Use of Cerium Oxide Nanoparticles Conjugated with MicroRNA-146a to Correct the Diabetic Wound Healing Impairment

Diabetic mouse model

All experimental protocols were approved by the Institutional Animal Care and Use Committee at University of Colorado Denver - Anschutz Medical Campus and followed the guidelines described in the NIH Guide for the Care and Use of Laboratory Animals. For all experiments, 12-week-old female mice homozygous for the Leprdb mutation (db/db) were used (BKS.Cg-Dock7m+/+Leprdb/J, strain No. 000642, Jackson Laboratory).

Cerium oxide nanoparticles synthesis and characterization

The CNP nanoparticles were synthesized using simple wet chemistry methods, as described previously.²⁴ In a typical synthesis, a stoichiometric amount of cerium nitrate hexahydrate (99.999% from Sigma-Aldrich) was added to 50 mL of deionized water (18.2 MV) and stirred for 1 hour. The cerium (III) ions in the solution were oxidized to cerium (IV) oxide using excess hydrogen peroxide. After that, the pH of the solution was kept below 3.5 1 N nitric acid to maintain the synthesized ceria nanoparticles in suspension. Crystalline nanoparticles of cerium oxide form immediately on oxidation. Nanoparticles were isolated by centrifugation at 8,000 g for 10 minutes. Ten μM of nanoceria was prepared from 5 mM of nanoceria suspension by small aliquot dilution in pure water. Hydrodynamic size and zeta potential measurements were carried out at varying pH using dynamic light scattering measurements from Zeta Sizer Nano (Malvern Instruments), which uses a laser with wavelength at 633 nm. Different pH values in the suspension were adjusted using either ammonium hydroxide or nitric acid.

Murine wound healing model and treatment

Mice were anesthetized with inhaled isofluorane. Each mouse was shaved and depilated before wounding. The dorsal skin was swabbed with alcohol and Betadine (Purdue Pharma). Each mouse underwent a single dorsal full-thickness wound (including panniculus carnosum) with an 8-mm punch biopsy (Miltex Inc). These wounds were treated with 50 μL of PBS, 100 ng of CNP alone, 10⁶ pfu of lent-imiR146a, or differing concentrations of CNP-miR146a (1; 10; 100; or 1,000 ng) at the time of wounding (Fig. 1A). All wounds were dressed with Tegaderm (3M), which was subsequently removed on postoperative day 2. Postoperatively, the mice received a subcutaneous injection of an analgesic, buprenorphine (Schering-Plough Animal Health Corp).

Murine wound area measurement

Photographs were obtained with a Nikon camera (Nikon) using a ruler for each image. ImageJ software (National Institutes of Health; http://rsbweb.nih.gov/ij/) was used to calculate the wound area of each mouse every other day. A blinded observer analyzed the size of each wound. Wound area was plotted as a function of time.

Biomechanical testing of murine skin

Biomechanical testing was performed on a full-thickness skin area from the treated mice, including the wounded area. Biomechanical testing was done 28 days post-healing on wounded skin treated with PBS or CNP-miR146a (1; 10; 100; or 1,000 ng). Skin samples immediately underwent testing after harvest; all samples had their subcutaneous tissue removed, and a uniform, dumbbell-shaped testing unit was stamped out using well-established, previously described techniques.^9,11 Cranial-caudal orientation was preserved. Two Verhoeff stain lines were placed on either end of the dumbbell shape, demarcating the gauge length of the testing area. The mean cross-sectional area of the samples tested was calculated as precisely as possible to ensure that an accurate modulus and maximum stress were calculated. The ends of the testing unit were fixed between sandpaper using a cyanoacrylate adhesive to prevent slipping. The testing unit was mounted in custom designed fixtures and tested in tension using an Instron 5543 test frame (Instron). Each testing unit underwent a previously established protocol. In short, the testing unit was preloaded to 0.005 Newtons (N), held for 120 seconds, and then subjected to a constant increase in force until the sample failed (as evidenced by tearing of the skin) and was no longer able to hold tension. Maximum stress to failure was calculated using linear regression from the linear region of the stress-strain curve. The modulus of elasticity was calculated as the slope of the stress-strain curve within the elastic region of the ramp to failure.

