Mouse BMSCs were cultured using the tissue-explants adherent method (12). BALB/c mice at 8 weeks of age were anesthetized with 10% chloral hydrate (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), and cuts were made in the femur and tibia. The bone marrow was flushed out from the femur and tibia punctures with Dulbecco's Modified Eagle's medium (DMEM; Promega Corporation, Madison, WI, USA), and centrifuged at 37°C to separate the cells (2,250 × g, 5 min). The cells were re-suspended in DMEM containing 10% FBS (Sigma-Aldrich; Merck KGaA), and maintained in a saturated humidified incubator at 37°C, in a 5% CO₂ atmosphere, to observe their growth. When the cells reached 80% confluence, they were digested, centrifuged at 37°C and 2,250 × g, and subcultured in DMEM without serum. To identify BMSCs, flow cytometry was used to detect MSC-specific markers: CD29, CD44, Sca-1, CD45 and CD34. When the BMSCs at the third passage grew to 80% confluence, the cells were harvested with trypsin digest solution. Following washing the cells with 4°C PBS three times, the cell density was adjusted to 1.0×10⁵/ml with 4°C PBS containing 0.2% BSA (Promega Corporation) and fixed in 4% formalin (Sigma-Aldrich; Merck KGaA). Then the cells were centrifuged at 350 × g for 5 min. Prior to antibody treatment, the cells were blocked with goat serum at room temperature for 10 min (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA). Following this, the supernatant was discarded, and the remaining cells were stained with primary monoclonal antibodies against mice CD29 (catalog no. 46-0291-82), CD44 (catalog no. 46-3476-87), Sca-1 (catalog no. 47-0544-36), CD45 (catalog no. 46-2217-44; 1:25; Invitrogen; Thermo Fisher Scientific, Inc.) and secondary CD34 (catalog no. 46-5562-33; 1:50; Invitrogen; Thermo Fisher Scientific, Inc.) at room temperature for 2 h, and then conjugated with fluorescein isothiocyanate, respectively. The cells were visualized with propodium (PI) staining. Finally, the expressions of these molecules on cells were detected by flow cytometry (FACScan; BD Biosciences, Franklin Lakes, NJ, USA).

A total of 24 BALB/c mice, with a male/female ratio of 1:2, weighing 20 to 25 g, (provided by the Xinxiang Medical University Laboratory Animal Center, Xinxiang, China) were accustomed to a 24 h light/dark cycle and had free access to food and water prior to fasting for 24 h, and then received a peritoneal injection of 70 mg/kg body weight streptozotocin (Sigma-Aldrich; Merck KGaA). The weight and blood glucose level changes were monitored every day, and mice with a blood glucose level of >16.7 mmol/l were selected as the hyperglycemia mice group. The mice were sacrificed following anesthesia with diethyl ether. The study was approved by the ethics committee of Xinxiang Medical University (Xinxiang, China).

The dorsi of the mice were shaved, and the mice were anesthetized via intramuscular injection of 10% chloral hydrate, followed by disinfection with 70% alcohol. Circular full-thickness sections of the skin, 1 cm in diameter, were excised from both sides of the back, and the wounds were stanched with sterile gauze.

The mice of the full-thickness skin wound model were randomly divided into three groups, with eight mice per group: A blank control group (no treatment); hyperglycemic mice with a full-thickness skin wound and no further treatment; and hyperglycemic mice with a full-thickness skin wound treated with BMSCs at a dose of 6×10⁷ cells/kg body weight.

Photographs of the wounds of the three groups of mice were captured with a digital camera at 3, 6, 9 and 12 days following treatment, and analyzed using Image-Pro Plus software (version, 6.0; Media Cybernetics, Inc., Rockville, MD, USA). The edge of the wound was marked and the wound area was measured. Then, the rate of wound healing was calculated three times, as follows: wound healing rate = [(initial wound area - wound area) / initial wound area]×100%.

A total of 15 days following treatment, new skin tissue samples were collected from the wounds of mice, fixed in 4% formaldehyde, paraffin-embedded, sliced and stained with hematoxylin and eosin for observation of the skin organization structure under inverted microscope.

