The experiments were performed in adherence with the Guide for the Care and Use of Laboratory Animals and were approved by the Ethics Committee of the University of Lübeck. Twenty female Balb/c mice, (6–8 weeks old, bodyweight 20–25 g, Takomi, Copenhagen, Denmark) were included in these experiments. The animals were kept separately in cages under controlled environmental conditions and had free access to standard laboratory rodent feed and water ad libitum. After thermal injury, the mice were randomly divided into five experimental groups of four animals each. In Groups 1 and 3, mice with dermal wounds received a daily subcutaneous 200 μL injection of 400 IU nanosize rhEPO/kg (NeoRecormon®, Roche Ltd, Mannheim, Germany). In Groups 2 and 4, control animals with dermal wounds received equivalent volumes of vehicle only (distilled water). The treatment duration was 7 days for Groups 1 and 2 and 14 days for groups 3 and 4. In addition, a fifth group of mice without dermal wounds served as controls. EPO administration was carried out according to other studies in mice.15
Scalding injury model
The animals were anesthetized with a mixture of ketamine hydrochloride 10 mg/kg (Bela-Pharm, Vechta, Germany) and xylazine 2.4 mg/kg (Bayer Health Care, Leverkusen, Germany) via intraperitoneal injection; these are suitable anesthetic agents for inducing skin-deep second-degree scalding injuries in a mouse model. After general anesthesia, the hair on the back was shaved and the skin washed and rinsed with sterile water. Creation of a second-degree scald injury was achieved after the area was marked with ink by a 2 × 3 cm window on the back of the mice and immersed in 80°C water for 6 seconds. Mice received 1 ml/kg saline Ringer’s solution immediately after scalding. The scald injury was covered with moist saline dressings to prevent animals from grooming. After thermal injury, the animals were randomized into the five treatment groups and received subcutaneous application of nanosized rhEPO or its vehicle in a blinded fashion. Post- scald sedation and analgesia was provided for the first 7 days with metamizole (1000 mg/L added to the drinking water). At day 7 or 14, respectively, the mice were sacrificed with a barbiturate overdose. The wound tissue or regenerated tissue was obtained for further gene expression, Western blot analyses, and extensive histological studies. The skin was excised from the scalded areas at the back of each mouse, divided into two segments, separately embedded, and frozen in liquid nitrogen as described below.
All evaluations were performed on routine hematoxylin and eosin stainings because these enable a qualitative description of distinct features of the wound areas and the tissue reaction upon the scald trauma, and morphometric evaluation of distinct parameters, such as presence or absence of re-epithelialization, width of epidermal layer, epidermal stratification, formation of rete ridges, presence of skin appendages (such as hair follicles), and presence of regeneration zones. All criteria used for histological scoring of wound healing are summarized in .
Histological scoring of wound regeneration
Gene expression analysis
After preparation of the skin samples, the RNA was obtained by precipitation with isopropanol. After centrifugation, the pellet was washed with 75% ethanol, dried, and diluted in RNase-free water.
RNA was purified and reverses transcribed using standard protocols (Qiagen, Promega, Madison, WI). Target gene products like EPOR (ENSMUST00000006397), βCR (ENS-MUST00000096356), cytokeratin 14 (CK14, ENSMUST00000007272), and glyceraldehyde-3-phosphate-dehydrogenase (GAPDH, ENSMUST00000073605, the housekeeping gene) were amplified by reverse transcriptase polymerase chain reaction using the following intron spanning primers: EPOR forward 5′GCGTTTCTGGTGTTCACTGC-3′ and reverse 5′-GGATGTGGGTGGTCATAGG-3′ (product length 334 base pairs, 40 cycles), βCR forward 5′- -3′ and reverse 5′- -3′ (product length 226 base pairs, 40 cycles), CK14 forward 5′- -3′ and reverse 5′- -3′ (product length 306 base pairs, 30 cycles), Col1 forward 5′- -3′ and reverse 5′- -3′ (product length 244 base pairs, 40 cycles), and GAPDH forward 5′- -3′ and reverse 5′- -3′ (product length 219 base pairs, 30 cycles).
