SEARCH

SEARCH BY CITATION

Keywords:

  • Adult stem cells;
  • Differentiation;
  • Stem cell-microenvironment interactions;
  • Seeding

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

Soft tissue loss presents an ongoing challenge in reconstructive surgery. Local stem cell application has recently been suggested as a possible novel therapy. In the present study we evaluated the potential of a silk fibroin-chitosan (SFCS) scaffold serving as a delivery vehicle for human adipose-derived stem cells (ASCs) in a murine soft tissue injury model. Green fluorescent protein (GFP)-labeled ASCs were seeded on SFCS scaffolds at a density of 1 × 105 ASCs per cm2 for 48 hours and then suture-inlaid to a 6-mm, full-thickness skin defect in 6-week-old male athymic mice. Wound healing was tracked for 2 weeks by planimetry. Histology was evaluated at 2 and 4 weeks. Our data show that the extent of wound closure was significantly enhanced in the ASC-SFCS group versus SFCS and no-graft controls at postoperative day 8 (90% ± 3% closure vs. 75% ± 11% and 55% ± 17%, respectively). Microvessel density at wound bed biopsy sites from 2 weeks postoperative was significantly higher in the ASC-SFCS group versus SFCS alone (7.5 ± 1.1 vs. 5.1 ± 1.0 vessels per high-power field). Engrafted stem cells were positive for the fibroblastic marker heat shock protein 47, smooth muscle actin, and von Willebrand factor at both 2 and 4 weeks. GFP-positive stem cells were also found to differentiate into epidermal epithelial cells at 4 weeks postoperative. In conclusion, human adipose-derived stem cells seeded on a silk fibroin-chitosan scaffold enhance wound healing and show differentiation into fibrovascular, endothelial, and epithelial components of restored tissue. STEM CELLS2009;27:250–258


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

Extensive soft tissue loss and architectural distortion present therapeutic difficulties in reconstructive surgery. To a limited degree, these challenges have been met by advances in materials science. Bioprosthetic scaffolds provide a structural framework and architectural milieu favorable for cell incorporation after implantation and the mechanical basis for the strength of a repair in the setting of soft tissue loss. The confluence of biocompatible matrices for strength elements and cell incorporation for tissue regeneration is assumed to be beneficial [1, 2].

Silk fibroin has long been investigated on the basis of its favorable biological characteristics in the context of surgical implantation [3–5]. It is known as a reliable suture material with mid-range degradation kinetics and solid mechanical strength [6]. Furthermore, various sheet and fiber conformations have been documented to favor cellular adhesion. Consistent with the principle of mimicking nature's attributes, advances in optimizing silk fibroin's qualities in terms of biocompatibility and cell support have been made by hybridization with the structural protein chitosan [7].

Chitosan is a naturally occurring polysaccharide composed of alternating acetylated and deacetylated D-glucosamine residues. It is generally inert in vivo and has favorable degradation kinetics, as with silk fibroin. Chitosan has been applied clinically in hemostatic wound dressings and is emerging as a promising constituent of novel biocompatible matrices in tissue engineering [8]. Chitosan mimics polysaccharide and glycosaminoglycan constituents of the extracellular matrix, enabling it to function as a substrate for cell adhesion, migration, and ultimately tissue incorporation [8].

The blend of silk fibroin with chitosan allows the well-known structural features of silk fibroin to benefit from the surface chemistry of chitosan to result in a scaffold with features ideal for cell attachment and population of the substratum. The resultant hybrid matrix has previously been reported to demonstrate excellent biocompatibility and practical utility in abdominal wall reconstruction as an index application that can be generalized to a variety of soft tissue repairs [7].

Mesenchymal stem cells of similar basic properties were initially described after isolation from bone marrow and have more recently been isolated from adipose, among other somatic tissues [9, 10]. Adipose-derived stem cells (ASCs) have been increasingly reported to confer benefits in vivo as agents of angiogenesis and multilineage restoration in the face of soft tissue defects [11–13]. Increasing evidence suggests that stem cells are residents of a microvascular niche, on stand-by for tissue repair as needed [14, 15]. However, with extensive tissue damage, the local pool of stem cells available for repair is putatively insufficient to fully correct the deficiency. Discarded adipose tissue obtained from liposuction procedures contains a significant number of mesenchymal stem cells accessed via a relatively low-risk surgical intervention, which can be applied therapeutically to counteract this deficiency [16, 17]. The objectives of our study were first to evaluate whether silk fibroin-chitosan (SFCS) could serve as an effective stem cell carrier, second to understand the impact of ASC-SFCS on wound healing, and finally to study the differentiation potential of transplanted human stem cells in the setting of a murine cutaneous injury model.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

