEGF than those cultured on Co–CS–HA. **p < 0.01.

Effects of scaffolds with or without ADSCs on angiogenesis in wound repair

Mice skin is thin and semitransparent, which allows macroscopic visualization of blood vessels. Close inspection of the images obtained under intense illumination revealed an apparently robust invasion of vascular tissue in the ADSC-scaffold groups compared with that in the other groups (Fig. 6a–g). Further, the blood vessels of the ADSC-SIS and ADSC-ADM groups were more robust than those of the ADSC-Co–CS–HA group; the blood vessels of the SIS and ADM groups were more robust than those of the Co–CS–HA group. The robustness of the blood vessels was in accordance with the microvessel densities (Fig. 6B), which were morphometrically assessed after immunohistochemical staining for CD31 (Fig. 6h–n). Microvessel density was significantly greater in the ADSC-scaffold groups than in the scaffold groups. Moreover, microvessel densities in the ADSC-SIS and ADSC-ADM groups were significantly greater than those in the ADSC-Co–CS–HA group. Microvessel densities in the SIS and ADM groups were significantly greater than those in the Co–CS–HA group. This was consistent with the results of subcutaneous implantation (Fig. 2A). The effect of interaction between the scaffolds and ADSCs was statistically significant. ADSCs generated greater angiogenesis promoting effect when they were seeded in SIS and ADM than when they were seeded in Co–CS–HA.

FIG. 6.

Effects of scaffolds with or without ADSCs on angiogenesis in wound repair. (A): Blood vessel visualization and immunofluorescent staining for CD31. After 3 weeks of transplantation, the skin of animals was removed and illuminated to analyze neovascularization (a–g). Scale bars represent 1 cm. Microvessel densities were morphometrically assessed after immunohistochemical staining for CD31 (red) (h–n). Nuclei were stained with Hoechst (blue). Scale bars represent 20 μm. (B): Microvessel density analysis. *p < 0.05, **p < 0.01. Color images available online at www.liebertonline.com/ten.

Survival and differentiation of ADSCs in vivo

At 3 weeks after transplantation, we were able to use immunofluorescence to analyze the survival of ADSCs, because the cells expressed GFP. Confocal microscopy showed that GFP-positive (green) ADSCs appeared in the ADSC-SIS (Fig. 7a), ADSC-ADM (b), and ADSC-Co–CS–HA (c) groups but not in the SIS (d), ADM (e), Co–CS–HA (f), or no-scaffold control (g) groups. Only a small proportion of the cells in the tissues were GFP-positive. Moreover, few ADSCs showed co-localization of GFP with CD31 staining (red) in the ADSC-SIS (h), ADSC-ADM (i), and ADSC-Co–CS–HA (j) groups. This indicated that few ADSCs spontaneously differentiated into a vascular endothelial phenotype.

FIG. 7.

Survival and differentiation of implanted ADSCs in vivo after 3 weeks. A small amount of green fluorescent protein-positive ADSCs (green) appeared in the ADSC-SIS (a), ADSC-ADM (b), and ADSC-Co–CS–HA (c) groups but not in the SIS (d), ADM (e), Co–CS–HA (f), or no-scaffold control (g) groups. Few green fluorescent protein-positive ADSCs co-localized with stain for CD31 (red) (h–j). Nuclei were stained with Hoechst (blue). Scale bars represent 20 μm. Color images available online at www.liebertonline.com/ten.

Wound healing rate

Analysis of wound healing rates was defined as the gross epithelialization of the wound bed. Photographs of the wound region were obtained on weeks 0, 1, 2, and 3 after the creation of excisional wounds (Fig. 8A). The photographs were analyzed using the Image J image analysis software. Closure of wounds in the ADSC-scaffold groups was significantly greater than that in the scaffold groups at week 1. There were no significant differences in the wound healing rates between the ADSC-scaffold groups and the nonseeded scaffold groups. The difference in the wound healing rate across the seven groups diminished over time. At 3 weeks, wounds in all the groups, except the control group, were healed.

FIG. 8.

