Cell Transplantation, Vol. 23, pp. 285–301, 2014 Printed in the USA. All rights reserved. Copyright 2014 Cognizant Comm. Corp.
0963-6897/14 $90.00 + .00 DOI: http://dx.doi.org/10.3727/096368912X663541 E-ISSN 1555-3892 www.cognizantcommunication.com
Transplantation of Cyclic Stretched Fibroblasts Accelerates the Wound-Healing Process in Streptozotocin-Induced Diabetic Mice Eunkyung Lee,*1 Do Yeon Kim,*†1 Eunkyung Chung,* Eun Ah Lee,* Ki-Sook Park,‡§2 and Youngsook Son*§2 *Department of Genetic Engineering, College of Life Science and Graduate School of Biotechnology, Kyung Hee University, Yongin, Korea †St. Peter’s Hospital and R&D Center, Cell & Bio, Inc., Seoul, Korea ‡East-West Medical Research Institute, Kyung Hee University, Seoul, Korea §College of Medicine, Kyung Hee University, Seoul, Korea
Mechanical stimulation is a known modulator of survival and proliferation for many cells, including endothelial cells, smooth muscle cells, and bone marrow-derived mesenchymal stem cells. In this study, we found that mechanical strain prevents apoptosis and increases the adhesive ability of dermal fibroblasts in vitro and thus confers the survival advantage in vivo after transplantation of fibroblasts into the full-thickness wound of diabetic mice. Cyclic stretch at a frequency of 0.5 Hz and maximum elongation of 20% stimulates cellular survival mediated by the activation of extracellular signal-regulated kinases (ERKs), c-Jun N-terminal kinases (JNKs), and the serine/threonine kinase Akt (AKT). Stretching of the fibroblasts increases the synthesis of extracellular matrix proteins and the formation of denser focal adhesion structures, both of which are required for fibroblast adhesion. The stretched fibroblasts also upregulate the expression of vascular endothelial growth factor (VEGF) and stromal cell-derived factor-1a (SDF-1a), which enhanced wound healing in vivo. Indeed, preconditioning with mechanical stretch allows better survival of the transplanted fibroblasts, when compared to unstretched control cells, in the wound environment of mice with streptozotocin-induced diabetes and thus accelerates the wound-healing process in these mice. Keywords: Cell survival; Cyclic stretch; Mechanical stimulus; Angiogenesis; Wound healing
INTRODUCTION Human cells receive mechanical stimuli throughout their life from the beginning of the early developmental stage. Mechanical stimuli affect various fundamental cellular processes, such as development, migration, determination of mitotic spindle position, proliferation, apoptosis, adhesive interaction, the extracellular matrix (ECM), paracrine factor production, and pathogenesis (6,10,14,16,17,20,24,31,48,53). Current studies on mechan ical stimuli focus on the molecular mechanisms by which cells sense the mechanical stimulus, transduce the mechanical signal into the intracellular chemical signal, and respond to the mechanical stimulus in a physiological environment (41,50). Many clinical trials on the transplantation of cells isolated from various tissues into wound sites have been performed (23). Survival and engraftment of the transplanted
cells are very critical to successful cell therapy. However, the rate of survival/engraftment is remarkably low owing to the unfavorable wound environment caused by inflammation and low blood flow following poor angiogenic response. One typical example of an unfavorable wound environment is the diabetic wound (15). Poor skin wound healing in diabetes has been thought to result in part from impaired neovascularization (7,15,29). Several studies yielded preliminary results to support wound-healing enhancement by transplantation of various stem cells or precursors, such as circulating endothelial precursor cells (EPCs) and bone marrow-derived mesenchymal stem cells, into diabetic skin wounds (3,5,44,46). The transplanted cells are able to heal the diabetic wounds through differentiation into vascular and nonvascular cells and/or production of paracrine factors. For example, transplanted EPCs stimulate the migration of endothelial cells by
Received July 19, 2012; final acceptance January 4, 2013. Online prepub date: February 4, 2013. 1 These authors provided equal contribution to this work and are co-first authors. 2 These authors provided equal contribution to this work as co-corresponding authors. Address correspondence to Youngsook Son, Ph.D., Department of Genetic Engineering, Kyung Hee University, Seocheondong, Kiheung-gu, Yongin, 446-701, Republic of Korea. Tel: +82-31-201-3822; Fax: +82-31-206-3829; E-mail:
[email protected] or Ki-Sook Park, Ph.D., East-West Medical Research Institute/College of Medicine, 25 Kyungheedae-ro, Dongdaemun-gu, Seoul, 130-701, Republic of Korea. Tel:+82-2-958-9368; Fax:+82-2-958-9083; E-mail:
[email protected]
285
286 lee ET AL.