Cell Isolation and Culture

Human adipose tissue was obtained from elective body contouring procedures in compliance with the guidelines of the Tulane University School of Medicine Institutional Review Board. Tissue was minced by sharp dissection in the case of solid tissue and directly exposed to enzymatic digestion with liposuction aspirates. Minced specimens were added to a solution of 0.07% blendzyme 3 (F. Hoffman-La Roche Ltd., Basel, Switzerland, http://www.roche.com), digested with mild agitation at 37°C for 60 minutes, passed through a 40-μm filter, and finally selected on the basis of adherence to 75 square centimeter surface area tissue culture flasks (Greiner Bio-One GmbH, Frickenhausen, Germany, http://www.gbo.com). Cells were grown in α-minimal essential medium supplemented with 20% fetal bovine serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 μg/ml streptomycin. Cells were incubated in a 5% CO2-containing chamber at 37°C, and medium was changed every 3 days. ASCs between passages 1 and 8 were used for all experiments. ASCs used in these experiments have previously been characterized by our group [11, 18]. Furthermore, the multilineage differentiation potential of these cells has also been demonstrated in our laboratory [18].

Synthesis of SFCS Grafts

The sericin coating of raw silk fiber donated by Dr. S. Hudson (North Carolina State University, Raleigh, NC) was removed via degumming. Sodium dodecyl sulfate (0.25%; wt/vol; Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com) and sodium carbonate (0.25%; wt/vol; Sigma-Aldrich) were dissolved and heated to 100°C. Silk was added at 1:100 (wt/vol) and heated for 1 hour, followed by draining of the alkaline soap solution. Degummed silk was rinsed in running distilled water, air-dried, and then dissolved in calcium nitrate tetrahydrate-methanol (molar ratio, 1:4:2 calcium/water/methyl alcohol) at 65°C. The silk fibroin (SF) was dissolved at 10% (wt/vol) concentration over a 3-hour period with continuous stirring.

Chitosan (CS) solution was prepared by 2% acetic acid dissolution of high molecular weight chitosan (82.7% deacetylation; Sigma-Aldrich). Under continuous stirring, SF and CS solutions were combined for preparation of a 75:25 (vol/vol) SF/CS blend, followed by mixing for 30 minutes and then 4-day dialysis (molecular weight cutoff, 6–8 kDa) against deionized water.

Forty milliliters of SFCS blend solution was added to a glass Petri dish and then nondirectionally frozen overnight at −80°C, followed by 2-day lyophilization. Dry samples were treated in a 50:50 (vol/vol) methanol:1 N sodium hydroxide (NaOH) solution for 15 minutes for SF crystallization and CS neutralization. Methanol:NaOH was then replaced by 1 N NaOH for 12–18 hours. NaOH was removed by dilution in phosphate-buffered saline (PBS; 1×) with sequential changes of solution hourly for 4 hours and then quarter-hourly until pH equilibration at 7.0 was reached. Samples were sterilized with 70% ethanol immersion for 12–18 hours and subsequently rinsed in sterile PBS prior to in vitro cell seeding and subsequent in vivo engraftment. Final graft thickness was 1.5 mm.

Green Fluorescent Protein Transfection for In Vivo Tracking of Stem Cells

ASCs at 60% confluence were transfected with a lentiviral vector carrying enhanced green fluorescent protein (eGFP) under the control of the cytomegalovirus promoter using an adaptation of standard protocols [19]. Briefly, 1 × 106 ASCs at passage 1 were exposed to 10 ml of vector-containing medium (1.2 × 106 virus particles per milliliter of medium) in a 75-cm2 flask. Eight micrograms of polybrene was added (Sigma-Aldrich). Vector-containing medium was replaced by fresh medium 24 hours later. Six days subsequent to infection, green fluorescent protein (GFP)-positive cells were subjected to flow cytometric sorting with stringent gating for successfully transfected fluorescent cells using the FACSVantage SE cell sorter (Becton, Dickinson and Company, Franklin Lakes, NJ, http://www.bd.com).

Stem Cell Loading of SFCS Grafts

Six-millimeter-diameter SFCS grafts were placed completely covering the well bottom in 96-well plates. Grafts were covered with 200-μl aliquots of medium alone in the SFCS group and with an equal volume of cell suspension containing 1 × 105 ASCs per cm2 in the ASC-SFCS group. Grafts were incubated under standard culture conditions for 24 hours, after which the overlying medium or cell suspension was aspirated. The graft was flipped to place the opposite surface facing up, and the corresponding medium or cell suspension solution was placed on the other side. Grafts were then incubated for 24 hours and transferred to the operating suite for surgical engraftment. Once on the operative field, grafts were transferred to a sterile six-well plate and washed gently in two 500-μl aliquots of PBS to remove any nonadherent cells or medium.