Wound healing rate. (A): Photographs of wound regions were taken on weeks 0, 1, 2, and 3 after the creation of excisional wounds. (B): Analysis of wound healing rates. *p < 0.05. Color images available online at www.liebertonline.com/ten.

Histology analysis

H&E staining of the wounded skins was performed at 21 days after surgery (Fig. 9). No significant difference in the skin structure was found between the different groups, except that re-epithelialization remained incomplete in the control group, but the wounds in other groups healed completely. These results indicated that ADSC-seeded scaffolds accelerate wound healing without an abnormal wound healing process, such as the formation of granulation or epidermal hyperplasia. The presence of microvessels, as evidenced by consistent luminal morphology and infiltration with erythrocytes, was also observed. The microvessel density in the ADSC-SIS and ADSC-ADM groups was greater than that in the ADSC-Co–CS–HA group. These results were consistent with wound bed cutaneous vascular infiltration and subsequent immunofluorescent findings (Fig. 6). Further, there were no evident inflammatory infiltrates, indicating excellent biocompatibility of the scaffolds. This was consistent with the results of subcutaneous implantation (Fig. 2). Traces of the scaffold could hardly be identified in all the groups, because the scaffold shared the integrity of the newly formed dermal layer, indicating the good biodegradation of the three kinds of scaffolds.

FIG. 9.

Histological images of experimental and control groups after transplantation for 3 weeks. Re-epithelialization was still incomplete in the control group (k, n), but the wounds in other groups had completely healed (a-j, l, and m). The black arrows represent location of new blood vessels. Scale bars represent 100 μm. Color images available online at www.liebertonline.com/ten.

Discussion

In this study, we attempted to promote the vascularization capacities of SIS, ADM, and Co–CS–HA by seeding ADSCs on them and compared the vascularization capacities of these ADSC-seeded scaffolds. Our findings demonstrated that ADSC-seeded scaffolds had greater vascularization capacities than the scaffolds without ADSCs. Moreover, compared with Co–CS–HA, SIS and ADM themselves promoted not only angiogenesis but also the angiogenic factor secretion ability of the ADSCs; the combination of these two ECM scaffolds and ADSCs produced synergistic angiogenesis promoting effects during wound repair.

Angiogenesis, or neovascularization, is defined as the growth of blood vessels and is mainly regulated by various growth factors, including VEGF, bFGF, HGF, and transforming growth factor-β.³² Of these growth factors, VEGF, which induces endothelial cell migration and proliferation, is considered the most potent angiogenic growth factor.⁶ After 3 weeks of subcutaneous transplantation, there were more new blood vessels within the implanted SIS and ADM than within the implanted Co–CS–HA; this indicated that SIS and ADM had greater vascularization capacities than those of Co–CS–HA. These results were consistent with those of the previous studies, which showed that SIS expressed bFGF, transforming growth factor-β, and VEGF²²; supported the adherence of microvascular endothelial cells³³; and promoted angiogenesis,²⁹ and that ADM supported vascular in-growth in vivo.³⁴ The greater vascularization capacities of SIS and ADM may be attributed to the angiogenic growth factors that are released by them during scaffold degradation. In addition, ECM is custom designed and manufactured by the resident cells, which is critical for cellular responses to signals in the cellular microenvironment and provides a supportive medium or conduit for the formation of blood vessels.²² We also detected the inflammatory cell infiltration within the implanted scaffolds. Minimal positive staining for CD4-, CD8-, and CD68-positive cells existed in the implanted SIS, ADM, and Co–CS–HA after single 28-day transplantation. These results indicated that all the scaffolds had good biocompatibility.