Western blot analysis
Total protein and RNA extraction from the skin samples was performed using the peqGOLD Trifast method according to the manufacturer’s protocol (peqlab, Erlangen, Germany). After skin sample preparation and precipitation of protein, the resulting pellet was dissolved in 1% sodium dodecyl sulfate solution. The concentration of soluble proteins was determined using the bicinchoninic acid assay (Pierce, Rockford, IL). We prepared the gel buffer for electrophoresis which comprises 120 mL and 200 mL of 1× sodium dodecyl sulfate buffer (prepared from 10× sodium dodecyl sulfate buffer, whereby 1 L contained 30 g Tris, 144 g glycine, 15 g sodium dodecyl sulfate) and filled the inner and lower chamber of electrophoresis apparatus. We checked the assembled electrophoresis apparatus with water for leakage. If no leakage is detected in 10 minutes, we removed the water. We inserted a filter paper to remove the remaining water from the electrophoresis apparatus. The wells were washed with sodium dodecyl sulfate buffer. The protein sample (up to 10 μg) solved in 1× sample buffer (2× sample buffer, ie, 2 mL 0.5 M Tris/HCl pH 6.8, 3.6 mL 86% glycerin, 0.31 g sodium dodecyl sulfate, 0.124 g DTT, a pinch of Bromophenol blue, adjusted to 10 mL with distilled water) was heated at 95°C for 5 minutes. Protein samples (20 μg) were analyzed by sodium dodecyl sulfate polyacrylamide gel electrophoresis in a 10% resolving gel at 120 V for one hour using the Thermo Scientific EC Classic Series Submarine gel system (Thermo Fisher Scientific, Schwerte, Germany) and visualized by the Coomassie brilliant blue staining method and then transferred by blotting to a 0.45 μm nitrocellulose membrane (Cell Signaling Technology, Danvers, MA) in cold transfer buffer (48 mM Tris pH 9, 39 mM glycine, 0,037% sodium dodecyl sulfate, 20% methanol). The primary antibodies against EPOR (sc-697), βCR, (sc-678), and growth hormone receptor (sc-20747) were from Santa Cruz Biotechnology (Santa Cruz, CA), against β-actin from Sigma (clone AC-74) and secondary horseradish peroxidase-conjugated antibodies (goat antirabbit immunoglobulin G) were from Dianova (Jackson) ImmunoResearch. The housekeeping β-actin signal, detected with secondary horseradish peroxidase-conjugated antibodies (goat antimouse immunoglobulin G) from Dianova, was used as the control for protein loading. The membranes were blocked with 5% nonfat dry milk in Tris-buffered saline 0.1% Tween-20 and 0.02% NaN3, for one hour at room temperature. The immunoblots were then incubated overnight at 4°C with the primary antibody and for one hour with the secondary antibody, analyzed using an enhanced chemiluminescence system (Cell Signaling Technology), and exposed on a Molecular Imager ChemiDoc XRS system (Bio-Rad, Hercules, CA). Appropriate mouse skin tissue specimens from the scalding experiments were freshly removed, immediately embedded in TissueTec (Jung, Nussloch, Germany), and frozen in liquid nitrogen. Frozen tissue sections of 10–14 μm thickness were cut on a cryotome (Leica CM 3050 S, Nussloch, Germany) and stained with hematoxylin (Mayer’s modification, Roth, Karlsruhe, Germany) and eosin using standard protocols.
The stainings were analyzed by light microscopy and images were taken on an Olympus IX 51 microscope (Olympus, Tokyo, Japan) using a CC12 camera and imaging software of the Olympus cell* family (cell* Imaging Software for Life Sciences Microscopy, Soft Imaging System GMbH, Münster, Germany). The mean values of all sections were allocated to the relevant score values for all individual specimens and all relevant groups, and used for statistical analysis. The scoring data served as a reference for comparison of both experimental groups. The histological scoring system ranged between 0 and 4 for wound closure and 0 and 2 for each of the three partial scores of epithelial covering (). The score values for epithelial covering were summarized (up to 6) and are represented graphically (). The specimens were evaluated in parallel by two independent researchers, including one pathologist, in a blinded manner ().
Figure 2 Histological scoring of the results of scald regeneration in mice treated with rhEPO or vehicle at different healing times (7 and 14 days). The final score of epithelial covering for each animal specimen is demonstrated as the sum of the three partial (more ...)
Figure 3 Representative hematoxylin and eosin stainings of scalded skin tissue samples from mice on day 7 after injury and continuous application of rhEPO or vehicle by subcutaneous injection. (A) Control group without EPO (vehicle only), mouse 1 (a–f); (more ...)
Comparable sections were evaluated every 100–200 μm depth at low-power (40×) and high-power magnification (100×). All tissue sections for each specimen (≥20) were systematically examined using a modified histological scoring system according to data in the literature.3
The histological parameters were examined and scored as wound closure in the form of assessment of the remaining wound area and re-epithelialization (presence or absence of epithelial covering, degree of epithelial covering relative to the number of epithelial cell layers as a monolayer or multilayer, and presence of conical structures).
In detail, the semiquantitative score for the epithelial covering with the described three parameters was made by systematic analysis. This was done on microscopic visual fields of whole sections with 40× or 100× magnification of each specimen (). The data for every section of the specimen were summarized, and the mean value was calculated and allocated to the relevant partial score value. The sum of the three partial score values was used as the final score value for each animal specimen, and was used for statistical analysis of the entire experimental group. We defined the quantitative score for wound closure to be the main criterion of wound healing, defined as the percentage of the remaining wound area when compared with the total skin area in the microscopic visual field of the section by 40× magnification. These analyses were performed on three random visual fields of each section.
Figure 1 Photographic example of the histological scoring method. The histological parameters of epidermal regeneration of mouse skin scored were (A) the presence or absence of epithelial covering, (B) presence of conical epithelial covering, and (C) degree of (more ...)
The data for the histological scores were analyzed using the unpaired Student’s t-test to compare the different experimental groups. The results are expressed as the mean ± standard error of the mean. Statistical significance was set at P < 0.05.