Surgical Procedure: Engraftment of ASC-SFCS Construct

In vivo portions of the study were conducted with the approval of the University of Texas M.D. Anderson Cancer Center Institutional Animal Care and Use Committee. A total of 17 male athymic mice of the strain nu/nu (Charles River Laboratories, Wilmington, MA, http://www.criver.com) were included in the study and placed into an airtight chamber for induction of anesthesia with 3% inhalational isoflurane (Isoflo; Abbott, Chicago, http://www.abbott.com) for 90 seconds. Animals were subsequently placed prone on the operating table and connected to a circuit delivering 1.5% inhalational isoflurane for maintenance anesthesia. The limbs were taped stationary, and the entire dorsum was scrubbed in 2% chlorhexidine gluconate solution (Chlorhexiderm Plus; DVM Pharmaceuticals, Inc, Miami, http://www.ivxanimalhealth.com). A 6-mm-diameter circular impression was made by placement of a punch biopsy instrument on the left paramedian dorsum of the mouse, followed by grasping and elevation of the circular region of tissue with an Adson forceps and sharp excision of the tissue with a Metzenbaum scissor. For the main study 10 animals were randomized to one of three treatment groups: no graft, SFCS alone, or ASC-SFCS. Animals in each group received one lesion and a graft-based repair depending on group randomization. For the gross analysis of vascular infiltration patterns, an additional two animals received an SFCS graft on the left paramedian dorsum and an ASC-SFCS graft on the contralateral side, with a third operative site receiving no graft at the midline caudal to the graft sites; sites were separated by 3 cm. Grafts were secured in place with 3–4 interrupted 6-0 prolene tacking sutures (Ethicon; Johnson & Johnson, New Brunswick, NJ, http://www.jnj.com). Operative sites were covered with a thin layer of 0.9% saline gel (Normlgel; Mölnlycke Healthcare, Gothenburg, Sweden, http://www.molnlycke.com), and mice were reversed from general anesthesia with supplemental 100% O2 support. Animals were observed for 15 minutes postoperative and then placed in individual cages to resume activity ad libitum.

Care of Experimental Animals in the Perioperative Period

Animals were housed in an air-filtered barrier facility and received autoclaved nutrition in a sterile working environment. Mice were housed one per cage postoperatively, fed chow ad libitum, and subjected to physical examinations daily. Postoperative analgesia was administered with subcutaneous injections of 0.5–2.5 mg/kg buprenorphine (Buprenex; Reckitt Benckiser Healthcare Ltd., Hull, U.K., http://www.reckittbenckiser.com) on postoperative days 0 and 1. Perioperative antibiosis was achieved with subcutaneous injections of 15 mg/kg cefazolin (Sandoz, Princeton, NJ, http://www.us.sandoz.com) on postoperative days 0, 1, and 2. General anesthesia was briefly induced as before on postoperative day 3 for suture removal.

Planimetry Analysis of Wound Closure Rates

Experimental animals were placed against the backdrop of a metric ruler and photographed on postoperative days 0–10 using a Canon high-resolution digital camera (PowerShot G7; Canon, Hong Kong, http://www.canon.com). Photographs were analyzed with the Metamorph software package (Molecular Devices Corp., Sunnyvale, CA, http://www.moleculardevices.com). Measurement of wound closure area was defined by the limits of grossly evident epithelialization, with all surface areas in a two-dimensional plane calibrated against the adjacent metric ruler. Four individual photomicrographic measurements were taken of each mouse. Percentage of wound closure (contraction) was defined as follows: (area at postoperative day 0 − area at postoperative day x)/(area at postoperative day 0) × 100. All 15 mice randomized between the group receiving no graft, the group receiving an ASC-SFCS graft (n = 5), and the group receiving SFCS alone (n = 5) were included in the wound healing analysis. All measurements were independently evaluated by two investigators blinded to the randomization.

Gross Study of Wound Bed Cutaneous Vascular Infiltration

As an index subgroup for comparison of the gross pattern of vascular infiltration, two animals received both SFCS and ASC-SFCS grafts, along with an additional wound receiving no graft. Animals were euthanized at postoperative day 9, and approximately 2 × 2 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 100 × 20 mm polystyrene cell culture dish (Corning Enterprises, Corning, NY, http://www.corning.com) and spread with moderate tension along the plate bottom, with the superficial surface facing up. Next, focused, standard incandescent illumination was directed immediately under the specimen into the polystyrene dish, illuminating the tissue density and vascular infiltration of wound bed biopsy specimens. Digital images were taken as with planimetry.