We also evaluated the influence of different scaffolds on the proliferation, apoptosis, and ability of secreting angiogenic growth factor of ADSCs. ADSCs cultured in SIS and ADM proliferated faster than those cultured in Co–CS–HA. None of the scaffolds induced significant cell apoptosis. ADSCs seeded in SIS and ADM secreted greater amounts of VEGF than those seeded in Co–CS–HA. These findings indicated that unlike Co–CS–HA, SIS and ADM enhanced the angiogenic efficacy of ADSCs. Previous studies showed that transfected cells, which overexpressed angiogenic growth factors, could be seeded in scaffolds to promote vascularization.^6,35 Our results suggested that selecting suitable scaffolds, which could promote angiogenic growth factor secretion by seeded cells, may avoid the need of transfection and fulfill the vascularization requirement. The different effects on ADSCs may be caused by the different growth factors in SIS and ADM. A previous study showed that bFGF could enhance the therapeutic angiogenic efficacy of human adipose-derived stromal cells by improving the survival of transplanted cells and secretion of angiogenic factors.³⁶ VEGF was also shown to enhance cell proliferation and angiogenesis.³⁷ Moreover, the structurally and functionally different molecules in the ECM scaffolds may support the proliferation and secretion of growth factors by the seeded cells.²¹ Further studies are required to investigate the mechanism underlying this effect.

At 3 weeks after transplantation, the ADSC-seeded scaffolds enhanced vascularization levels compared with the unseeded scaffolds. These results were similar to the results of previous studies performed using other kinds of stem cells.^2,11,12 This indicated that ADSCs could be used as a cell source to enhance vascularization capacities of scaffolds. In addition, ADSC-SIS and ADSC-ADM had greater vascularization capacities than ADSC-Co–CS–HA. In the ADSC-SIS and ADSC-ADM groups, the angiogenesis promoting effect in wound repair was thought to be caused by both ADSCs and scaffolds, because our results indicated that compared with Co–CS–HA, SIS and ADM themselves not only enhanced angiogenesis but also promoted VEGF secretion by ADSCs. We found that the effect of interaction between scaffolds and ADSCs existed; ADSCs generated greater angiogenesis promoting effect on wound healing when they were seeded in SIS and ADM than when they were seeded in Co–CS–HA. Therefore, the combination of these two ECM scaffolds and ADSCs would synergistically promote angiogenesis. These findings suggest that for angiogenesis and the angiogenic potential of ADSCs, SIS and ADM containing multiple ECM components, which are secreted by the resident cells, are superior to Co–CS–HA prepared with the three kinds of ECM components; further, ECM scaffolds such as SIS and ADM may also be superior to composite scaffolds prepared with one or several ECM components. Although it has been hypothesized that the angiogenic potential of ADSCs is a combined result of their ability to produce angiogenic growth factors and to differentiate into endothelial cells,⁸ in this study, a small amount of GFP-positive ADSCs was detected after 3 weeks of transplantation and only few GFP-positive cells had differentiated into vascular endothelial phenotype, indicating that the survival and differentiation of ADSCs into vascular endothelial phenotype is low. Thus, paracrine mechanisms of ADSCs play an important role in angiogenesis.

At 7 days after transplantation, ADSC-seeded scaffolds significantly enhanced wound healing compared with the nonseeded scaffolds. Neovascularization is a crucial step in the wound healing process³⁸; the formation of new blood vessels is necessary to sustain the newly formed granulation tissue and the survival of keratinocytes.³⁹ Thus, ADSCs could enhance wound healing by promoting angiogenesis. Meanwhile, ADSCs have been shown to enhance wound healing through differentiation into skin cells^24,25 and secretion of growth factors that stimulate collagen synthesis and migration of dermal fibroblasts.⁴⁰ We proposed that ADSCs enhance wound healing via multiple ways. However, the difference in the wound healing rate across the groups diminished over time, and the wounds nearly healed at day 21 in all the groups. Wound closure may occur at a faster rate in mice, because the cell proliferation rate is multifold faster in a small animal model than in humans; hence, self-regeneration of the skin has been shown to occur rapidly in a murine wound model.³⁰

A limitation of our study was that although we confirmed that the amounts of angiogenic growth factor secretion by ADSCs cultured in different scaffolds varied, we did not detect this difference in vivo. Unfortunately, since the ADSCs were harvested from C57BL/6-GFP mice and the wounds were created in C57BL/6 mice, it was technically very difficult to determine this difference in vivo. Further studies are required to maximize the angiogenic and regenerative potential of ADSCs. Modification of the microenvironment in scaffolds, combination of growth factors or other types of cells, and gene therapy via genetic modification are being increasingly used to address this problem.

