**2.4. Model of burn injury**

Under the same anesthesia and dorsum preparation described above, the dorsum of mice was immersed in an 80°C bath for 10 seconds to produce a scald burn. Fluid resuscitation was achieved through a subcutaneous 1 mL saline injection, treated with different agents (last used EPO Z in comparison with EPO alpha), divided into three groups, and sacrificed on Days 3, 6, and 12, respectively. Burned skin has been divided into two flaps that have been, respectively, used for molecular essays and histology.

Molecular essays are consisted in cytoplasmic protein dosage (Bio‐Rad Protein Assay (Bio‐Rad Lab, Richmond, CA, USA), spectrophotometry, using albumin as a standard); Western blot for GFs and cell cycle molecules; histology measured the presence of inflammatory infiltrates, necrosis, and repair in standard hematoxylin (eosin, trichromic, and immunohistochemistry were used to visualize and quantify alpha‐smooth positive cells such as a response to VEGF in both neoangiogenesis and neovasculogenesis).

Statistical analysis was conducted with parametric essays for repeated measures (ANOVA) and bonferroni test was used to evaluate intergroup positivity, with a *p* ¼ 0.05 considered as statistically significant.

Our experimental studies have pointed out some important features of antioxidant molecules in impaired wound healing (diabetic mice), as well as the role of some cytokine‐related molecules and endogenous products belonging to natural immunity cascade [2] in normal and impaired wound healing [9–13].