Standard Histology

Tissue specimens were obtained from the operative site by sharp dissection and placed either into 10% formalin for fixation and subsequent paraffin processing and embedding or into molds for immersion in Tissue-Tek OCT Compound (Sakura Finetek, Torrance, CA, http://www.sakura.com) freezing medium on dry ice for frozen block preparation. In addition, specimens of normal skin were taken from skin 4 cm distant from the operative site to conduct a survey of any possible regional distribution of cells beyond the immediate delivery and repair zone. Histology sections were prepared from tissue specimens harvested at 2 weeks postoperatively, with one animal from each group remaining intact for tissue harvest at 4 weeks. Serial sectioning was performed (5 μm for frozen sections and 4 μm for paraffin sections), and representative sections underwent H&E staining for examination of tissue architecture. Additional serial sections were prepared as unstained slides for subsequent immunofluorescent studies. H&E-stained slides were studied as follows: z-sections were analyzed using a Zeiss Axiovert microscope (Carl Zeiss, Oberkochen, Germany, http://www.zeiss.com) equipped with a Canon G7 high-resolution digital camera adaptor for image acquisition.

Microvascular Density Analysis

For microvessel density determination, paraffin sections were deparaffinized and stained with primary antibody directed against von Willebrand factor (1:100 dilution; polyclonal rabbit-anti-human; Dako, Glostrup, Denmark, http://www.dako.com). Briefly, slides were deparaffinized in serial xylene and ethanol washes, followed by exposure to target retrieval solution at 95°C for 40 minutes. Slides were subsequently cooled, washed serially in PBS, blocked for 1 hour in 10% donkey serum at room temperature, incubated in primary antibody solution at 37°C for 1 hour, washed, incubated in secondary antibody solution (1:1,000 dilution; Alexa Fluor 594 donkey anti-rabbit IgG; Invitrogen, Carlsbad, CA, http://www.invitrogen.com), washed, incubated in 4,6-diamidino-2-phenylindole (DAPI) to stain nuclei, and mounted with coverslips for viewing. Quantification assay was based on adaptation of established protocols [20]. Ten serial sections were examined for each animal, with five high-power fields examined per slide. Vessels with a diameter of ≤50 μm were counted, with appropriate diameter vessels beyond a clear bifurcation point counted as two distinct microvessels.

Immunofluorescent Histology

Immunohistochemistry analysis based on fluorescent conjugation of the signal was carried out to identify engrafted ASCs on the basis of eGFP signal and to identify the phenotype of engrafted stem cells with specific antibodies against additional antigens. Primary antibodies against the following epitopes were used in this series of experiments: smooth muscle actin (SMA; ab32575; rabbit-anti-human; 1:500), GFP (ab6556 and ab6673; rabbit-anti-human and goat-anti-human, respectively; 1:200), Ki67 (ab833 and ab15580; rabbit-anti-human; 1:50), heat shock protein 47 (HSP-47; ab13510; mouse-anti-human; 1:1,000), cytokeratin 19 (ab15463; rabbit-anti-human; 1:200) (Abcam, Cambridge, MA, http://www.abcam.com), and von Willebrand factor (vWF; A0082; rabbit-anti-human; 1:1,000; Chemicon, Billerica, MA, http://www.chemicon.com). Prior to antibody staining, slides were washed in PBS and fixed with 4% paraformaldehyde for 10 minutes at room temperature. Next, slides were washed with PBS containing 0.3% Triton X-100 (Sigma-Aldrich) and subjected to blocking solution of 10% donkey serum for 30 minutes at room temperature. Primary antibody incubation followed for 1 hour at 37°C, followed by PBS wash and incubation with either an Alexa Fluor 488 donkey anti-goat IgG secondary antibody (Invitrogen), Alexa Fluor 594 donkey anti-rabbit IgG, or Alexa Fluor 594 donkey anti-mouse IgG, species-matched according to primary antibody species used. Secondary antibodies were applied at a dilution factor of 1:1,000 for 1 hour at room temperature. DAPI nuclear dye was applied for 15 minutes at room temperature, and slides were again washed and were mounted with coverslips. Slides were analyzed and photomicrographs taken using a confocal microscope (Olympus, Center Valley, PA, http://www.olympus-global.com).