Swine excisional wound model and treatment

A Yorkshire white male pig was used in this study. All procedures were approved by the Ohio State University Institutional Laboratory Animal Care and Use Committee (ILACUC). To induce diabetes, the pig was injected with streptozocin (STZ) (75 mg/kg body weight) only once. To avoid hypoglycemia due to insulin released by the destroyed beta cells, 200 mL of a 5% glucose solution were given over a period of 1 hour after STZ application. Four weeks after STZ injection, the pig (47 kg) was sedated by Telazol (Butler Schein) and anesthetized by mask with isoflurane (3% to 4%). The dorsal region was shaved. The skin was surgically prepared with alternating chlorhexidine 2% and alcohol 70% (Butler Schein) scrubs. Under such aseptic conditions, 2 sets of 5 full-thickness excisional wounds (1”×1”) were established on the back of pigs with a depth of approximate 2 cm, to reach the subcutaneous fat in all wounds. A total of 5 wounds were created on each side of the back. A scaled plastic template was used to create the wounds at fixed distance from each other and from the spine (4 cm), starting below the lower border of the scapula on the dorsum of the pig. Wounds from 1 side of the back were treated with a CNP-miR146a (100 ng) followed by dressing with Tegaderm, while the wounds from the contralateral side of the back were dressed with Tegaderm only and received only PBS (control). Digital photographs were taken on days 0, 3, 7, 10, and 14 for measurement of the wound area. The pig was maintained on 12-hour light–dark cycles and was euthanized 14 days post-wounding. All wounds were digitally photographed in the presence of a standard reference ruler. Wound area was calculated using the ImageJ software. The pig was maintained on standard chow ad libitum, fasted overnight before the procedures, and housed individually in the animal facility. The pig was maintained on 12-hour light–dark cycles and was euthanized after the completion of experiments.

Swine wounds histologic analysis

At day 14 post-wounding, the animal was euthanized and the wound-edge tissue (2 cm from the edge) was collected in formalin from the control (PBS) and treated (1,300 ng CNP-miR146a). Five-micron microtomed sections were placed on positive charged slides. Some slides underwent Masson’s Trichrome staining. The rest of the slides were deparaffinized. Once sections were hydrated they were placed in citrate buffer pH 6.0, and heat-induced epitope retrieval was performed using the Biocare Medical’s Decloaker. The slides were then stained using Leica’s Bond Rx instrument, and all reagents were applied robotically onto slides. The slides were incubated with primary CD45 and CD31 antibodies (BD Biosciences) at 1:50 dilutions. They were developed with a Vectastain Elite ABC kit with secondary antibody (Vector Laboratories). A blinded observer analyzed the numbers of CD31-positive and CD45-positive cells per high-power field.

Statistical analysis

The t-test was used to analyze the data. All data were expressed as the mean ± SEM and a value of p < 0.05 was considered significant.

Treatment of diabetic wounds with CNP-miR146a accelerates healing

Figure 1B shows the percentage of original wound area at day 7 post-wounding for nondiabetic wounds (PBS), diabetic wounds (PBS), and diabetic wounds treated with 100 ng CNP, 10⁶ PFU LentimiR-146a, or 100 ng of CNP-miR146a. The percentage of open wounds among diabetic wounds is significantly higher than in nondiabetic wounds at day 7 post-wounding. Treatment with 100 ng of CNP-miR146a significantly enhances diabetic wound healing. On the other hand, CNP or miR-146a alone did not improve healing. Treatment of diabetic wounds with the CNP-miR146a resulted in a significant improvement of wound healing. Diabetic wounds treated with PBS (0 ng) were significantly larger at days 14 and 16 compared to diabetic wounds treated with CNP-miR146a (Fig. 1C). Our data showed that diabetic wounds treated with PBS or 1 ng CNP-miR146a took 18 days to heal. Treatment with 10, 100, or 1,000 ng of CNP+miR-146a effectively enhanced the rate of wound closure, and wounds were fully closed at day 14 post-wounding (Fig. 1D).

