Blocking AGE-RAGE Signaling Improved Functional Disorders of Macrophages in Diabetic Wound

Advanced glycosylation end products (AGEs) accumulate in diabetic wounds. Interactions between AGEs and their receptor (RAGE) leads to dermatologic problems in diabetes. Macrophage, which plays important roles in wound healing, highly expresses RAGE. Therefore, we investigated whether RAGE-expressing macrophages might be responsible for impaired wound healing on diabetes. We used anti-RAGE antibody applied topically on diabetic wounds. After confirming that wound healing was improved in anti-RAGE antibody group compared with normal mice, our results showed that macrophages appeared insufficient in the early stage and fading away slowly in the later proliferative phase compared with the control group, which was ameliorated in anti-RAGE antibody-applied wounds. Blocking AGE-RAGE signaling also increased neutrophils phagocytized by macrophages and promoted the phenotypic switch of macrophages from proinflammatory to prohealing activities. In vitro, phagocytosis of THP-1 (M0) and lipopolysaccharide- (LPS-) induced (M1) macrophages was impaired by treatment with AGEs, while IL-4- and IL-13-induced (M2) macrophages was not. Finally, AGEs increased the proinflammatory response of M1 macrophages, while inhibiting the polarization and anti-inflammatory functions of M2 macrophages. In conclusion, inhibition of AGE-RAGE signaling improved functional disorders of macrophages in the early inflammatory phase, which promoted the healing of wounds in diabetic mice.


Introduction
Morbidity resulting from diabetes mellitus is rapidly increasing worldwide and constitutes a burden to our global society [1]. Impaired wound healing is a serious complication of this disease, and it results in severe pain and reduced quality of life. There is compelling evidence that AGEs accumulate in these wounds because of certain biochemical features associated with diabetes. They are thought to contribute significantly to the pathology associated with impaired wound healing [2][3][4].
RAGE is a receptor for AGEs, and it is expressed in a variety of cells. It is particularly enriched in macrophages. Recent clinical and experimental research has shown that blocking the AGE-RAGE signaling interaction enhances angiogenesis, increases granulation of tissues, and promotes faster reepithelialization in wounds. This helps to promote diabetic wound healing [5].
AGEs accumulate in the diabetic derma and contribute to impaired wound healing, together with macrophages, which express high levels of the receptor RAGE, suggests that AGE-RAGE signaling might underlie the macrophage dysfunction that is a hallmark of impaired wound healing in diabetics. Here, we studied the functional changes of macrophages during wound healing in a diabetic mouse model. We also explored the influence of AGEs on THP-1 macrophages and their relationship with AGE-RAGE signaling.
These results improve our understanding of the association between AGE-RAGE signaling and the functional dysregulation of macrophages in impaired wound healing. These findings may aid the development of macrophage-based therapies for this diabetic complication in the future.

Materials and Methods
2.1. Induction of Diabetes in Mice and Wounding. Male C57BL/6 mice (8-10 weeks old, 20-25 g) were obtained from the Experimental Animal Center of Rui Jin Hospital in Shanghai, China. All experimental procedures were in compliance with laboratory institutional guidelines and the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
To induce diabetes, a daily intraperitoneal injection of STZ (Sigma-Aldrich, St. Louis, MO, USA) at a dose of 65 mg/kg body weight was administered for 5 consecutive days. Blood glucose measurements were performed for 8 weeks after the injections. When polyuria, polydipsia, polyphagia, weight loss, and elevated blood glucose (16.7 mmol/L) were observed, mice were deemed to be in a diabetic state.
To introduce wounds in these animals, control and diabetic mice (n = 72 each) were anesthetized with an intraperitoneal injection of sodium pentobarbital (60 mg/kg body weight). The animals were shaved dorsally and a depilatory agent was used to remove the remaining hair. The surgical area was washed with benzalkonium bromide. One fullthickness excisional wound (9 mm diameter) was created by a sterilized punch. Throughout the experiment, all mice were individually caged and a semipermeable transparent dressing (Tegaderm; 3M Health Care, St. Paul, MN) covered the wound, which was replaced every 2 days until day 11. Diabetic and control mice were randomly assigned to three groups in which different topical treatments were applied to the wounds (the saline group (C), the rabbit IgG isotype (Bioss, Beijing, China) group (I), and the anti-RAGE antibody (Abcam, UK) group (R)). From days 0-10 after wounding, diabetic mice were treated with saline, mouse IgG isotype, or anti-RAGE antibody every 2 days. Normal control mice were topically treated with saline at the same time intervals. Mouse IgG isotype or anti-RAGE antibody (20 μg/dose each) was applied directly underneath the Tegaderm dressing. In the other group, the same volume of saline was applied as a control. A preliminary analysis suggested that topically use of anti-RAGE, rabbit IgG had no difference on wound closure rates on normal mice.