Statistical Analysis

Values are presented as means ± SD. All analysis was performed using the Statistical Program for Social Science for Windows (SPSS, Chicago, http://www.spss.com). Statistical independent-samples t test was performed with equal variances not assumed to determine significance of all data. p ≤ .05 was considered to define statistical significance.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

Planimetry Analysis of Wound Closure Rates

Wound closure was measured by planimetric analysis and revealed a wound closure at postoperative day 6 of 46% ± 15% in the control group receiving no graft, 58% ± 9% in the SFCS group, and 72% ± 5% in the ASC-SFCS group (p ≤ .05). Postoperative day 8 values were 55% ± 17% in the no-graft group, 75% ± 11% in the SFCS group, and 90% ± 3% in the ASC-SFCS group (p ≤ .05). As wound closure accelerated toward complete epithelialization by postoperative day 14 in all groups, the trend of enhanced closure in the ASC-SFCS group continued, but it became less pronounced as all wounds neared complete healing (Fig. 1).

thumbnail image

Figure 1. Immediate post-op and 2-week images of human ASC-seeded SFCS scaffold engrafted into a 6-mm-diameter, full-thickness cutaneous wound in athymic male mice, H&E photomicrograph from region of regenerated wound bed tissue, and demonstration of significantly enhanced microvessel density in ASC-SFCS engraftment biopsies at 2 weeks post-op. (A): Graft immediately post-op. Scale bar = 5 mm. (B): Healed wound bed at 2 weeks post-op. Scale bar = 5 mm. (C): H&E image from healed wound bed region. Large arrow indicates vascular structure with residual erythrocytes; ∗ indicates papillary dermis; small arrow indicates intact epidermal epithelium. Scale bar = 100 μm. (D): Mean microvessel density in the ASC-SFCS group at 2 weeks post-op was 7.5 ± 1.1 vessels per high-power field, whereas microvessel density in the SFCS group at 2 weeks was 5.1 ± 1.0 vessels per high-power field. (E): Representative image showing 4,6-diamidino-2-phenylindole stain highlighting nuclei (blue) and stain for von Willebrand factor (red) colocalizing with nuclear signal and outlining the luminal cross-sections of microvessels in dermal tissue biopsy from 2 weeks post-op in ASC-SFCS-engrafted mice. Scale bar = 50 μm. Abbreviations: ASC, adipose-derived stem cell; post-op, postoperative; SFCS, silk fibroin-chitosan.

Download figure to PowerPoint

Gross Study of Wound Bed Cutaneous Vascular Infiltration

Wound bed analysis of fresh tissue mounts demonstrated a markedly enhanced extent of wound closure in the ASC-SFCS group in comparison with both the SFCS and no-graft control groups at postoperative day 9 (Fig. 2). Close inspection of images under intense illumination revealed an apparently more robust invasion of vascular tissue, characterized by hyperemia, in the SFCS and ASC-SFCS groups versus no-graft controls. Furthermore, the extent of vascular infiltration of the surrounding tissue in the region of the operative site was greater in qualitative magnitude in the ASC-SFCS group versus the SFCS group (Fig. 2).

thumbnail image

Figure 2. Whole tissue mounts of regenerated wound bed biopsies from a single animal excised at 9 days postoperative demonstrate enhanced vascular infiltration at wound sites receiving silk fibroin-chitosan (SFCS) scaffold grafts versus no-graft control, with further gross hyperemia and vascular infiltration noted in a specimen from adipose-derived stem cell (ASC)-SFCS engraftment site. (A): No-graft control. (B): SFCS engraftment site. (C): ASC-SFCS engraftment site. Scale bars = 5 mm.

Download figure to PowerPoint

Standard Histology

Routine histological examination of tissue biopsies from the wound bed region, based on H&E staining, revealed completely closed wounds in both the SFCS and ASC-SFCS groups by 2 weeks postoperative, consistent with planimetry observations (Fig. 1). Close examination revealed the restoration of an intact epidermal epithelium in all cases. The completeness of the restored epithelium did not differ significantly between the SFCS and ASC-SFCS groups at 2 weeks. The presence of vascular structures, as evidenced by consistent luminal morphology and infiltration with erythrocytes, was noted among restored soft tissue of the midstrata and deep strata of the reticular dermis. Findings of such vascular structures correlate with subsequent immunofluorescent findings. Trace fibers of silk-chitosan were seen among dermal elements in wound bed biopsy but were rare, suggesting degradation of the majority of SFCS carrier by 2 weeks. There was no evident inflammatory infiltrate (no polymorphonuclear cell infiltration, no giant cells noted) on any H&E stains of wound bed biopsies at 2 weeks, indicating excellent biocompatibility of engrafted SFCS.

