Oxytocin effects on experimental skin wound healing.

OBJECTIVE: Oxytocin (OXY) has significant effects on mammalian behavior. Next to its role in lactation and social interactions, it is described to support better wound healing as well. However, direct OXY effects on wound healing and the regeneration of the microvascular network are still not clarified. We therefore examined the effects of OXY and an OXY receptor antagonist [atosiban (ATO)] on skin wound healing, focusing on epithelialization and neovascularization.
METHODS: Skin wound healing has been assessed using intravital fluorescence microscopy in a model of full dermal thickness wounds in the dorsal skin fold chamber of hairless mice. Animals received repetitive low or high doses of OXY or ATO. Morphological and cellular characterization of skin tissue repair was performed by histology and in vitro cell assays.
RESULTS: The assessment of skin tissue repair using this therapy regimen showed that OXY and ATO had no major influence on epithelialization, neovascularization, wound cellularity, or inflammation. Moreover, OXY and ATO did neither stimulate nor deteriorate keratinocyte or fibroblast migration and proliferation.
CONCLUSION: In summary, this study is the first to demonstrate that OXY application does not impair skin wound healing or cell behavior. However, until now, the used transmitter system seems not to be clarified in detail, and it might be proposed that it is associated with the stress response of the organism to various stimuli. ©2017 Sorg H., et al., published by De Gruyter, Berlin/Boston.

Abbreviations: ATO, atosiban; bFGF, basic fibroblast growth factor; CAE, chloroacetate esterase; FMD, functional microvascular density; HD, high dose; LD, low dose; OXY, oxytoxin; RBCV, red blood cell velocity.

Introduction

Skin wound healing is a well-orchestrated survival mechanism that can be influenced by various circumstances leading to a better or worse course of healing [1], [2]. Recently, there is also mounting evidence that psychological effects (i.e. stress, social support, positive affect, and environmental enrichment) might also interact with wound healing [3]. The hormone oxytocin (OXY) is best known for its role in lactation, parturition, and uterine contraction. However, much research work has been done on its peripheral psychosocial roles and effects not directly related to gestation or reproduction [4], [5], [6]. OXY is a nonapeptide that is synthesized in neurosecretory cells of the supraoptic and paraventricular nuclei of the hypothalamus. From there, it is transported via axons to the neuronal part of the pituitary gland and then stored and released, by touch and warmth, as needed [7]. OXY has significant effects on mammalian behavior, especially for emotional as well as social bonding in relationships and sexuality [5], [7], [8], [9], [10], [11]. Additionally, different animal models proved that OXY supported better social interactions (i.e. less stress), which were associated with overall better healing performances [12], [13]. If applied repetitively, the effects on other transmitter systems could demonstrate long-lasting effects that show interaction in both growth and healing actions [7], [12], [13], [14]. Herein, it could be demonstrated that OXY application in rats leads to an acute anti-inflammatory response [15]. In different in vitro assays, it could also be seen that it acts as an antioxidant substance [16] and stimulates the proliferation of osteoblasts, pituicytes, blastocysts, myofibroblasts, and dermal as well as malignant endothelial cells [17], [18], [19], [20]. However, its direct effects on physiological skin wound healing and the regeneration of the microvascular network in the skin are still not clarified in detail. Therefore, we examined the effects of OXY and atosiban (ATO), an OXY receptor antagonist, during skin wound healing in an in vivo animal model of full dermal thickness wounds. Skin wound healing has been further characterized by histology and immunohistochemistry as well as by different in vitro cell assays.