Diabetic skin maintained its biomechanical properties post-healing

At 4 weeks after wounding, healed murine diabetic wounds treated with (1, 10, or 100 ng) CNP-miR146a showed a trend toward improved biomechanical properties compared with the PBS-treated wounds; however, it did not reach significance (Fig. 2A). The average modulus for the healed skin treated with PBS was 11.09 ± 5.98 MPa (Fig. 2B). For the healed skin treated with 1, 10, 100, and 1,000 ng of CNP-miR146a, the average modulus was 13.5 MPa, 13.71 MPa, 12.6, and 8.5 MPa, respectively (Fig. 2B).

Diabetic skin did not develop any pathologic abnormalities after treatment

Histopathologic microscopic examination of skin samples from mice treated with PBS or different doses of CNP-miR146a did not reveal any pathology or lesions (Fig. 3). In addition, no soft tissue abnormalities were noted in the majority of tissues; multiple soft tissues were microscopically reviewed, there was excess adipose tissue, but nothing appeared abnormal for this strain of diabetic mouse.

CNP-miR146a improves healing in a clinically relevant model of diabetes

To induce diabetes, the pig was injected with STZ (75mg/kg body weight) only once, as shown in Figure 4A. Wounds were created at 1 month post-STZ injection. Five of 10 wounds on the right side were treated with CNP-miR146a at day 0, as shown in Figure 4B. Digital imaging of the wounds was performed on days 0, 3, 7, 10, and 14 for measurement of wound area. Figure 4C shows representative digital photographs of 2 wounds at day 14, 1 treated with PBS and another with CNP-miR146a. These photographs show that at day 14 post-wounding, the wound treated with CNP-miR146a is clearly smaller in size compared with the wound treated with PBS.

Figure 4D shows graphic analysis of the areas of all 10 wounds, measured by digital planimetry. At day 10, the PBS-treated wounds had an average surface area of 6.8 cm² ± 0.46; the surface area of the CNP-miR146a treated wounds was 4.8 cm² ± 0.24. By day 14, the average surface areas of the PBS and CNP-miR146a treated wounds were 4.63 cm² ± 0.21 and 3.06 cm² ± 0.28, respectively (Data expressed as mean ± SEM, *p < 0.05).

CNP-miR146a decreases inflammation and increases angiogenesis in diabetic wounds

The effect of CNP-miR146a treatment on inflammation was assessed using immunohistochemistry for CD45, the common leukocyte antigen. Representative photos of immunoperoxidase staining for CD45 at 14 days in swine diabetic wounds treated with PBS or CNP-miR146a are demonstrated in Figure 5A. Quantitative analysis of diabetic wounds treated with CNP-miR146a demonstrated a significant decrease in CD45-positive cells compared with PBS-treated diabetic wounds (Fig. 5B).

The effect of CNP-miR146a treatment on angiogenesis was assessed using immunohistochemistry for endothelial marker CD31. Representative photos of immunoperoxidase staining for CD31 at 14 days in swine diabetic wounds treated with PBS or CNP-miR146a are shown in Figure 5A. Quantitative analysis of diabetic wounds treated with CNP-miR146a demonstrated a significant increase in the number of vessels compared with PBS-treated diabetic wounds (Fig. 5B).

The effect of the CNP-miR146a on fibrosis was assessed at 14 days using trichrome staining. Representative sections from both PBS- and CNP-miR146a-treated wounds are displayed in Figure 5. Treatment with CNP-miR146a did not result in a significant change in fibrosis compared with the PBS-treated diabetic wounds.