releasing vascular endothelial growth factor (VEGF) and stromal derived factor-1 (SDF-1), both of which enhance angiogenesis (5,22,49). However, a major problem in cell therapy for diabetes is low survival rates of transplanted cells in the endogenous diabetic wound environment due to factors such as low oxygen level, high glucose level, and accumulated reactive oxygen species (ROS) (4,30,47). Therefore, several trials to promote the survival of the transplanted cells have been attempted (23). A strategy to enhance the survival of transplanted cells might be to precondition the cells with mechanical strain prior to transplantation because mechanical strain affects cellular proliferation and apoptosis of various cells, including human mesenchymal stem cells (19,21), endothelial cells (27,37), and human skin keratinocytes (52). The mechanical strain is transduced inside the cells through various intracellular signals to affect apoptosis and proliferation (20). Mechanical strain has also been reported to induce proliferation of human dermal fibroblasts through the activation of extracellular signal-regulated kinase (ERK), p38 mitogen-activated protein kinase (MAPK), and V-akt murine thymoma viral oncogene homolog 1 (AKT) (36). However, whether mechanical strain increases the cellular survival of dermal fibroblasts remains unclear. Adult skin wound healing is a complex process requiring the collaborative efforts of several different types of cells. During the wound-healing process, each of the various types of cells contributes to proliferation, migration, and contraction, as well as the synthesis of growth factors, matrix elements, and proteases at a wound site (34). In particular, dermal fibroblasts play essential roles in the repair of skin wounds through remodeling of the wound bed by synthesis of new ECM and growth factors and the formation of thick actin bundles (43). In this study, we examined the effects of cyclic equibiaxial stretch on the cellular proliferation, apoptosis, adhesive ability, and paracrine factor production of human dermal fibroblasts. Moreover, we evaluated whether preconditioning of fibroblasts with mechanical stretch provided an advantage in cell survival following transplantation and accelerated the wound-healing process in diabetic mice. MATERIALS AND METHODS Primary Cell Culture Human dermal fibroblasts (hDFs) isolated from circumcised foreskins were purchased from MCTT (Seoul, Korea), and mouse enhanced green fluorescent protein (EGFP)–dermal fibroblasts (mDFs) were isolated from 8- to 10-week-old FVB.Cg-Tg(ACTB-EGFP)B5Nagy/J strain [EGFP directed to widespread tissues by the cytomegalovirus-immediate early (CMV-IE) enhancer/ chicken b-actin/rabbit b-globin hybrid promoter strain of Friend leukemia virus, strain B] mice (The Jackson
Laboratories, Bar Harbor, ME, USA) through the treatment of type I collagenase (Washington Biochemical Corp., Freehold, NJ, USA). The fibroblasts were cultured in fibroblast growth media (FGM2) containing 2% fetal bovine serum, 5 mg/ml of insulin, 1 ng/ml of basic fibroblast growth factor (bFGF; all from Lonza, Walkersville, MD, USA) at 37°C in a humidified atmosphere of 5% CO2. The medium was exchanged every 2 days, and the cells were split at 80–90% confluent state by 0.05% trypsin/0.02% ethylenediaminetetraacetic acid (EDTA; Invitrogen, Carlsbad, CA, USA) treatment. The cells at passages between four and six were used for this study. The experiments were approved by the animal experiment ethics committee of Kyung Hee University [KHUASP(SU)-07-007]. Application of the Cyclic Equibiaxial Stretch to the Cells hDFs and mouse EGFP–mDFs were plated at a density of 1.5 × 105 cells per well on the type I collagenprecoated silicon membrane of the six-well plate (Flexcell Inc., Hillsborough, NC, USA) and allowed to attach for 24 h in FGM2. If necessary, the cells were starved with the serum-free fibroblast basal media (FBM) (Lonza, San Diego, CA, USA) for 24 h before applying the strain. The cells on the flexible silicone membrane repeti tively received the cyclic equibiaxial stretch for 24 h at 0.5 Hz frequency and 20% maximum elongation according to the manufacturer’s manual for the Flexercell® FX-4000™ apparatus (Flexcell Inc.) as described in Figure 1A and B. With the Flexercell apparatus, the diameter of the circular silicone membrane is enlarged 1.2-fold by periodic vacuum loadings, which could result in a 1.44-fold increase of total cell culture surfaces (pr2). The cells at the center of the circle do not receive any stretch stimulus, but the cells at the periphery receive maximal stretch (Fig.1C, D). The stretch effects, such