Materials and methods

Animals and wounding

Male homozygous SKH-1-h hairless mice (12–16 weeks old) with a body weight (bw) of 30–40 g were used in the study. The animals were housed in standard laboratories with a 12 h light-dark cycle and had free ad libitum access to standard laboratory food and water. The Animal Ethics Committee of the State Provincial Office of Mecklenburg-West Pomerania approved the animal experiments (permit no. LALLF M-V/TSD/7221.3-1.1-001/08) and the experiments were carried out in accordance with the permits of the Laboratory Animal Centre of the University Medicine Rostock practices and the Institutional Animal Care and Use Committee guidelines. For intravital fluorescence microscopy of skin wound healing, the dorsal skin fold chamber preparation in mice was used as described previously in detail by our group [21]. In brief, mice were anesthetized intraperitoneally with a mixture of ketamine (90 mg/kg bw; 10% ketamine; Bela-Pharm, Vechta, Germany) and xylazine (25 mg/kg bw; 2% Rompun; Bayer Health Care, Leverkusen, Germany). Two symmetrical titanium frames were implanted to sandwich the extended double layer of the skin. The creation of a full dermal thickness wound was achieved after marking the area with a standardized circular ink stamp (2.5 mm in diameter) and by removing the complete skin down to the panniculus carnosus, thus creating a wound area of 3–6 mm2. The nonwounded skin of the opposite side still consisted of epidermis, dermis, and striated skin muscle. The wounded site was covered with a removable glass coverslip incorporated in one of the titanium frames. This model allows to visualize the process of revascularization, angiogenesis, and vessel regression in the skin by means of intravital fluorescence microscopy. Additionally, this technique allows the repeated study of the continuing process of skin repair over a period of 2–4 weeks. By implantation of the two titanium frames, we could avoid wound contraction by positioning the skin in between frames and by this allowing microscopy on a plane tissue level [21].

Experimental groups and protocol

A total of 36 animals with dermal wounds were included into the study and randomly allocated into five experimental groups. Animals received a daily intraperitoneal injection of either low-dose OXY (LD OXY; 1 mg/kg bw; n=9; Sigma-Aldrich, Steinheim, Germany) or high-dose OXY (HD OXY; 10 mg/kg bw; n=6; Sigma-Aldrich). An additional set of animals were treated with a low dose of an OXY receptor inhibitor ATO (LD ATO; Tractocile, 1 mg/kg bw; n=7; Ferring GmbH, Kiel, Germany) or a high dose (HD ATO; 10 mg/kg bw; n=6; Ferring) to evaluate the effect of an OXY receptor antagonist on wound healing. Control animals received equivalent volumes of physiological saline (0.9% NaCl; 12.5 mL/kg; n=8). Animals were studied by intravital fluorescence microscopy on days 3, 6, 9, and 12 after wounding. At the end of the experiments, blood and wound tissue was collected for subsequent laboratory analysis.

Microscopic analysis of wound repair and microcirculation

All procedures were performed in ketamine/xylazine-anesthetized mice as described above. The analysis of wound epithelialization was performed under a stereomicroscope using planimetric techniques (IC-A; Leica Microsystems GmbH, Wetzlar, Germany). Wound closure was considered complete when the entire surface area was covered with tissue. The analysis of angiogenesis and microcirculation was performed with the use of an intravital fluorescence epi-illumination microscope (Axiotech vario; Zeiss, Jena, Germany). Contrast enhancement for microvessel imaging was achieved after retrobulbar injection of 0.1 mL of 2% fluorescein isothiocyanate (FITC)-labeled dextran (molecular weight 150 kDa; Sigma Chemical, Deisenhofen, Germany) [21] and allowed the analysis of microvessel diameter, red blood cell velocity (RBCV), and functional microvessel density (FMD; Figure 1). Quantitative offline analysis of the videotaped images was performed by means of a computer-assisted image analysis system (CapImage; Dr Zeintl Software, Heidelberg, Germany). For the analysis of the different microvascular parameters, two to four consecutive areas have been used with clear visibility, which could be reached by focusing through the wound considering the three-dimensional aspect of the wound. In the respective areas, two to four different vessels, which belonged to the chosen area for the FMD measurements, have been used for the analysis of the diameter and the RBCV. The presented data show the mean values of the analyzed areas. RBCV was determined using the line-shift diagram method (modified “frame-to-frame” measurement, CapImage [22]), including the Baker-Wayland factor (1.6) for consideration of the parabolic profile of blood in microvessels [23]. FMD was defined as the total length of RBC-perfused microvessels per observation area and was given in cm/cm2. The three-dimensional aspect of the wound was considered by focusing through the depth of the wound.

Figure 1:

Representative intravital fluorescence microscopic images of microcirculation on day 9 of wound healing in an HD OXY-treated animal (A–C).