Skin tissue samples around the wound were collected from each group before and at 1–6 days once daily following treatment, weighed and crushed with liquid nitrogen. Radioimmunoprecipitation assay buffer (Sigma-Aldrich; Merck KGaA) was added to the crushed samples to obtain the lysate, which was centrifuged (3,913 × g, 10 min) to collect the supernatant for determination of IL-6 and TNF-α protein concentrations with ELISA kits according to the manufacturer's protocol (catalog no. KHO0411; Elabscience Biotechnology Co., Ltd., Wuhan, China).

All statistical analyses were performed using SPSS software version 19 (IBM SPSS, Armonk, NY, USA). Data are expressed as the mean ± standard error of the mean. The differences between groups were tested using Student's t-test or one-way analysis of variance, followed by Fisher's least significant difference test. P<0.05 was considered to indicate a statistically significant difference.

Newly generated cells could be seen after 5–7 days. The cells grew, adhering to the wall, and were shuttle-shaped or polygonal; their distribution was mostly non-uniform (Fig. 1). Examination by flow cytometry demonstrated that the phenotype of the cells conformed with the typical phenotypic characteristics of BMSCs: CD29(+), CD44(+), Sca-1(+), CD45(−) and CD34(−) (Fig. 2). Specifically, CD29, CD44 and Sca-1 expression constituted >95% of antigen expression, and expression of CD45 and CD34 expression accounted for <2%.

Fig. 3 presented the wound healing rates of the three groups. Obvious scar tissue appeared around the wounds of the blank control group at 3 days following treatment, whereas the hyperglycemia group and hyperglycemia + stem cells group presented no scar tissue. At 6 days following treatment, obvious scar tissue was still visible in the blank control group, however, there were clear signs of healing in the blank group and hyperglycemia + stem cells group. At 9 days, the wound healing percentages of the blank control group, and the hyperglycemia + stem cells groups, demonstrated significant differences, when compared with the hyperglycemia group. At 12 days, the wounds of the blank control group were almost completely healed. In addition, the hyperglycemic mice treated with stem cells demonstrated good wound healing; yet the hyperglycemia group still had a large area that had not healed.

A total of 15 days following treatment, hematoxylin and eosin staining was applied to the new skin tissue of the wounds of the three groups, which were observed with an inverted microscope. In the blank control group, the basal layer of new skin tissue was uniform, the horny and granular layers were thick, demonstrating good fusion with the wound edge (Fig. 4A). In the hyperglycemia group, the new skin wound tissue basal layer was flat, and the prickle cell layer, granular layer, and hierarchical structure of the horny layer were visible, but the horny layer and granular layer were thinner and incomplete, presenting poor fusion with the wound edge (Fig. 4B). In the hyperglycemic mice treated with stem cells, the new skin tissue basal layer was uniform, the horny layer and granular layer were thinner, and fusion with the wound edge was better (Fig. 4C).

Fig. 5A demonstrated the concentration of IL-6 in the skin tissue homogenate of mice. Prior to trauma, the IL-6 concentration in the skin tissue of normal mice and hyperglycemic mice were very low (<5 ng/g), which dramatically increased in all three groups following trauma. Over time, the concentration of IL-6 in the blank control group gradually reduced to a visibly consistent level at 3 days following trauma. In the hyperglycemic group, the concentration of IL-6 decreased slightly from day 1 to day 6 within the scope of 45–28 ng/g. In the hyperglycemic mice treated with stem cells, the concentration of IL-6 decreased which was slightly higher than that of the blank group (P=0.521).

Fig. 5B presents the concentrations of TNF-α in the skin tissue homogenates of the mice. Prior to trauma, the concentrations of TNF-α in the skin tissue of normal mice and hyperglycemic mice were 10–20 ng/g, and demonstrated a dramatic increase in all three groups following trauma. As time increased, the concentration of TNF-α in the blank group gradually reduced to approximating the normal level at 3 days following trauma. In the hyperglycemic group, the concentration of TNF-α remained higher than 30 ng/g throughout the study period. In the hyperglycemic mice treated with stem cells, the concentration of TNF-α significantly decreased, which was slightly higher than that of the blank group.