Analysis of Wound
Healing. On days 0, 1, 3, 7, and 14 after wounding, animals were anesthetized with an intraperitoneal injection of sodium pentobarbital. The edge of the wound was traced onto a transparent plastic membrane, which was then scanned for analysis using ImageJ 1.49v software.

Preparation of Wound
Tissue. The wound and surrounding tissues (a margin of approximately 5 mm into the unwounded skin) from animals in each group were excised on 1, 3, 7, and 14 days after wounding. The excised tissues included the subcutaneous fat underneath the wound. Samples were fixed in 10% neutral-buffered formalin and stored at −80°C for future analysis.

Hematoxylin and Eosin (H&E) Staining and
Immunohistochemistry. Wound tissue sections (4 μm thick) were then deparaffinized in xylene, rehydrated through a graded alcohol series into phosphate-buffered saline (PBS). To quantify neutrophils, sections were stained with H&E and mounted in resin. Six tissue sections from each group were randomly chosen and imaged with five fields per slice at 400x magnification using a Zeiss microscope (Axioskop 2 Plus, Germany).
For immunohistochemical analyses, sections were deparaffinized and pancreatin (1 : 3; Maxim, Fuzhou, China) for antigen retrieval was performed. Sections were incubated with rabbit polyclonal anti-AGE (1 : 10000; Abcam, UK) or mouse monoclonal anti-CD68 (1 : 400; Abcam, UK) at 4°C overnight. Following incubation with an HRP-labeled secondary antibody (Dako, Denmark), chromogenic development was performed using diaminobenzidine, and sections were then counterstained with hematoxylin. Stained cells at the wound edge were manually counted on Zeiss at 200x magnification. The images were captured using a Zeiss and processed by SPOT imaging software (Diagnostic Instruments, Sterling Heights, MI).

Transmission Electron Microscopy (TEM).
Tissues obtained on days 1 and 3 after wounding were fixed, dehydrated, and embedded in Araldite CY212. Ultrathin sections were stained with uranyl acetate and lead citrate and visualized by TEM (CM-120 BioTwio, Philips).

Cell Culture.
A human acute monocytic leukemia cell line (THP-1) was obtained from the Cell Bank of the Chinese Academy of Sciences. Cells were cultured at 37°C in a humidified incubator under 5% CO 2 in RPMI 1640 medium (Hyclone, USA) supplemented with 10% FBS. To induce differentiation of macrophages, THP-1 cells were cultured in the presence of 100 ng/mL phorbol-12-myristate-13 acetate (PMA; Sigma-Aldrich, St. Louis, MO, USA) for 24 h.

Transfection and Simulation.
Stealth RNA interference duplexes against RAGE were designed and synthesized (GenePharma Technologies, Shanghai, China). After differentiation, RAGE siRNA (50 nmol/L) was delivered into THP-1 macrophages using Lipofectamine® 3000 (Invitrogen, Basel, Switzerland) according to the manufacturer's instructions. To assess the efficiency of the transfection method, fluorescein isothiocyanate-labeled nonspecific siRNAs were used to show that up to 85% of the THP-1 macrophages were successfully transformed.
2.9. Phagocytosis Assay. Phagocytosis was examined by assessing the cellular uptake of 2.0 μm-sized FITC-latex beads (Sigma, US). Beads were incubated with cells at 37°C for 30 min. Cells were then washed with precooled PBS and analyzed via flow cytometry (FC, Beckman Coulter, US).

Statistical analysis
All data were represented as the means ± SD and analyzed with SPSS for Mac 21.0 (SPSS, Chicago, IL, USA). Analysis of variance (ANOVA) and Student's t-test were applied to determine the statistical significance of differences between groups. Statistical significance was defined as p < 0 05.

Confirmation of the Diabetic State and Expression
Localization of AGEs and RAGE. All mice receiving multiple injections of STZ displayed features specific to diabetes (polyuria, polydipsia, polyphagia, weight loss, and elevated blood glucose). Body weights in the STZ-injected group dropped from 21.9 ± 1.17 g to 18.1 ± 1.32 g (p < 0 05). During the 8 weeks following these injections, blood glucose levels also dramatically increased from 5.4 ± 0.7 mmol/L to 27.7 ± 4.1 mmol/L (p < 0 05).
Immunohistochemical staining for AGEs revealed faint signal in the dermal matrices and cells of control mice. In contrast, prominent signal was detected in the matrices, cells, and basement membranes of vessels of the skin in STZ-injected diabetic mice (Figure 1(a)).
To characterize the expression of RAGE on macrophages, we examined the normal and diabetic wound sections for the expression of RAGE/CD68 by immunofluorescence staining. Colocalization showed that macrophages from both normal and diabetic wound tissues expressed RAGE (Figure 1(b)).