The process of wound healing shows distinct patterns of newly formed microvascular networks (A). At first, the regeneration of the microvascular network creates an inner circular ring of vessels at the direct wound margin (A; higher magnification in B). The inner ring is surrounded by outer radially localized vessels, supplying the vasculature within the inner circular ring (A; higher magnification in C). Dashed line (A) marks the edge of the wound at the day of wounding; continuous line indicates border between radial and circular microvessels. Scale bar, 200 μm (A) and 80 μm (B and C).

Histology and immunohistochemistry

The titanium frames were explanted on day 12 and the sandwiched skin was fixed in 4% phosphate-buffered formalin for 3 days and then embedded in paraffin. From the paraffin-embedded tissue blocks, 4 μm sections were serially cut and stained with hematoxylin-eosin (H&E) for the assessment of routine histology and for wound cellularity as a parameter for the resolution of the granulation and inflammation tissue. For this purpose, digitized images were taken in grayscale format and analyzed by the CapImage software. A random sample of a wound image was taken, and the gray level necessary to mark the complete area of the nuclei of the cells was determined. This gray value served as threshold value for the analysis of all further images. For each wound, an identical area of approximately 1.0 mm2 was analyzed. Values for cellularity are given as area in mm2.

Leukocytes were stained with the AS-D chloroacetate esterase (CAE) technique and identified by positive staining and morphology within the granulation tissue. For CD31 staining, goat polyclonal anti-CD31 (1:50; Santa Cruz Biotechnology, Heidelberg, Germany) was used as primary antibody at 4°C overnight followed by exposure to a secondary antibody at room temperature (LSAB Kit; Dako Deutschland GmbH, Hamburg, Germany) according to the manufacturer’s instructions. New fuchsin (Dako Deutschland) was used as chromogen. All sections were counterstained with hemalaun and examined by light microscopy (Axioskop 40; Zeiss). The numbers of CAE-positive cells were counted in five to six directly neighboring high-power fields (HPF) within the wound granulation tissue and are given as n/HPF. The wound granulation tissue as well as the initial wound margins could be exactly differentiated by a clear demarcation of the granulation tissue against the noninjured skin [21]. Microvessel density was assessed by counting vascular lumina with CD31-positive endothelial lining within the wound granulation tissue and is given as n/HPF.

Cell proliferation assay

To evaluate the influence of OXY on cell proliferation, we assessed the proliferative activity of fibroblasts (L929; mouse) and keratinocytes (HaCaT; human) by the water-soluble tetrazolium salt (WST-1) assay (Roche Diagnostics, Mannheim, Germany). Cells were seeded into 96-well microtiter plates at a concentration of 4×103 cells/well in Dulbecco’s modified Eagle’s medium (DMEM low-glucose; PAA, Cölbe, Germany), 10% fetal calf serum (FCS), and 1% penicillin/streptomycin supplemented with 10% FCS, basic fibroblast growth factor (bFGF; 20 ng/mL; R&D Systems GmbH, Wiesbaden, Germany), or OXY at three different concentrations of 10 nmol/mL (OXY 10; Sigma-Aldrich), 100 nmol/mL (OXY 100; Sigma-Aldrich), and 1000 nmol/mL (OXY 1000; Sigma-Aldrich). After incubation for 48 h, cells were washed with PBS and the cell proliferation reagent WST-1 was added to the cell culture medium followed by incubation for 4 h with repetitive measurements of optical density every 30 min. Sample absorbance was analyzed using a bichromatic ELISA reader (TECAN CM Sunrise; Grödig, Salzburg, Austria) at 450 nm. All experiments were performed in triplicate.