Conclusion

In this study, we demonstrated that ADSCs could be used as a cell source to promote vascularization capacities of scaffolds. Moreover, compared with Co–CS–HA, SIS and ADM themselves not only promoted angiogenesis but also enhanced angiogenic growth factor secretion by the ADSCs; the combination of ECM scaffolds and ADSCs exhibited synergistic effects for promoting angiogenesis. These results suggested that when using stem cells to promote the vascularization capacity of scaffolds, both the vascularization capacity of the scaffold itself and its influence on the angiogenic potential of stem cells should be considered.

Footnotes

Acknowledgments

This study was supported by the funding support from National High Technology Research and Development Program of China (2006AA02A119) and the Nature Science Foundation of China (30973285).

Disclosure Statement

No competing financial interests exist.

References

Machens

H.G.

, Berger

A.C.

, Mailaender

Bioartificial skin. Cells Tissues Organs, 167:88. 2000.

Egaña

J.T.

, Fierro

F.A.

, Krüger

, Bornhäuser

, Huss

, Lavandero

, Machens

H.G.

Use of human mesenchymal cells to improve vascularization in a mouse model for scaffold-based dermal regeneration. Tissue Eng Part A, 15:1191. 2009.

Yao

, Roderfeld

, Rath

, Roeb

, Bernhagen

, Steffens

The impact of proteinase-induced matrix degradation on the release of VEGF from heparinized collagen matrices. Biomaterials, 27:1608. 2006.

Pieper

J.S.

, Hafmans

, van Wachem

P.B.

, van Luyn

M.J.

, Brouwer

L.A.

, Veerkamp

J.H.

, van Kuppevelt

T.H.

Loading of collagen-heparan sulfate matrices with bFGF promotes angiogenesis and tissue generation in rats. J Biomed Mater Res, 62:185. 2002.

Shi

, Han

, Mao

, Ma

, Gao

Enhanced angiogenesis in porous collagen-chitosan scaffolds loaded with angiogenin. Tissue Eng Part A, 14:1775. 2008.

Jabbarzadeh

, Starnes

, Khan

Y.M.

, Jiang

, Wirtel

A.J.

, Deng

, Lv

, Nair

L.S.

, Doty

S.B.

, Laurencin

C.T.

Induction of angiogenesis in tissue-engineered scaffolds designed for bone repair: a combined gene therapy-cell transplantation approach. Proc Natl Acad Sci USA, 105:11099. 2008.

Rosilio

, Benoit

J.P.

, Deyme

, Thies

, Madelmont

A physicochemical study of the morphology of progesterone-loaded microspheres fabricated from poly(D,L-lactide-co-glycolide)

J Biomed Mater Res, 25:667. 1991.

Murohara

, Shintani

, Kondo

Autologous adipose-derived regenerative cells for therapeutic angiogenesis. Curr Pharm Des, 15:2784. 2009.

Holder

W.D.

, Gruber

H.E.

, Roland

W.D.

, Moore

A.L.

, Culberson

C.R.

, Burg

K.J.L.

, Mooney

D.J.

Increased vascularization and heterogeneity of vascular structures occurring in polyglycolide matrices containing aortic endothelial cells implanted in the rat. Tissue Eng, 3:149. 1997.

10.

Rickert

, Moses

M.A.

, Lendlein

, Kelch

, Franke

R.P.

The importance of angiogenesis in the interaction between polymeric biomaterials and surrounding tissue. Clin Hemorheol Microcirc, 28:175. 2003.

11.

Salem

, Ciba

, Rapoport

D.H.

, Egana

J.T.

, Reithmayer

, Kadry

, Machens

H.G.

, Kruse

The influence of pancreas-derived stem cells on scaffold based skin regeneration. Biomaterials, 30:789. 2009.

12.

Egaña

J.T.

, Danner

, Kremer

, Rapoport

D.H.