Microvascular Density Analysis

Wound bed biopsy specimens taken at 2 weeks showed significantly greater vascular density in animals treated with ASC-SFCS grafts versus animals receiving acellular SFCS grafts (Fig. 1). Mean microvessel density in the ASC-SFCS group at 2 weeks postoperative was 7.5 ± 1.1 vessels per high-power field, whereas density in the SFCS group at 2 weeks was 5.1 ± 1.0 vessels per high-power field (p ≤ .05).

Immunofluorescent Histology

Engrafted GFP cells were identified throughout the various substrata of the dermis and cutaneous appendages at 2 and 4 weeks postoperatively (Figs. 3–6). Engrafted ASCs were identified, on the basis of GFP signal, at 2 weeks postoperative staining positive for the nuclear marker of proliferation Ki67, indicating ongoing proliferation of engrafted ASCs (Fig. 3). Some engrafted ASCs were observed to display a fibroblastic phenotype, on the basis of GFP/HSP-47 costaining at 2 weeks (Fig. 4); others were noted as recapitulating elements of linear vascular structures, costaining for the vascular smooth muscle marker SMA and GFP (Fig. 5). In addition, engrafted cells were seen incorporating into winding linear structures consistent with microvascular elements and coexpressing positive signals for GFP and the endothelial marker vWF (Fig. 5). At 2 weeks, GFP cells were not observed incorporated into regenerated epidermal epithelium.

thumbnail image

Figure 3. Demonstration of engrafted human adipose-derived stem cells delivered via silk fibroin-chitosan (SFCS) scaffold in dermal tissue biopsy 2 weeks postoperative expressing the marker of proliferation Ki67. (A): 4,6-Diamidino-2-phenylindole staining indicates nuclei (blue). (B): Engrafted stem cells are indicated by green fluorescent protein staining (green). (C): Red staining indicates Ki67 signal overlapping with nucleus. (D): Overlay image demonstrates colocalization of all signals. Scale bars = 20 μm.

Download figure to PowerPoint

thumbnail image

Figure 4. Engrafted human adipose-derived stem cells (ASCs) express the marker of fibroblastic phenotype heat shock protein 47 (HSP-47). Representative images demonstrate HSP-47 costaining of engrafted green fluorescent protein (GFP)-positive ASCs. (A): 4,6-Diamidino-2-phenylindole staining indicates nuclei (blue). (B): Positive identification of engrafted ASCs in dermal tissue biopsy is indicated by GFP staining (green). (C): Red stain indicates HSP-47. (D): Overlay demonstrates colocalization of GFP and HSP-47 signal. Scale bars = 20 μm.

Download figure to PowerPoint

thumbnail image

Figure 5. Spontaneous participation of engrafted human adipose-derived stem cells (ASCs) in recapitulation of vascular elements in murine soft tissue injury model after delivery via cutaneous engraftment of cell-seeded silk fibroin-chitosan (SFCS) graft. (A): 4,6-Diamidino-2-phenylindole (DAPI) staining indicates nuclei. (B): Green staining indicates engrafted ASCs. (C): DAPI stain for nuclei. (D): Green staining indicates engrafted stem cells. (E): Red staining indicates smooth muscle actin (SMA) in cross-sectional luminal structure consistent with vessel. (F): Overlay demonstrates colocalization of green fluorescent protein (GFP)/SMA signal, indicating differentiation of stem cells. (G): Red signal indicates positive staining for von Willebrand factor (vWF). (H): Overlay indicates colocalization of GFP/vWF signal, indicating ASC differentiation into endothelial components of restored vascular elements. Scale bars = 20 μm.

Download figure to PowerPoint

thumbnail image

Figure 6. Participation of engrafted human adipose-derived stem cells (ASCs) delivered via silk fibroin-chitosan (SFCS) scaffold in restored epidermal epithelium at 4 weeks postoperative. (A): 4,6-Diamidino-2-phenylindole staining indicates nuclei (blue). (B): Engrafted ASCs are indicated by positive stain for green fluorescent protein (green). (C): Red stain indicates positive staining for cytokeratin 19, a marker of epidermal epithelial phenotype. (D): Overlay demonstrates colocalization of signal, indicating differentiation of engrafted ASCs into ectodermal-lineage epithelial components at 4 weeks postoperative. Scale bars = 20 μm.