In vitro cell migration assay

To evaluate the effect of OXY on cell migration, fibroblasts (L929) and keratinocytes (HaCaT) were seeded onto Petri dishes (5×105 cells/dish) and grown to confluence in DMEM (DMEM low-glucose; PAA), 10% FCS, and 1% penicillin/streptomycin. After the removal of the medium, cell monolayers were scratched with a pipette tip (10 μL) to produce an artificial wound. Then, cells were re-exposed to medium supplemented with either 10% FCS or OXY at three different concentrations of 10 nmol/mL (OXY 10), 100 nmol/mL (OXY 100), and 1000 nmol/mL (OXY 1000). As positive stimulant for cell migration, we used bFGF (20 ng/mL; R&D Systems). The scratch area was photographed immediately as well as 24, 48, and 72 h after scratching. The digitized images were taken in grayscale format and analyzed densitometrically by the CapImage software. A random sample of a scratch area was taken, and the gray level necessary to mark the complete area of the cells, migrated into the scratch area, was determined. This gray value served as the threshold value for the analysis of all further images. Values of cell migration into the scratch area are given as the area covered by cells in percent of the initial scratch area. All experiments were performed in triplicate.

Statistical analysis

All data are given as means±SEM. Data were analyzed for normality and equal variance across groups. Results for continuous variables are presented as means±SEM. Because measurements of wound size were made several times on the same sample within five independent treatment groups (control, LD OXY, HD OXY, LD ATO, and HD ATO), we applied the GLM repeated-measures analysis of variance (ANOVA) for the statistical analysis of the data to test the null hypotheses about the effects of both the between-subject factor (treatment) and the within-subject factor (time). Overall statistical significance was set at p<0.05. For clarity and rapid interpretation of data, only the significant differences for comparison between groups at a single point of time are given. Statistics were performed using the software package SigmaStat (version 10.0; Jandel Corporation, San Rafael, CA, USA).

Results

Wound epithelialization

In all groups, the size of the wound area was comparable, with an average area of 4.9±0.2 mm2 immediately after wounding. Planimetric analysis of the wound area at the subsequent time points showed a continuous increase in epithelialization in the control group with a 51±3% wound coverage on day 6 and complete wound closure (99±1%) on day 12 after wounding (Figure 2). LD OXY or HD OXY application did not impair wound epithelialization, as wound coverage on day 6 showed 49±2% and 45±3% and an almost complete wound closure (99±1% and 98±2%) on day 12 (Figure 2). In LD ATO- and HD ATO-treated groups, similarly as observed in the control or OXY group, continuous epithelialization could be seen on day 6 (54±3% and 51±1%) and day 12 (98±1% and 98±2%), respectively (Figure 2).

Figure 2:

Analysis of wound epithelialization.

(A) Photomacroscopic images and quantitative planimetric analysis of wounds during regeneration, displaying the continuous process of wound closure with complete epithelialization on day 12. Left, skin fold chamber directly after wounding of the control group; right, wounds of the HD ATO group (dotted line, initial wound area; continuous line, wound area on day 6). (B) Quantitative analysis of wound epithelialization on days 3, 6, 9, and 12 in mice treated daily with saline (control; 0.9% NaCl; 12.5 mL/kg bw; n=8), LD OXY (1 mg/kg bw; n=9), HD OXY (10 mg/kg bw; n=6), LD ATO (1 mg/kg bw; n=7), or HD ATO (10 mg/kg bw; n=6). Data are means±SEM. No statistically significant differences.

Microvascular parameters

As described previously [21], the neovascularization process of the healing skin follows a distinct regeneration pattern with inner and outer rings of vessels, which has been named “sola cutis se reficientis”. Whereas the inner ring consists of circular vessels directly at the wound edges of the re-epithelializing wound surface, characterized by large and irregular vessel diameters, the outer area of the newly formed epithelium is described by radially running vessels with smaller and less heterogeneous diameters. The outer radial vessels supply the circular vessels at the wound margin and grow incessantly to the wound center until complete wound closure. In control animals, diameters of circular vessels increased until day 6 (~18 μm) and showed finally a diameter reduction toward ~15 μm on day 12, indicating vessel remodeling and maturation (Figures 1 and 3A). Circular vessels of LD OXY, HD OXY, and LD ATO animals showed similar diameters compared to controls over the 12-day observation period. In contrast, circular vessels of HD ATO showed an increase in vessel diameter (~22 μm) on day 6 (Figure 3A) but then demonstrated a maturation process to ~15 μm as seen in controls (Figure 3A). Radial vessels in the outer area of the re-epithelializing skin were smaller in diameter than the circular ones and could not be observed in wounds on day 3 (Figure 3B). Radial vessels did not significantly differ in diameter among groups on days 6, 9, and 12 after wounding, ranging between 11 and 12 μm (Figure 3B).