, Lohmeyer

J.A.

, Dye

J.F.

, Hopfner

, Lavandero

, Kruse

, Machens

H.G.

The use of glandular-derived stem cells to improve vascularization in scaffold-mediated dermal regeneration. Biomaterials, 30:5918. 2009.

13.

Miranville

, Heeschen

, Sengenès

, Curat

C.A.

, Busse

, Bouloumié

Improvement of postnatal neovascularization by human adipose tissue-derived stem cells. Circulation, 110:349. 2004.

14.

Planat-Benard

, Silvestre

J.S.

, Cousin

, André

, Nibbelink

, Tamarat

, Clergue

, Manneville

, Saillan-Barreau

, Duriez

, Tedgui

, Levy

, Pénicaud

, Casteilla

Plasticity of human adipose lineage cells toward endothelial cells: physiological and therapeutic perspectives. Circulation, 109:656. 2004.

15.

Moon

M.H.

, Kim

S.Y.

, Kim

Y.J.

, Kim

S.J.

, Lee

J.B.

, Bae

Y.C.

, Sung

S.M.

, Jung

J.S.

Human adipose tissue-derived mesenchymal stem cells improve postnatal neovascularization in a mouse model of hindlimb ischemia. Cell Physiol Biochem, 17:279. 2006.

16.

Kondo

, Shintani

, Shibata

, Murakami

, Imaizumi

, Kitagawa

, Murohara

Implantation of adipose-derived regenerative cells enhances ischemia-induced angiogenesis. Arterioscler Thromb Vasc Biol, 29:61. 2009.

17.

Rehman

, Traktuev

, Li

, Merfeld-Clauss

, Temm-Grove

C.J.

, Bovenkerk

J.E.

, Pell

C.L.

, Johnstone

B.H.

, Considine

R.V.

, March

K.L.

Secretion of angiogenic and antiapoptotic factors by human adipose stromal cells. Circulation, 109:1292. 2004.

18.

Nakagami

, Maeda

, Morishita

, Iguchi

, Nishikawa

, Takami

, Kikuchi

, Saito

, Tamai

, Ogihara

, Kaneda

Novel autologous cell therapy in ischemic limb disease through growth factor secretion by cultured adipose tissue-derived stromal cells. Arterioscler Thromb Vasc Biol, 25:2542. 2005.

19.

Zuk

P.A.

, Zhu

, Ashjian

, De Ugarte

D.A.

, Huang

J.I.

, Mizuno

, Alfonso

Z.C.

, Fraser

J.K.

, Benhaim

, Hedrick

M.H.

Human adipose tissue is a source of multipotent stem cells. Mol Biol Cell, 13:4279. 2002.

20.

De Ugarte

D.A.

, Morizono

, Elbarbary

, Alfonso

, Zuk

P.A.

, Zhu

, Dragoo

J.L.

, Ashjian

, Thomas

, Benhaim

, Chen

, Fraser

, Hedrick

M.H.

Comparison of multi-lineage cells from human adipose tissue and bone marrow. Cells Tissues Organs, 174:101. 2003.

21.

Badylak

S.F.

, Freytes

D.O.

, Gilbert

T.W.

Extracellular matrix as a biological scaffold material: structure and function. Acta Biomater, 5:1. 2009.

22.

Badylak

S.F.

The extracellular matrix as a biologic scaffold material. Biomaterials, 28:3587. 2007.

23.

Wang

, Zhang

, Lu

, Zhang

, Ma

, Rong

, Yu

, Jin

Cross-linked collagen-chondroitin sulfate-hyaluronic acid imitating ECM as scaffold for dermal tissue engineering. Tissue Eng Part C Methods, 16:269. 2010.

24.

Altman

A.M.

, Matthias

, Yan

, Song

Y.H.

, Bai

, Chiu

E.S.

, Slakey

D.P.

, Alt

E.U.

Dermal matrix as a carrier for in vivo delivery of human adipose-derived stem cells. Biomaterials, 29:1431. 2008.

25.

Altman

A.M.

, Yan

, Matthias

, Bai

, Rios

, Mathur

A.B.

, Song

Y.H.

, Alt

E.U.

Human adipose-derived stem cells seeded on a silk fibroin-chitosan scaffold enhance wound repair in a murine soft tissue injury model. Stem Cells, 27:250. 2009.

26.

Flynn

L.E.

, Prestwich

G.D.

, Semple

J.L.

, Woodhouse

K.A.

Proliferation and differentiation of adipose-derived stem cells on naturally derived scaffolds. Biomaterials, 29:1862. 2008.

27.

Woo

K.M.

, Seo

, Zhang

, Ma

P.X.

Suppression of apoptosis by enhanced protein adsorption on polymer/hydroxyapatite composite scaffolds. Biomaterials, 28:2622. 2007.

28.

, Genetos

D.C.

, Leach

J.K.

Osteogenesis and trophic factor secretion are influenced by the composition of hydroxyapatite/poly(lactide-co-glycolide) composite scaffolds. Tissue Eng Part A, 16:127. 2010.

29.

Tan

M.Y.

, Zhi

, Wei

R.Q.

, Huang

Y.C.

, Zhou

K.P.

, Tan

, Deng

, Luo

J.C.

, Li

X.Q.

, Xie

H.Q.

, Yang

Z.M.

Repair of infarcted myocardium using mesenchymal stem cell seeded small intestinal submucosa in rabbits. Biomaterials, 30:3234. 2009.

30.

Zhang

, Deng

, Wang

, Yang

, Guo

, Li

, Ma

, Yu

, Zhang

, Jin

Expansion and delivery of human fibroblasts on micronized acellular dermal matrix for skin regeneration. Biomaterials, 30:2666. 2009.

31.

Liu

, Fan

, Wang

, Toh

S.L.

, Goh

J.C.

The interaction between a combined knitted silk scaffold and microporous silk sponge with human mesenchymal stem cells for ligament tissue engineering. Biomaterials, 29:662. 2008.

32.

Carmeliet

Angiogenesis in life, disease and medicine. Nature, 438:932. 2005.

33.

Badylak

, Liang

, Record

, Tullius

, Hodde

Endothelial cell adherence to small intestinal submucosa: an acellular bioscaffold. Biomaterials, 20:2257. 1999.

34.

Silverman

R.P.

, Li

E.N.

, Holton

L.H.

3rd , Sawan

K.T.

, Goldberg

N.H.

Ventral hernia repair using allogenic acellular dermal matrix in a swine model. Hernia, 8:336. 2004.

35.

Bleiziffer

, Eriksson

, Yao

, Horch

R.E.

, Kneser

Gene transfer strategies in tissue engineering. J Cell Mol Med, 11:206. 2007.

36.

Bhang

S.H.

, Cho

S.W.

, Lim

J.M.

, Kang

J.M.

, Lee

T.J.

, Yang

H.S.

, Song

Y.S.

, Park

M.H.

, Kim

H.S.

, Yoo

K.J.

, Jang

, Langer

, Anderson

D.G.

, Kim

B.S.

Locally delivered growth factor enhances the angiogenic efficacy of adipose-derived stromal cells transplanted to ischemic limbs. Stem Cells, 27:1976. 2009.

37.

Ferrara

, Gerber

H.P.

, LeCouter

The biology of VEGF and its receptors. Nat Med, 9:669. 2003.

38.

Singer

A.J.

, Clark

R.A.

Cutaneous wound healing. N Engl J Med, 341:738. 1999.

39.

, Chen

, Scott

P.G.

, Tredget

E.E.

Mesenchymal stem cells enhance wound healing through differentiation and angiogenesis. Stem Cells, 25:2648. 2007.

40.

Kim

W.S.

, Park

B.S.

, Sung

J.H.

, Yang

J.M.

, Park

S.B.

, Kwak

S.J.

, Park

J.S.

Wound healing effect of adipose-derived stem cells: a critical role of secretory factors on human dermal fibroblasts. J Dermatol Sci, 48:15. 2007.