Download figure to PowerPoint

Additional immunofluorescent studies undertaken on biopsies taken at 4 weeks postoperative revealed the same findings observed at 2 weeks (GFP/Ki67+/+ cells, GFP/HSP-47+/+ cells, GFP/SMA+/+, and GFP/vWF+/+ cells in structural conformations consistent with vascular structures). In addition to these findings, engrafted GFP-positive cells were observed incorporating as components of the regenerated epidermal epithelium at 4 weeks on the basis of GFP cells costaining for the cytokeratin marker of epidermal epithelium cytokeratin 19 (CK19) (Fig. 6).

Examination of biopsies of normal skin in regions 4 cm distant from the site of ASC-SFCS engraftment revealed no presence of engrafted ASCs at 2 weeks (Fig. 7). Likewise, at 4 weeks, serial sections probing for regionally distributed ASCs on the basis of GFP signal revealed no cells distant from the operative site, indicating a primary focal persistence of ASCs engrafting at the local site of soft tissue reconstruction.

thumbnail image

Figure 7. Dermal biopsies 4 cm distant from the site of adipose-derived stem cell (ASC) delivery via silk fibroin-chitosan (SFCS) scaffold and wound biopsies from SFCS graft controls demonstrate the lack of green fluorescent protein (GFP) signal, indicating focal persistence of stem cells and lack of stem cell engraftment in non-cell-seeded, SFCS-graft-only controls. (AD): 4,6-Diamidino-2-phenylindole staining indicates nuclei (blue); probe for green fluorescence shows a lack of signal, indicating no engrafted ASCs at this distant tissue site; overlay with bright-field image highlights the absence of stem cell GFP signal at a 4-cm-distant anatomic site. (EH): Wound bed biopsies from 2 weeks postoperative show the absence of GFP signal in SFCS controls. Shown are nuclei (blue), GFP green (no signal), and smooth muscle actin (red). (IL): Wound bed biopsies from 4 weeks postoperative show the absence of GFP signal in recapitulated epithelium in SFCS-only control animals. Shown are nuclei (blue), GFP (green; no signal), and cytokeratin 19 (red). Scale bars = 20 μm.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

We report here for first time the successful use of a 75:25 silk fibroin-chitosan blend scaffold for the seeding and in vivo delivery of human adipose-derived stem cells in a murine cutaneous wound model. Our data indicate that this delivery technique not only is feasible but confers the physiologic benefits of accelerated wound closure. In this study the effect of ASC-seeded scaffold grafts versus unseeded scaffold and no-graft controls was investigated, and findings were consistent with the observation of enhanced wound closure with stem cell therapy, as reported previously using an alternative scaffold delivery approach [11]. In the present study, we show that ASCs engraft, proliferate, and differentiate. Immunohistochemistry suggests that the transplanted cells differentiate into fibroblastic, vascular, and epithelial phenotypes in their new microenvironment.

Our data support and amend the potential utility of SFCS as a biocompatible scaffold [7]. SFCS has been reported to have the potential to facilitate local vascular ingrowth [7]. Furthermore, our initial experiments demonstrated the ability of culture-expanded ASCs to adhere to a SFCS substrate in the range of 75% adhesion by 1 hour postseeding, with adherent stem cells occupying both surface and three-dimensional elements of the scaffold (unpublished data). The phenomenon of ASC-SFCS enhancing vascular in-growth was observed here, with SFCS enhancing vascular infiltration qualitatively versus no-graft controls in our subgroup analysis of whole tissue mounts excised at postoperative day 9. Quantitative analysis of stem cell-seeded grafts versus SFCS grafts further demonstrated vascular enhancement. The differentiation capability of seeded stem cells delivered on SFCS complements SFCS's attributes of biocompatibility.

Our findings regarding the planimetric analysis of wound healing are consistent with recent reports indicating enhanced wound closure in a similar murine system after application of bone marrow or adipose-derived mesenchymal stem cells [11, 12, 21]. The histology at 2 and 4 weeks reflects engraftment and spontaneous differentiation. Of particular interest is the observation that human ASCs are induced to differentiate along multiple lineages of tissue regeneration in the mouse microenvironment. This finding is in line with our previous results demonstrating that the transfection of human ASCs with liposome-enveloped xenogenic protein from a neonatal rat tissue preparation can induce differentiation of stem cells along the directed lineage [22]. Taken together, these observations support the hypothesis that the inductive biochemical and structural cues of the microenvironment are conserved across species and that a silk fibroin-chitosan delivery vehicle can provide a beneficial niche in supporting migration, proliferation, and differentiation of the applied cells.