Figure 3:

Quantitative analysis of diameters (μm) and FMD (cm/cm2) in circular vessels (A and C) and radial vessels (B and D).

Animals were treated daily with saline (control; 0.9% NaCl; 12.5 mL/kg bw; n=8), LD OXY (1 mg/kg bw; n=9), HD OXY (10 mg/kg bw; n=6), LD ATO (1 mg/kg bw; n=7), or HD ATO (10 mg/kg bw; n=6). Data are means±SEM. *p<0.05 vs. control; #p<0.05 vs. HD OXY.

In saline-treated controls, but also in all OXY and all ATO animals, the FMD of circular vessels around the wound margin constantly decreased from ~150 to 50 cm/cm2 (Figure 3C). The FMD of radial vessels remained nearly constant in all groups with values ranging from 98 to 126 cm/cm2 but has been lower in OXY and ATO groups on day 9 compared to the control group (Figure 3D).

RBCV in the newly formed microvasculature did not significantly differ among groups and days and ranged between 176 and 328 μm/s for circular vessels and 319–448 μm/s in radial vessels, which did not significantly differ to values in normal skin (Table 1).

Table 1:

Quantitative analysis of RBCV in circular and radial vessels as well as in vessels in nontraumatized skin distal of the wound (DOW) in mice on days 3–12 after wounding treated daily with saline (control; 0.9% NaCl; 12.5 mL/kg bw; n=8), LD OXY (1 mg/kg bw; n=9), HD OXY (10 mg/kg bw; n=6), LD ATO (1 mg/kg bw; n=7), or HD ATO (10 mg/kg bw; n=6).

Groups	RBCV (μm/s)
	Circular vessels				Radial vessels				Vessels DOW
	Day 3	Day 6	Day 9	Day 12	Day 3	Day 6	Day 9	Day 12	Day 3	Day 6	Day 9	Day 12
Control	328±28	188±14	176±12	179±30	–	436±17	400±36	389±19	470±33	467±34	456±39	388±32
LD OXY	264±19	177±17	187±6	190±25	–	388±34	422±24	438±59	400±19	429±23	406±27	397±37
HD OXY	213±25	209±18	169±18	180±24	–	419±33	412±35	329±22	428±30	416±47	426±28	424±41
LD ATO	244±26	219±14	167±17	183±23	–	319±19	380±25	343±21	402±33	474±53	369±14	404±14
HD ATO	227±29	196±10	169±13	190±10	–	353±15	375±31	448±22	449±44	379±21	388±10	402±18

Data are means±SEM.

Wound tissue histology and immunohistochemistry

The threshold-based assessment of cellularity on H&E-stained paraffin sections revealed no significant different values in wounds of animals with LD OXY and HD OXY as well as LD ATO and HD ATO when compared to controls or among each other (Figure 4). These results support the hypothesis that OXY as well as ATO do not interfere with wound granulation tissue formation during the inflammatory phase of skin wound healing (Figure 4). In line with this, the granulation tissue of OXY- or ATO-treated animals presented with decreased leukocytic tissue infiltration on day 12 after wounding, but without statistically significant differences among groups (Figure 5). Interestingly, the HD OXY group showed the lowest leukocytic infiltration in the granulation tissue. The quantification of the neovascular process in the regenerating skin has been assessed by the morphometric quantification of the expression of the endothelial cell marker CD31 within the area of the granulation tissue (Figure 6). Although there have only been marginal differences in the in vivo analysis of angiogenesis, immunostaining for microvascular density showed significantly higher values in HD ATO (47±8 microvessels/HPF)-treated animals compared to HD OXY (23±5) and LD ATO (15±1; p≤0.05), whereas all other groups did not significantly differ among each other (Figure 6).

Figure 4:

Quantitative analysis (A) and representative images (B–D) of H&E staining for cellularity in wound tissue specimens on day 12 after wounding.

Animals were treated daily with saline (control; 0.9% NaCl; 12.5 mL/kg bw; n=8), LD OXY (1 mg/kg bw; n=9), HD OXY (10 mg/kg bw; n=6), LD ATO (1 mg/kg bw; n=7), or HD ATO (10 mg/kg bw; n=6). Data are means±SEM. No statistically significant differences. Scale bar, 80 μm.

Figure 5:

Quantitative analysis (A) and representative images (B–D) of leukocyte infiltration (AS-D CAE) in wound tissue specimens on day 12 after wounding.

Animals were treated daily with saline (control; 0.9% NaCl; 12.5 mL/kg bw; n=8), LD OXY (1 mg/kg bw; n=9), HD OXY (10 mg/kg bw; n=6), LD ATO (1 mg/kg bw; n=7), or HD ATO (10 mg/kg bw; n=6). Data are means±SEM. No statistically significant differences. Scale bar, 80 μm.

Figure 6:

Representative images (A) and quantitative analysis (B) of CD31-stained endothelial lining to determine microvessel density in wound tissue specimens after wounding.

Animals were treated daily with saline (control; 0.9% NaCl; 12.5 mL/kg bw; n=8), LD OXY (1 mg/kg bw; n=9), HD OXY (10 mg/kg bw; n=6), LD ATO (1 mg/kg bw; n=7), or HD ATO (10 mg/kg bw; n=6). Data are means±SEM. *p<0.05 vs. HD OXY; #p<0.05 vs. LD ATO. Scale bar, 120 μm.

In vitro cell proliferation

We analyzed the effects of OXY on fibroblast and keratinocyte proliferation using the WST-1 proliferation assay. The exposure of fibroblasts to 10% FCS as well as to OXY 10 significantly increased the proliferation rate in comparison to OXY 100 and OXY 1000 (Figure 7). Interestingly, the exposure to bFGF was not able to increase the proliferation rate of fibroblasts during the observation time frame of 240 min (Figure 7). Keratinocytes, however, seemed not to be influenced by OXY supplementation, as they did not significantly increase their proliferation activity as given by the absence of differences among groups (Figure 7).

Figure 7:

Quantitative assessment of in vitro cell proliferation by the WST-1 assay of (A) fibroblasts (L929) and (B) keratinocytes (HaCaT) over an observation period of 240 min.

Cells (4×103 cells/well) were treated with DMEM supplemented with 10% FCS, bFGF (20 ng/mL), or OXY in three different concentrations with 10 nmol/mL (OXY 10), 100 nmol/mL (OXY 100), and 1000 nmol/mL (OXY 1000). After incubation for 48 h, cells were washed with PBS and the cell proliferation reagent WST-1 was added to the cell culture medium followed by incubation for 4 h with repetitive measurements of optical density every 30 min. All experiments were performed in triplicate. Data are means±SEM. *p<0.05 vs. bFGF, OXY 100, and OXY 1000.

In vitro cell migration

The migrational viability of fibroblasts as assessed by means of the wound scratch assay was significantly increased by bFGF compared to controls (10% FCS) and OXY 10 at early time points (24–48 h after scratch). Fibroblasts that were exposed to OXY 1000 in the wound scratch assay showed significantly increased wound coverage at 48 h after scratch. At 72 h, the areas covered with cells were almost ≥80% and did not show significant differences (Figure 8). Of interest, however, there was no difference between the different supplements in keratinocyte migration during the first 48 h, whereas the exposure of keratinocytes to OXY 100 significantly impaired in vitro wound closure compared to 10% FCS, OXY 10, and OXY 1000 (Figure 8).

Figure 8:

Quantitative assessment of in vitro cell migration using the wound scratch assay with (A) fibroblasts (L929) and (B) keratinocytes (HaCaT) over an observation period of 72 h.

Cells were grown to confluence in 10% FCS on Petri dishes and then scratched with a pipette tip (C; scratch 0 h). Cells were treated with DMEM supplemented with 10% FCS, bFGF (20 ng/mL), or OXY in three different concentrations with 10 nmol/mL (OXY 10), 100 nmol/mL (OXY 100), and 1000 nmol/mL (OXY 1000). Cells were photographed immediately and 24, 48, and 72 h after the scratch, as representatively shown for fibroblasts among groups at the time point of 72 h after scratching (C). Scale bar, 400 μm. Data are means±SEM. *p<0.05 vs. control; #p<0.05 vs. bFGF; ßp<0.05 vs. OXY 10; §p<0.05 vs. OXY 10; $p<0.05 vs. OXY 100.

