Increased ROS and Persistent Pro-Inflammatory Responses in a Diabetic Wound Healing Model (db/db): Implications for Delayed Wound Healing

Abstract

Diabetes and its complications, including impaired wound healing, present a critical clinical challenge and burden for the U.S. healthcare system, with costs of over USD 13 billion annually. Hyperglycemia and chronic inflammation in diabetic wounds increase reactive oxygen species (ROS) production, inducing oxidative stress and perpetuating inflammation, which delays healing. This study investigates inflammation, oxidative stress, and the roles of cellular populations in a diabetic wound healing mouse model (db/db). Given that diabetes leads to persistent inflammation and impaired fibroblast function, we also examined how diabetes influences superoxide production in dermal fibroblasts. Blood, dermal fibroblasts, and wound tissue were collected from 12-week-old female diabetic (Db) and heterozygous (Hz) mice. Electron paramagnetic resonance (EPR) spectroscopy revealed higher superoxide levels in diabetic blood, dermal fibroblasts, and wounds compared to controls. In diabetic wounds, immunohistochemistry and flow cytometry showed increased leukocyte infiltration and reduced macrophage presence, with a higher proportion of pro-inflammatory Ly6C^hi macrophages. These results suggest that elevated superoxide production and persistent inflammation contribute to impaired fibroblast function and delayed wound healing in diabetes. By identifying the contributions of ROS and Ly6C^hi macrophages to oxidative stress and chronic inflammation, this study offers insights into therapeutic strategies. These findings highlight the importance of addressing systemic oxidative stress alongside localized inflammation to improve wound healing outcomes in diabetic patients and advance diabetic wound care strategies.

Keywords: EPR; chronic inflammation; db/db mouse model; dermal fibroblast; diabetes; reactive oxygen species (ROS); wound healing.

Figure 1

Increased superoxide levels in the blood and dermal fibroblasts of diabetic mice compared to heterozygous control mice. Whole blood was collected via cardiac puncture and fibroblasts were isolated from the skin of 12-week-old female control heterozygous (Hz) and diabetic (Db) mice. A total of 300 µL of the blood was treated with a CMH probe (0.25 mM) for 10 min at room temperature. Cultured fibroblasts were treated with a CMH probe (0.25 µM) and incubated for 50 min at 5% CO₂ and 37 °C. A total of 50 µL of blood or cell suspension was loaded in a capillary tube and EPR measurements were conducted immediately. EPR measurements were conducted at room temperature and spectra were recorded using an X-band spectrometer EMXnano (Bruker). The amount of superoxide was determined based on the 1:1 reaction of CMH with superoxide to form CM^● and quantified by the SpinFit and SpinCount modules (Bruker). (A) Concentration of CM^● in blood. (B) Concentration of CM^● in fibroblasts normalized to protein concentration. Data were analyzed with Prism software version 10.4.2 using an unpaired t-test and are expressed as mean ± SEM; * p < 0.05, ** p < 0.01 (n = 6–7).

Figure 2

Increased superoxide levels in wounds from diabetic mice compared to heterozygous control mice. Wound tissue superoxide was measured 7 days after wounding. Skin from the wounds of heterozygous (Hz) and diabetic (Db) mice was homogenized in sucrose buffer, with a 1:6 ratio of tissue to buffer. The skin homogenate was treated with a CMH probe (0.25 mM) and incubated for 20 min at room temperature; then, 150 µL of the mixture was loaded into PTFE tubing and flash-frozen in liquid nitrogen. EPR measurements were conducted at 77 K and spectra were recorded using the X-band spectrometer EMXnano (Bruker). (A) Representative EPR spectra of the nitroxide CM^● in wounds from Db (red trace) and Hz (black trace). (B) CM^● concentration was obtained by double integration followed by SpinCount. Data were analyzed with Prism software version 10.4.2 using an unpaired t-test and are expressed as mean ± SEM; * p < 0.05 (n = 9–13).

Figure 3

Increased total leukocyte and neutrophil infiltration but decreased macrophage infiltration in diabetic wounds compared to heterozygous controls. Diabetic and heterozygous wounds were assessed for inflammatory cell infiltration by immunohistochemistry, specifically staining for (A) CD45 (leukocytes), (B) myeloperoxidase (MPO, neutrophils), and (C) F480 (macrophages). High-power fields (HPFs, 200× magnification). Data were analyzed with Prism software version 10.4.2 using an unpaired t-test and are expressed as mean ± SEM; * p < 0.05, ** p < 0.01 (n = 4–5).

Figure 4

Fewer macrophages and a higher proportion of Ly6C^hi macrophages in diabetic wounds compared to heterozygous controls. The myeloid cell phenotype was assessed by flow cytometry in diabetic and heterozygous wounds: (A) neutrophil (CD64− Ly6G+), (B) monocyte/macrophage (CD64+ Ly6G-), and (C) monocyte/macrophage Ly6C^hi populations. Data were analyzed with Prism software version 10.4.2 using an unpaired t-test and are expressed as mean ± SEM; * p < 0.05, ** p < 0.01 (n = 6–7).