as apoptosis and bromodeoxyuridine (BrdU) incorporation, were observed at the two-thirds distant area from the center of the circle, and all of the fibroblasts on the silicon membrane were used both in in vitro replating experiments and in vivo experiments. Western Blot Analysis For Western blot analysis, the cells were rinsed with ice-cold phosphate-buffered saline (PBS; WelGENE, Daegu, Korea) three times and disrupted by 100 μl of 2× lysis buffer (40 mM Tris-HCl pH 7.5, 300 mM NaCl, 2 mM Na2EDTA, 2 mM EGTA, 2% Triton-X 100, 5 mM sodium pyrophosphate, 2 mM b-glycerophosphate, 2 mM Na3VO4, and 2 μg/ml leupeptin, all from Cell Signaling Technology Inc., Beverly, MA, USA) per well of six-well plates. The cell lysates were incubated on ice for 20 min and centrifuged at 10,000 × g for 30 min at 4°C, and the supernatants were saved. Total protein amount was measured by Quick Start™ Bradford dye reagent (BioRad
STRETCHED FIBROBLASTS ACCELERATE WOUND HEALING
287
Figure 1. A device for in vitro application of the cyclic stretch and the analysis of radial distance-dependent orientation of dermal fibroblast applied with the stretch. The operating principle (A) and the top view (B) of the Flexcell® FX-4000™ Tension System are shown. (C) Cells and actin fibers were reorganized by the cyclic stretch. The actin fibers were parallel to the orientation of the cells applied with the cyclic stretch. Double-head arrows indicate the direction of the stretch. The cells were stained with tetramethylrhodamine isothiocyanate (TRITC)-conjugated phalloidin in order to show the actin fibers. (D) Orientation of cells was determined at several distances from the center of the silicone membrane after the application of stretch. More than 400 cells were analyzed to determine the orientation of cells at each distance, and the results are shown by mean ± SD. The broken line indicates the radial strain at each distance, while the stretch is applied.
288 lee ET AL.
Lab., Inc., Hercules, CA, USA), and the same amount of proteins of each sample was denatured and resolved by 10% SDS polyacrylamide gel electrophoresis (SDSPAGE; Sigma, St. Louis, MO, USA) and electrophoretically transferred to nitrocellulose membranes (Protran®, Whatman, GE Healthcare, Freiburg, Germany). Blots were blocked by filtered 5% nonfat dry milk (BD Pharmingen, San Jose, CA, USA) in 200 mM Tris-buffer saline pH 7.6 (TBS; Sigma), incubated with primary antibodies in TBS containing 0.1% Tween-20 (Sigma) overnight at 4°C, incubated with horseradish peroxidase (HRP)conjugated secondary antibody (BioRad Lab., Inc.), and visualized by the ECL detection system (Santa Cruz Inc., St. Louis, MO, USA). Primary antibodies against panERKs, phospho-ERKs, phospho-AKT, and phosphoc-Jun N-terminal kinases (JNKs) were purchased from Cell Signaling Technology, Inc., and an antibody against a-tubulin was purchased from Sigma. For Western blot analysis of fibronectin and type I collagen, samples were prepared by the use of the collected culture media and separated in 8% Tris-HCl SDS PAGE gels (Sigma). The number of the cells in each culture sample had been determined, and it had been used as the internal control to determine the volume of media that was used for Western blot analysis. Clone HFN7.1 hybridoma (American Type Culture Collection, Rockville, MD, USA) culture supernatant was used for Western blot of fibronectin, and antitype I collagen antibody was purchased from Biodesign Inc. (Kennebunk, Maine, Germany). Terminal Deoxynucleotidyltransferase-Mediated dUTPFITC Nick-End Labeling Assay (TUNEL Assay) Apoptotic cells were detected by TUNEL assay with fluorescein system (Roche, Indianapolis, IN, USA). The cells were washed with PBS three times and fixed in 3.7% paraformaldehyde (Sigma) in PBS for 20 min at 4°C. After extensive washing with PBS, the cells were permeabilized with 0.2% Triton X-100 (Sigma) in PBS for 5 min, incubated at 37°C for 60 min with an equilibration buffer containing 2-deoxynucleotide 5¢-triphosphate and terminal deoxynucleotidyltransferase (TdT) enzyme, according to the manufacturer’s protocol. After the TUNEL assay, the cells were counterstained with 4¢,6-diamidino-2phenylindole (DAPI; Sigma), mounted with Vectashield mounting medium (Vector Lab., Burlingame, CA, USA), and immediately examined by fluorescence microscope (Leica Microsystems, Inc., Bannockburn, IL, USA). For the quantitative analysis of apoptotic cells, 12 fields per sample were analyzed (~400 cells/field). The cells showing green fluorescence in the nucleus were counted as apoptotic cells. Results were expressed as a percentage of apoptotic cells among DAPI-positive cells.
5-Bromo-2¢-Deoxyuridine (BrdU) Incorporation Assay The cellular proliferation was measured by DNA incorporation of BrdU. Briefly, the cells were labeled with 20 μM BrdU (Sigma) for 24 h of the mechanical stretch application. After fixing the cells as described above, the cells were incubated with 2 N HCl (Duksan Pure Chemical Company, Ansan, Korea) for 30 min, washed with PBS, incubated with mouse anti-BrdU antibody (Roche, Indianapolis, IN, USA), incubated with fluorescein-conjugated secondary antibody (Roche), counterstained with propidium iodide (PI; Sigma), and mounted with Vectashield mounting medium. The cells with green fluorescence in the nucleus were counted as proliferating cells. For the quantification of proliferating cells, 12 fields per sample were analyzed (~400 cells/field). Results were expressed as a percentage of BrdU-positive cells among PI-positive cells. Inhibitor Study PD98059 (Millipore Corporation, Billerica, MA, USA), a specific inhibitor of the ERK1/2 activation by mitogenactivated protein kinase kinase (MAPKK or MEK), LY294002 (Millipore), a selective inhibitor of phosphatidylinositol 3-kinase (PI3K), and SP600125 (Millipore), an anthrapyrazolone inhibitor of Jun N-terminal kinase (JNK), were dissolved in dimethyl sulfoxide (DMSO; Sigma). The cells were preincubated with PD98059 (100 μM), LY294002 (10 μM), and SP600125 (1 μM) for 1 h prior to the application of the cyclic stretch. ELISA of VEGF and SDF-1a in Culture Supernatant of hDFs During Mechanical Stretch Application VEGF or SDF-1a in the culture supernatants was measured by sandwich ELISA purchased from R&D Systems (Abingdon, UK). The human recombinant VEGF, SDF-1a, or test samples were added to the wells and then incubated for 2 h at room temperature. The plates were incubated with HRP-conjugated polyclonal antihuman VEGF or SDF-1 antibody at room temperature for 2 h. The color reaction was induced by the addition of substrate solution and was stopped 20 min later by the addition of 2 N of sulfuric acid. An automated microplate reader (BioRad Lab. Inc.) was used to measure the optical density (O.D.) at a wavelength of 450 nm. Between each of these steps, the plates were washed four times with PBS containing 0.1% Tween 20. Immunocytochemistry hDFs cultured on the type I collagen precoated silicone membrane were fixed with 3.7% formaldehyde for 20 min and permeabilized with 0.2% Triton X-100 in PBS for 5 min at room temperature. The silicone membrane of a well was subdivided into six pieces and
STRETCHED FIBROBLASTS ACCELERATE WOUND HEALING
289
stained using a general procedure for immunocytochemistry. Immunocytochemistry was performed by the use of the following primary antibodies: anti-a-smooth muscle actin (a-SMA; Dako, Burlingame, CA, USA), anti-atubulin (Sigma), or anti-vinculin (Sigma). In order to stain the actin fibers, cells were incubated with tetramethylrhodamine isothiocyanate (TRITC)-conjugated phalloidin (Sigma). Images were collected using a confocal microscope (Leica Microsystems, Inc.).
of the frozen tissue crossing the longest diameter of the wound was obtained. The entire wound site of the section was subdivided into four divisions, and EGFP-positive cells were counted under the fluorescence microscope (Leica Microsystems, Inc.).
Induction of Experimental Diabetes Streptozotocin (STZ, Sigma) at a dose of 50 mg/kg in 50 mM sodium citrate (Sigma), pH 4.5, was intraperitoneally injected daily for 5 days to wild-type male FVB mice at 8–12 weeks of age in order to induce the experimental diabetes. At 16 days after the final injection, serum glucose was measured from the blood of the tail vein using the glucose meter, Accucheck (Roche, Indianapolis, IN, USA). Mice were considered diabetic if they had hyperglycemia (>300 mg/dl), and it was checked weekly that the mice remained hyperglycemic for at least 5 weeks. At 3 weeks after the final injection of STZ, the full-thickness wound was made on the dorsal skin of the mice with hyperglycemia of 300~400 mg/dl to perform transplantation experiments. Transplantation of EGFP–Dermal Fibroblasts in Wound The full-thickness excision wound was made on the dorsal skin of the STZ-induced diabetic mouse using an 8-mm-diameter punch (Kai Europe GmbH, Solingen, Germany) biopsy while sparing the underlying panniculus carnosus muscle after depilation under anesthesia with ketamine (106.25 μg/kg; YuhanYanghang, Seoul, Korea). In order to transplant mouse EGFP–fibroblasts applied with the stretch or not, all of the fibroblasts on the silicon membrane were harvested with treatment of 0.3% type I collagenase. At a density of 6 × 105 cells/40 μl, the wound received a mixture of poloxamer F127 (Sigma) hydrogel (or fibrin glue; Greenplast®; Green Cross Corp., Yongin, Korea) and mouse EGFP–fibroblasts, which had been applied with the stretch or not. Wound sites were covered with Mepitel (Mölnlyke Health Care AB, Gothenburg, Sweden) and then Tegaderm (3M, London, ON, Canada), and the animals were kept in individual cages. At 2 or 7 days posttransplantation, the mice were euthanized by inhalation of CO2 gas. The wound tissues, including normal adjacent tissue and fascia, were dissected out. Then, the wounds were prefixed with 3.7% formaldehyde solution in PBS overnight and divided into two pieces, which were embedded in paraffin or OCT compound (Shandon Cryomatrix™, Thermo Fisher Scientific Inc., Waltham, MA, USA), respectively. A series of serial sections (10 μm)
Histological Scoring of Wound Healing After fixation of the wound tissues in formalin, the samples were embedded in paraffin and sectioned (5 μm). For histological evaluation, sections were deparaffinized and rehydrated followed by Masson’s trichrome staining (Sigma) or hematoxylin/eosin (H&E; Sigma) staining. In order to evaluate neovascularization, sections were subjected to immunohistochemical staining against Griffonia simplicifolia 1 (GS-1) lectin (Sigma). All slides were then evaluated under light microscopy by two independent observers, using a semiquantitative score to evaluate vascularity, granulation, and dermal/epidermal regeneration as previously described (1). We used four-point scales to evaluate vascularity (1, severely altered angiogenesis with one or two vessels per site, endothelial edema, thrombosis, and/or hemorrhage; 2, moderately altered angiogenesis with three to four vessels per site, moderate edema and hemorrhage, but absence of thrombosis; 3, mildly altered angiogenesis with five to six vessels per site, moderate edema, but absence of thrombosis and hemorrhage; and 4, normal angiogenesis with more than seven vessels per site with only mild edema but absence of thrombosis and hemorrhage) and granulation tissue formation (1, thin granulation layer; 2, moderate granulation layer; 3, thick granulation layer; and 4, very thick granulation layer) and a three-point scale to evaluate dermal and epidermal regeneration (1, little regeneration; 2, moderate regeneration; and 3, complete regeneration). Statistical Analysis Data were presented as means ± standard deviation. Statistical analysis was performed using t test and calculating one-tailed p values. If necessary, differences between experimental groups were compared by ANOVA followed by Bonferroni’s test. Significance of in vivo transplantation experiments was determined with the Wilcoxon signed-rank test. Values of p