In terms of the phenotypic fate of engrafted cells, it is notable that engrafted cells differentiate along two discrete germ lines, namely mesoderm, as seen with fibrovascular phenotypes, and an ectoderm lineage, as seen at 4 weeks with the participation of engrafted cells in epithelial proliferation. In vitro studies of these mesodermal markers among ASCs in previous studies have indicated a high degree of mesodermal marker positivity (SMA, HSP-47), whereas epithelial markers (CK19) are initially found to be negative (unpublished data). These data are consistent with the frequent finding of ASC differentiation into mesodermal lineages, as evidenced by fibrovascular elements 2 weeks postoperative both in our study and in recent reports [12, 21]. Furthermore, the delayed engraftment of the applied stem cells in epithelial growth confirms previous findings and might indicate a delayed infiltration of ASCs into the proliferating basal epithelial stratum [11]. This could also reflect a more complex and divergent route of differentiation, with ASCs finally contributing to both mesodermal and ectodermal phenotypes.

The clear engraftment of ASCs into regenerating tissue in our study differs from previous reports, where the engraftment of therapeutically introduced mesenchymal stem cells has been either unobservable or observable only at low levels [23]. The prolonged engraftment of ASCs seen in this study may be supported by humoral factors released by the ASCs. Previous findings from our group indicate that ASCs produce significant amounts of insulin-like growth factor-1 and vascular endothelial growth factor [24]. Other researchers have further stressed the importance of heterogeneous biochemical and paracrine effector functions among components of mesenchymal stem cell populations [14, 25, 26]. Qualitatively, ASCs are seen among engrafted wound bed biopsies in our system, although the quantitative extent of this engraftment was beyond the scope of this study. Given the potential that paracrine effects are the primary mediators of wound closure effects, it would be reasonable for future studies to quantify the extent of ASC engraftment in such a system. Such a follow-up analysis should increase our understanding of the relative contributions of paracrine versus direct cell differentiation effects and could also be logically extended to consider ASC-SFCS efficacy in the compromised wound setting.

In our model system of soft tissue injury and disruption, transient inflammation may facilitate the initial incorporation and differentiation of engrafted stem cells. Only trace remnants of SFCS fibers were observed at 2 weeks, consistent with degradation kinetics of this substrate within this approximate time frame. This time scale of SFCS degradation in vivo is generally consistent with previous work establishing the utility of this scaffold for soft tissue repair applications [7]. Although we have observed excellent biocompatibility of the stem cell carrier graft, the possibility that inflammatory effects could contribute to enhanced wound closure must be further considered as a possibility and cannot be fully ruled out. This scenario is unlikely, however, given the immunodeficient background of the nude mouse and the absence of histological findings consistent with ongoing inflammation in our 2-week wound bed biopsies.

Although our histology data solidly indicate engraftment of ASCs in wound bed biopsies, with differentiation into both fibrovascular and epithelial elements by 4 weeks, alternative explanations for this phenomenon must be considered. Specifically, the possibility of ASC fusion with vascular or epithelial elements cannot be completely ruled out. The impact of cell fusion versus direct differentiation on the beneficial effects of therapy with stem cells is not yet fully understood and is the subject of some controversy. However, evidence exists suggesting that fusion might not be the primary mechanism at work with respect to the benefits that tissue-derived cellular therapy confers in the context of tissue injury [27].

The functional improvement seen in our study, measured by accelerated rates of wound closure, likely reflects both paracrine and direct cellular mechanisms. The cytokine production of ASCs both in vitro and in pathological tissue settings has been well documented [12, 26, 28–30]. In terms of the enhanced vascular density seen in this study, paracrine mechanisms are probable functional mediators. Furthermore, with ASCs involved in neo-vascular bed establishment at the site of wound repair, there is likely also an autocrine loop functioning, with local ASCs producing angiogenic factors that act on themselves or neighboring ASCs in reestablishment of vascularization.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

In this study, we have successfully applied human ASC-seeded SFCS as a cytoprosthetic hybrid for reconstructive support in a cutaneous wound healing model. ASC-SFCS supports the delivery and engraftment of stem cells, as well as differentiation into fibrovascular and epithelial components. ASC-SFCS holds promise as a future delivery vehicle for stem cells in clinical reconstructive settings.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

We thank Jackie Furr and Sherita Daniel of the University of Texas M.D. Anderson Cancer Center core histology research laboratory for expert assistance in preparing histological sections. This work was supported in part by Grant 543102 from the Alliance of Cardiovascular Researchers (to E.U.A.) and by American Heart Association Southeast Affiliate Award 0555331B (to Y.-H.S.).

DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES

The authors indicate no potential conflicts of interest.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. ACKNOWLEDGMENTS
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES