**2. Methods**

estimates-of-diabetes-and-its-burden-in-the-united-states.pdf). This burden of disease is the result of a progressive disease process associated with numerous complications, such as retinopathy, neuropathy, and nephropathy, as well as the development of chronic wounds. Chronic wounds remain the leading cause of hospital admissions and nontraumatic lower extremity amputations in patients with diabetes, with nearly 80% of all amputations performed in patients with diabetes preceded by a diabetic wound [1]. Risk factors for the development of diabetic wounds include the presence of the "pathogenic triad" of neuropathy, ischemia, and trauma [2–4]. In addition, foot deformities, lower extremity edema, and use of inappropriate

Despite clinical strategies aimed at prevention and early detection of diabetic neuropathy and diabetic wounds, patients with diabetes continue to have a 12–25% lifetime risk of developing a chronic diabetic wound due to poor patient compliance and progression of the diabetic phenotype [5, 6]. Projections indicate that by 2025 diabetes will affect over 300 million people worldwide, with a rising proportion of the burden borne by patients in developing countries [7, 8]. Given this anticipated rise in the burden of disease attributable to diabetes, and the potential concomitant rise in development of chronic wounds, the impetus for developing

After injury, patients with diabetes suffer from an impaired wound healing response; the standard wound healing response, in which the tissue passes through consecutive, but overlapping phases of coagulation, inflammation, proliferation, and remodeling, is disrupted [2]. Notably, patients with diabetes demonstrate a phenotype characterized by decreased angiogenesis, impaired leukocyte migration, decreased growth factor production, sustained inflammation, impaired fibroblast function, and imbalance of extracellular matrix deposition and remodeling, and delayed wound healing [9]. A longer duration of diabetes has been associated with a greater risk of impaired wound healing; consistent with this observation, duration of diabetes has been correlated with more profound diabetic neuropathy, increasingly compromised biomechanical properties of diabetic skin, and an increasingly delayed rate of endothelial progenitor cell proliferation. While the cytokines, chemokines, and cellular components of the wound healing response have been extensively studied, recent attention has focused upon the role that microRNAs (miRNAs) play in the impaired healing of diabetic

MiRNAs are a class of small, noncoding RNA molecules, 20-22 nucleotides long, that regulate gene expression at the posttranscriptional level [10]. Complementary binding of miRNAs to the 3'-untranslated region (UTR) of their target messenger RNA (mRNA) results in posttranscriptional repression and/or mRNA degradation, thereby regulating the expression of downstream targets [11]. MiRNA are thought to regulate over one-third of all physiologic processes, including regulation of cell cycle progression, apoptosis, cytokine transcription, cellular metabolic function, signal transduction, proliferation, and determination of cell fate [12, 13]. MicroRNAs are being investigated for their role as biomarkers, diagnostic tools, and therapies in a variety of disease states, including diabetes and numerous types of cancer [14, 15]. Key to the development of the diabetic phenotype is the dysregulation of the expression of miRNA that regulate inflammation, extracellular matrix composition, and angiogenesis.

footwear also contribute to the development of diabetic wounds [2–4].

4 Wound Healing - New insights into Ancient Challenges

more effective wound care strategies cannot be understated.

wounds [3].

### **2.1. Murine model of diabetic wound healing**

We utilized a murine model of diabetes, using genetically diabetic female mice homozygous for the leptin receptor mutation (db/db, C57BKS.Cg‐m‐/‐*Leprdb/*J) and age‐matched, nondia‐ betic, heterozygous controls (db/‐), obtained from the Jackson Laboratory (Bar Harbor, ME). Leptin is a centrally acting hormone that modulates both satiety and body fat content [19, 20]. Mice homozygous for the leptin receptor mutation develop hyperphagia, obesity, hypergly‐ cemia, insulin resistance, and hyperlipidemia, leading to their extensive use as a murine model of diabetes [19, 20]. Between 4 and 6 weeks of age, the diabetic mice are significantly more obese and hyperglycemic than their nondiabetic counterparts (**Figure 1**), and by 10–14 weeks of age, diabetic mice weigh greater than 45 g, with serum glucose levels in excess of 400 mg/dL, whereas nondiabetic mice weighed less than 25 g, with serum blood glucose levels less than 250 mg/dL [21]. Most noteworthy for our analysis, diabetic mice also demonstrate significantly delayed wound healing when compared with nondiabetic mice, requiring up to 5 days longer to close cutaneous wounds (**Figure 2**) [21, 22].

**Figure 1. Weight (A) and serum blood glucose (B) of diabetic and nondiabetic mice. (A)** Weight of diabetic (C57BKS.Cg‐m‐/‐*Leprdb/*J) and nondiabetic (C57BKS.Cg‐m‐/+*Leprdb/*J) mice at different ages. **(B)** Serum blood glucose levels in diabetic and nondiabetic mice at different ages. Asterisk (\*) indicates *p* < 0.05 for any given time point (Image reproduced from Zgheib *et al.* [21] with permission of the authors.)

**Figure 2. Wound closure in diabetic and nondiabetic mice.** The percent (%) of original wound area remaining at vari‐ ous time points during the wound healing process in diabetic and nondiabetic mice. Asterisk (\*) indicates *p* < 0.05 and (\*\*) indicates *p* < 0.001.

To obtain a baseline, unwounded skin from the dorsum of diabetic and nondiabetic mice was harvested, and the cranial‐caudal orientation was preserved for the purposes of biomechanical testing. Diabetic and nondiabetic mice between 10 and 14 weeks of age had full thickness, excisional dermal wounds created using an 8 mm dermal punch biopsy (Miltex, Inc., York, PA) through the panniculus carnosus, as previously described [17, 21]. Following the initial wounding, mice were treated with either phosphate buffered saline (PBS), 106 mesenchymal stem cells (MSCs),and 1 × 106 plaque forming units (PFU) of either an empty lentiviral vector or a lentiviral construct containing a mutated stromal cell derived factor (SDF‐1α) transgene, or 10 μL of 100 ng/mL of recombinant human SDF‐1α protein. The wounded skin was harvested at 1, 3, 7, 14, and 21 days after initial wounding and animals were euthanized by inhalation of carbon dioxide followed by cervical dislocation.

### **2.2. Isolation of MSCs**

MSCs were isolated from the femurs and tibia of adult C57BL/6TbN (act‐GFP) OsbY01 transgenic (GFP) mice, as described previously [17]. Mononuclear cells were then separated by density gradient separation using Ficoll before being suspended at a density of 2.5 × 104 cells/μL in PBS [23].

### **2.3. Human skin analysis**

Human skin samples were obtained via the National Disease Research Interchange (NDRI). Human skin samples measuring 5 × 5 cm were obtained from the anterior portion of the lower extremities within 8 h postmortem. The samples were obtained from patients who were between 45 and 75 years old, without significant comorbidity, malignancy, or a history of radiation or chemotherapy. While it was known whether or not the patients carried an existing diagnosis of diabetes, as well as what medications the individual was currently taking, data regarding the duration of diabetes, degree of blood glucose control, or intensity of sun exposure was not available. Following receipt of the sample, subcutaneous tissues were sharply excised from the dermis and the skin samples intended for biomechanical analysis were immediately flash frozen in liquid nitrogen.

### **2.4. Culture of dermal fibroblasts**

Human skin samples and murine skin samples were processed in order to enable the culture of dermal fibroblasts for further *in vitro* analysis. Skin samples were washed in 70% ethanol for 2 min, followed by 3 washes in PBS. Subcutaneous tissues were sharply excised from overlying dermis, and the remaining dermis was minced before being digested in 2 mL of 1000 u/mL Collagenase for 1 h at 37°C. The sample was then centrifuged at 1000 rpm for 5 min, the supernatant was removed, and the sample was then digested in 5 mL of dispase (1.9 u/mL) for 30 min at 37°C. The sample was again centrifuged at 1000 rpm for 5 min, the supernatant was removed, and the remaining pieces of dermis were mixed with 10 mL of DMEM with 10% fetal bovine serum (FBS, Life Technologies, Calrsbad, CA) and 1% antibiotic‐antimycotic (Life Technologies, Calrsbad, CA) and plated in a 100 mm tissue culture plate (Corning Incorporate, Corning, NY). When fibroblasts were observed to proliferate and adhere to the tissue culture plate independent of the pieces of dermis, the dermis was removed, the plate was washed twice with 10 mL PBS, and the culture media was replaced. Cells between passage 1 and 4 were used for these experiments.

### **2.5. Lentiviral construction**

**Figure 2. Wound closure in diabetic and nondiabetic mice.** The percent (%) of original wound area remaining at vari‐ ous time points during the wound healing process in diabetic and nondiabetic mice. Asterisk (\*) indicates *p* < 0.05 and

To obtain a baseline, unwounded skin from the dorsum of diabetic and nondiabetic mice was harvested, and the cranial‐caudal orientation was preserved for the purposes of biomechanical testing. Diabetic and nondiabetic mice between 10 and 14 weeks of age had full thickness, excisional dermal wounds created using an 8 mm dermal punch biopsy (Miltex, Inc., York, PA) through the panniculus carnosus, as previously described [17, 21]. Following the initial

or a lentiviral construct containing a mutated stromal cell derived factor (SDF‐1α) transgene, or 10 μL of 100 ng/mL of recombinant human SDF‐1α protein. The wounded skin was harvested at 1, 3, 7, 14, and 21 days after initial wounding and animals were euthanized by

MSCs were isolated from the femurs and tibia of adult C57BL/6TbN (act‐GFP) OsbY01 transgenic (GFP) mice, as described previously [17]. Mononuclear cells were then separated by density gradient separation using Ficoll before being suspended at a density of 2.5 × 104

Human skin samples were obtained via the National Disease Research Interchange (NDRI). Human skin samples measuring 5 × 5 cm were obtained from the anterior portion of the lower extremities within 8 h postmortem. The samples were obtained from patients who were between 45 and 75 years old, without significant comorbidity, malignancy, or a history of radiation or chemotherapy. While it was known whether or not the patients carried an existing diagnosis of diabetes, as well as what medications the individual was currently taking, data regarding the duration of diabetes, degree of blood glucose control, or intensity of sun

plaque forming units (PFU) of either an empty lentiviral vector

mesenchymal

wounding, mice were treated with either phosphate buffered saline (PBS), 106

inhalation of carbon dioxide followed by cervical dislocation.

(\*\*) indicates *p* < 0.001.

stem cells (MSCs),and 1 × 106

6 Wound Healing - New insights into Ancient Challenges

**2.2. Isolation of MSCs**

cells/μL in PBS [23].

**2.3. Human skin analysis**

The SDF‐1α mutant that was utilized binds the CXCR4 receptor without activating it [24, 25]. According to the manufacturer's instruction, a complementary DNA (cDNA) library was prepared from mouse tissues using TRIzol and Superscript (Invitrogen, Carlsbad, CA). After sequence analysis confirmed the SDF1α cDNA and the SDF1α inhibitor (SDFαi), the CS‐CG HIV‐1 transfer plasmid was used to generate a self‐inactivating lentiviral vector designed to convey expression of either the green fluorescent protein reporter gene (GFP, Clontech Laboratories, Mountain View, CA) or the mutant SDF1αi with the GFP reporter gene under the control of a cytomegalovirus promoter [25–27]. The ability of the plasmid to effectively transfect cells was tested on murine dermal fibroblasts *in vitro* prior to *in vivo* use.

### **2.6. Biomechanical testing**

In order to assess the biomechanical properties of murine and human diabetic and nondiabetic skin, all samples immediately underwent biomechanical testing after harvesting. Biomechan‐ ical testing was performed on the diabetic and nondiabetic murine samples at baseline, 4 weeks, and 7 weeks after wounding. Prior to testing, the subcutaneous tissues were removed, and a uniform dumbbell‐shaped testing unit was stamped out from the sampled skin [28, 29]. The cranial–caudal orientation of each sample was maintained, and the healed wound, if present, was centrally located within the testing unit. Two lines of Verhoeff stain were placed on either end of the sample, thereby identifying the length of the testing area. The mean cross‐ sectional area of each sample was then measured using a custom laser band [30]. Each sample underwent testing to calculate the modulus of elasticity and the maximum stress until failure, as previously described [29].

### **2.7. Immunohistochemistry**

Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide in methanol. GFP was detected with application of rabbit anti‐GFP primary antiserum (1:100; Invitrogen, Carlsbad, CA) followed by biotinylated antirabbit secondary antibody (1:200), and the slides were developed using avidin–biotin complex (Vector Laboratories, Burlingame, CA). Slides were counterstained with hematoxylin. For double‐immunofluorescent staining, sections were blocked with 1% sodium borohydrate (Sigma–Aldrich, St. Louis, MO) in PBS. Platelet endo‐ thelial cell adhesion molecule‐1 (CD31) was used an endothelial marker. CD31 was detected by using rat anti‐mouse CD31 antibody (1:20 dilution; Invitrogen, Carlsbad, CA and AbCam, Cambridge, MA) followed by biotinylated rabbit anti‐rat secondary antibody and detection with the alkaline phosphatase detection system (Vector Laboratories, Burlingame, CA). GFP was detected by using the same rabbit anti‐GFP primary antiserum followed by Alexa‐ Fluor‐488‐conjugated F (ab') 2 secondary antibodies (1:100; Invitrogen). Slides were counter‐ stained with 4,6‐diamidino‐2‐phenylindole (Vector Laboratories, Burlingame, CA).

Sections were stained for cluster of differentiation (CD) 45, which was used as a cellular marker for inflammatory cells, and antigen retrieval was performed using 13 Antigen Retrieval Citra (BioGenex HK086‐9k, Fremont, CA) in a decloaking chamber using the factory default settings (Biocare Medical, Concard, CA). Slides were rinsed three times in 0.1% PBST (0.1% Triton100 in PBS), blocked in 20% goat serum in 0.4% PBST for 60 min at room temperature, and incubated overnight at room temperature with primary antibodies against CD 45 (rabbit polyclonal ab10558, 1:100; Abcam, Cambridge, MA) in 5% goat serum and 0.1% PBST. The following day, slides were washed three times in 0.1% PBST and incubated with the appro‐ priate anti‐rabbit biotinylated secondary antibodies for 1 h at room temperature. Slides were washed three times in 0.1% PBST and mounted in Vectashield (Vector Laboratories, Burlin‐ game, CA). The slides were then incubated with avidin–biotin–peroxidase complex (Vector Laboratory) and developed, as described by the manufacturer. The mean number of either CD45 or CD31 positive cells per high power field (hpf) was calculated as the average of 5 hpf per slide.

### **2.8. Enzyme-linked immunosorbent assay and Western Blot**

Skin samples had their subcutaneous tissues removed before being flash frozen in liquid nitrogen immediately after harvest. Collagen content was quantified by Western Blot. Skin samples were cut into 1‐mm pieces and homogenized in a 0.5 M acetic acid solution containing 1× protease inhibitor cocktail and 5 mmol/L EDTA, in order to solubilize total collagen. The solution was then centrifuged for 4 h at 4°C at 16.1 × 103 G, the lipid layer was aspirated, and this process was repeated. Bicinchoninic acid protein assay was used to quantify the total protein concentration of the supernatant, and Western blot was performed using standard techniques, and final protein concentrations were standardized to 1 μg/2 μL using a sample buffer before being boiled at 95°C for 5 min. Tris–acetate gels (3–8%) were run at 150 V for 1 h, transferred to 30 V overnight at 4°C, rinsed with Tris-buffered saline, and were subsequently blocked with 5% milk in Tris-buffered saline with 0.1% Tween for 1 h. Collagen I antibody (Abcam Inc., Cambridge, MA) was diluted to 1:2000 (collagen III to 1:1000) and blots were incubated for 1 h at room temperature, washed with Tris-buffered saline with 0.1% Tween, and incubated with secondary antibody (anti-rabbit IgG horseradish peroxidase; GE Healthcare, Marlborough, MA) at 1:10,000 for 1 h at room temperature. Western blots were washed with Tris-buffered saline with 0.1% Tween and then Tris-buffered saline. The blots were then developed using a standard chemiluminescence solution (ECL solution A and B; GE Healthcare, Marlborough, MA) and incubated for 1 min.

### **2.9. Real-time polymerase chain reaction**

underwent testing to calculate the modulus of elasticity and the maximum stress until failure,

Endogenous peroxidase activity was blocked with 0.3% hydrogen peroxide in methanol. GFP was detected with application of rabbit anti‐GFP primary antiserum (1:100; Invitrogen, Carlsbad, CA) followed by biotinylated antirabbit secondary antibody (1:200), and the slides were developed using avidin–biotin complex (Vector Laboratories, Burlingame, CA). Slides were counterstained with hematoxylin. For double‐immunofluorescent staining, sections were blocked with 1% sodium borohydrate (Sigma–Aldrich, St. Louis, MO) in PBS. Platelet endo‐ thelial cell adhesion molecule‐1 (CD31) was used an endothelial marker. CD31 was detected by using rat anti‐mouse CD31 antibody (1:20 dilution; Invitrogen, Carlsbad, CA and AbCam, Cambridge, MA) followed by biotinylated rabbit anti‐rat secondary antibody and detection with the alkaline phosphatase detection system (Vector Laboratories, Burlingame, CA). GFP was detected by using the same rabbit anti‐GFP primary antiserum followed by Alexa‐ Fluor‐488‐conjugated F (ab') 2 secondary antibodies (1:100; Invitrogen). Slides were counter‐

stained with 4,6‐diamidino‐2‐phenylindole (Vector Laboratories, Burlingame, CA).

**2.8. Enzyme-linked immunosorbent assay and Western Blot**

solution was then centrifuged for 4 h at 4°C at 16.1 × 103

Sections were stained for cluster of differentiation (CD) 45, which was used as a cellular marker for inflammatory cells, and antigen retrieval was performed using 13 Antigen Retrieval Citra (BioGenex HK086‐9k, Fremont, CA) in a decloaking chamber using the factory default settings (Biocare Medical, Concard, CA). Slides were rinsed three times in 0.1% PBST (0.1% Triton100 in PBS), blocked in 20% goat serum in 0.4% PBST for 60 min at room temperature, and incubated overnight at room temperature with primary antibodies against CD 45 (rabbit polyclonal ab10558, 1:100; Abcam, Cambridge, MA) in 5% goat serum and 0.1% PBST. The following day, slides were washed three times in 0.1% PBST and incubated with the appro‐ priate anti‐rabbit biotinylated secondary antibodies for 1 h at room temperature. Slides were washed three times in 0.1% PBST and mounted in Vectashield (Vector Laboratories, Burlin‐ game, CA). The slides were then incubated with avidin–biotin–peroxidase complex (Vector Laboratory) and developed, as described by the manufacturer. The mean number of either CD45 or CD31 positive cells per high power field (hpf) was calculated as the average of 5 hpf

Skin samples had their subcutaneous tissues removed before being flash frozen in liquid nitrogen immediately after harvest. Collagen content was quantified by Western Blot. Skin samples were cut into 1‐mm pieces and homogenized in a 0.5 M acetic acid solution containing 1× protease inhibitor cocktail and 5 mmol/L EDTA, in order to solubilize total collagen. The

this process was repeated. Bicinchoninic acid protein assay was used to quantify the total protein concentration of the supernatant, and Western blot was performed using standard techniques, and final protein concentrations were standardized to 1 μg/2 μL using a sample buffer before being boiled at 95°C for 5 min. Tris–acetate gels (3–8%) were run at 150 V for 1 h,

G, the lipid layer was aspirated, and

as previously described [29].

8 Wound Healing - New insights into Ancient Challenges

**2.7. Immunohistochemistry**

per slide.

Total RNA was extracted and purified from skin samples after homogenization in TRIzol (Invitrogen and Life Technologies, Carlsbad, CA), following the manufacturer's instructions. RNA was converted into cDNA using the SuperScript First-Strand Synthesis System (Invitrogen and Life Technologies, Carlsbad, CA). Random primers were used for the reverse transcription reaction and real-time quantitative polymerase chain reaction (PCR) was performed with either the CFX96 real-time PCR thermal cycler (Bio-Rad, Hercules, CA) or the ABI 7900 realtime PCR thermal cycler (Applied Biosystems, Foster City, CA). These samples were amplified in triplicate using primers for NFkB, IRAK1, TRAF6, IL-6, MIP-2 (IL-8), col1a2, col3a1, VEGF, BCL-2, HIF-1α, miR-146a, miR-15b, and miR-29a (TaqMan gene expression assay, Applied Biosystems, Foster City, CA). Internal normalization was achieved by using the 18 s housekeeping gene as an internal control for mRNA and the U6 housekeeping gene as an internal control for miRNA.