Topical Application of Anti-RAGE Antibody Accelerated
Diabetic Wound Healing. To examine the effects of inhibiting AGE-RAGE signaling, we confirmed that topical application of an anti-RAGE antibody accelerated wound closure in diabetic wounds. A significant difference was observed at 7 d after wounding (Figure 1(c)).

4.3.
Blocking AGE-RAGE Interaction Improved the Removal of Neutrophils on Diabetic Mice. Neutrophils are the first blood-borne nucleated cells to infiltrate into injury or infection sites, where they produce multiple effector molecules [19]. As the inflammatory phase of diabetic wound healing exists, a persistence of overloaded reactive oxygen species leads to continuous damages. We examined the potential changes in neutrophil infiltration after inhibition of RAGE. We quantified the number of neutrophils present at 3 d postwounding, which is a key time point for neutrophil resolution. The amount of neutrophils significantly increased in the wounds of diabetic mice, and this was partially rescued by the application of the anti-RAGE antibody (Figure 2).

AGE-RAGE Signaling Delayed Macrophage Infiltration
and Neutrophil Being Phagocytized In Vivo. Macrophages phagocytize apoptotic neutrophils to prevent further tissue damage, thus resulting in resolution of wound inflammation and promoting wound healing [20,21]. After finding out that the infiltration of neutrophils was incresed in diabetic mice, which was modulated by AGE-RAGE signaling, we examined whether macrophage infiltration and phagocytosis were disrupted in diabetic mice. Macrophage infiltration was assessed by immunohistochemical staining for CD68 (Figure 3(a)). One day postwounding, group N displayed the presence of many CD68 + cells, while groups C and I had significantly fewer. Meanwhile, group R had a large quantity of CD68 + cells, but fewer than group N. At 14 d after wounding, groups N and R displayed significantly fewer CD68 + cells compared to groups C and I.
Immunohistochemical staining for CD68 + macrophages suggested that inhibition of the AGE-RAGE interaction partly reversed a phenomenon described previously as "macrophages arriving slowly and fading away slowly" [8] ( Figure 3(b)).
Samples collected on days 1 and 3 after wounding were examined for the presence of macrophages by TEM. At 1 d, groups N and R displayed macrophage infiltration at the wound edge, while groups C and I did not (Figure 4(a)). At 3 d, groups N and R showed the expected uptake of neutrophils by macrophages at the wound edge. In contrast, groups C and I possessed macrophages that lack any signs of phagocytosis (Figure 4(b)).

AGE-RAGE Signaling Contributes to Polarization
Disorder in Macrophages In Vivo. Because macrophages involved in diabetic wound healing display a switch disorder between proinflammatory and prohealing phenotypes [10,22], we examined whether blocking the AGE-RAGE interaction could reverse these defects. We tested the expression of various markers for macrophages (CD68), the proinflammatory phenotype (iNOS), and the prohealing phenotype (CD206) using immunofluorescence at 7 d postwounding, which is a critical time point during the proliferative phase (Figures 5(a) and 5(b)). Compared to group N, we observed that the number of iNOS + cells increased significantly, and this was rescued by treatment with the anti-RAGE antibody ( Figure 5(c)). Therefore, inhibiting AGE-RAGE signaling might reduce the switch disorder of macrophage polarization.  Figure 7). WB showed that AGE-RAGE signaling improved the TNF-α secretion of M1 macrophages, and these M1 macrophages also displayed the ability to secrete the growth factor PDGF. PDGF is typically decreased in the presence of AGEs, but to a lesser extent after blocking AGE-RAGE. However, the expression of the M1 macrophage marker iNOS was not affected among these groups (Figure 8(a)). With respect to M2 macrophages, AGE-RAGE signaling inhibited the expression of the M2 marker CD206 and PDGF secretion (Figure 8(b)). By FC analysis, we detected the expression of CD68, iNOS, and CD206 in M1 and M2 macrophages. More than 99% of the CD68 + cells expressed iNOS in all six experimental groups (Figure 8(c)). In addition, the expression of CD206 in M2 macrophages was downregulated after stimulation with AGEs, which was rescued by blocking the AGE-RAGE interaction (Figure 8(d)).

Discussion
Based on the results in vivo and in vitro, we found that AGE-RAGE signaling correlated with the disruption of macrophage function, including inhibition of phagocytosis and cytokine secretion. These processes play vital roles in the resolution of local inflammation and wound healing. Perturbing these functions may disrupt the healing process in diabetic wounds of human patients. Furthermore, we suggest that the early phase of the inflammatory process is an ideal control point for targeting macrophage-based therapies.
A disease-associated microenvironment state marked by AGEs accumulation in diabetes has been reported by multiple lines of evidences [2], which indicates that the AGE-RAGE interaction contributes to failed wound healing on diabetes. First of all, the accumulation of AGEs and the ability of the anti-RAGE antibody to block the AGE-RAGE interaction were confirmed in our STZ-induced diabetic mouse model. Oxidative stress is thought to occur during the early inflammatory phase of diabetic wounds, and these lesions are often described as being "stuck" in the inflammatory phase [21]. Consistent with this, we found that AGE-RAGE signaling devoted to insufficient macrophages infiltration and phagocytized neutrophils on diabetic mice. In addition to a deficiency in the numbers of macrophages, we also explored whether phagocytosis in macrophages normally devoted to clearing neutrophils was affected by AGEs. Using an in vitro phagocytosis assay, we found that AGEs impaired phagocytosis in THP-1 macrophages and LPSinduced M1 macrophages via AGE-RAGE signaling. Interestingly, IL-4-and IL-13-induced M2 macrophages displayed similar levels of phagocytosis as those observed in M1 macrophages, which were not affected by AGEs.
Ignoring the polarization phenotypes that we observed, these cells were clearly all of the macrophage identity, regardless of whether they were induced by particular cytokines. This explains why M2 macrophages also display similar phagocytic dynamics compared to M1 macrophages. We suggest that when M2 macrophages participate in wound healing, the secretion of various factors aid proliferation and regeneration. As for these macrophages' resistance to the effects of AGEs, this warrants further research. In vivo and in vitro experiments clearly showed that AGE-RAGE signaling impaired macrophage phagocytosis, which is important for shutting down inflammatory responses during wound healing.
Macrophages are known to switch from a proinflammatory phenotype to a wound-healing phenotype to resolve inflammation and initiate the healing process. However, macrophages in diabetic wounds undergo a switch disorder between inflammation and healing [23][24][25]. By immunofluorescence, we found that blocking the AGE-RAGE interaction improved defects in macrophage polarization, which were persistent in M1 macrophages. This also served to delay the appearance of M2 macrophages in vivo. To further establish that disruptions in macrophage function are related to AGE-RAGE signaling, polarization and cytokine secretion levels were also assessed in vitro. WB and ELISA results suggested that AGE-RAGE signaling improves the proinflammatory functions of M1 macrophages and impairs the antiinflammatory functions of M2 macrophages. Meanwhile, the expression of M2 macrophage markers suggests that AGE-RAGE interactions inhibit the polarization of these cells. Furthermore, the expression of an M1 macrophage  marker did not differ among the six experimental groups by WB. Meanwhile, FC analysis found that over 99% of the THP-1 macrophages expressed M1 macrophage markers after stimulation by LPS. That may explain why WB analysis detected no differences in expression levels among the test groups. Additionally, since there were a roughly equivalent number of M1 macrophages, the increase in expression of inflammatory factors by WB indicates that AGE-RAGE signaling improves the ability of cells to secrete these factors, which might be the consequence of impaired phagocytosis. Interestingly, WB and ELISA detected the secretion of the growth factor PDGF in M1 macrophages, while M2 macrophages secreted another inflammatory factor, TNFα. Although few in numbers, FC analysis identified some macrophages that coexpressed CD68, iNOS, and CD206. Other investigators have suggested that wound-specific macrophages possess hybrid M1/M2 activation phenotypes that make them highly plastic, conferring the ability to switch between inflammatory and wound healing functions [26][27][28]. All these observations indicate that there are limitations to the M1-M2 definition. All these facts proved that M1-M2 definition by stimuli encountered in vitro had its limitation to apply in vivo. What is more is that as the proliferation phase has been hindered by functional disorders of macrophages in the early inflammatory phase, it is unrealistic to figure out the proper time points and interventions. Thus, in complex and dynamic cellular microenvironments, the early inflammatory phase during wound healing is a more convenient and ideal control point for macrophagebased therapies compared to the proliferation phase.

Conclusions
Taken together, the current study demonstrates that blocking the AGE-RAGE interaction improves the function of macrophages, thus promoting diabetic wound healing. Unfortunately, we could not identify the specific functional dysregulations of macrophages in vivo, and this remains a pressing topic for future research. Regardless, our findings suggest that there is a close relationship among AGE-RAGE, wound healing, and macrophage physiology. Additionally, this study offers insights that could help develop therapies that target early-acting, wound-specific macrophages in patients suffering from diabetic lesions. Future work should focus on the mechanisms controlling functional dysregulation of macrophages during wound healing in vivo.