Laboratory analysis

Blood samples of saline-treated controls as well as of OXY- and ATO-treated animals revealed physiological but not statistically significant different values for erythrocytes, leukocytes, platelets, as well as hemoglobin and hematocrit (data not shown).

Discussion

The here presented study examined two major questions: (1) how systemic OXY application affects the repair process of full dermal thickness wounds in mice, specifically the epithelialization of the skin and its effects on keratinocytes and fibroblasts, and (2) how OXY influences the distinct effects on skin neovascularization. Based on these questions, repetitive systemic OXY application has been assessed in a model of full dermal thickness wound healing in the mouse dorsal skin fold chamber, as reported previously by our group [21], [24]. The model of the dorsal skin fold chamber in rodents displays a standardized technique, which allows a multifaceted use in experimental research [25], [26]. Besides this technique, a variety of other in vivo [27], [28], [29], [30] and in vitro [31], [32] models exist; however, only a limited number of models include the possibility of direct microcirculatory analysis during the process of dermal restoration [33], [34], [35]. In general, animal models are only able to mimic physiological as well as pathological human wound healing problems, with dehiscence, ischemia, ulceration, infection, and scarring [36]. However, variances in the tissue architecture, immune system functions, and general physiology among animals in contrast to humans must be taken into consideration [21].

In the here used approach, the analysis of skin wound healing in a systematically and standardized manner was performed. As the regeneration or repair process of the skin is not limited to epithelial recovery, the used model further allows the quantitative assessment of microcirculation and neovascularization of the healing skin in one and the same animal over time. Using this skin wound healing model, we could determine the specific regeneration profile of newly formed microvascular networks in the healing skin in all animals under investigation as regularly seen in other studies [21], [24]. As newly formed vessels underlie a distinct maturation process over time, as indicated by a reduction of vessel diameter as well as a decrease of vessel density [24], [37], OXY as well as ATO treatment did not interfere with the physiological process of new blood vessel formation in skin defects. This underscores that OXY application causes no harmful effects on angiogenesis during skin regeneration.

For the here presented study, we orientated us in the dosing regimens of previously done work in the context of OXY [8], [9], [13], [14], [15], [38], [39] and its selective OXY receptor antagonist ATO [15], [40], [41], [42], [43]. In contrast, treatment with the OXY antagonist ATO elevates oxidative stress in the hearts of the newborn rats [40], abolished the cardio preconditioning effect of OXY [41], and reversed the OXY effect on gastric ischemia-reperfusion injury [42].

As demonstrated in a previously published work, the OXY receptor could be verified in the wound microenvironment [44]. Here, the receptor is expressed in human foreskin fibroblasts as well as in dermal fibroblasts and keratinocytes [45], [46], [47]. However, OXY itself is expressed in human skin and can be predominantly localized in the epidermis [47]. Polymerase chain reaction studies confirmed the expression of OXY in both skin and cultured epidermal keratinocytes [47]. Moreover, OXY is expressed not only in keratinocytes [45] but also in human skin-derived dermal fibroblasts [46]. Immunohistochemical stainings of OXY in human skin confirmed an expression of OXY in all epidermal layers [46]. The OXY receptor, however, was preferentially expressed in the basal layers [46]. Deing et al. could also demonstrate that the inhibition of the OXY receptor signaling revealed an impact of OXY on the modulation of oxidative stress and cytokine release in dermal fibroblasts and keratinocytes [46]. Furthermore, OXY receptor knockdown was associated with an increased susceptibility to oxidative stress of skin cells [46]. Interestingly, the OXY receptor has also been described in rat adipocytes [48]. It can therefore be assumed that the given concentrations of OXY and ATO in the here presented study were sufficient enough to differentiate between certain effects of the treatment or its antagonism as well as versus the control group, only receiving saline treatment.

The healing of skin wounds is one of the most multifaceted biological processes in the mammalian organism, comprising different timely overlapping phases [1], [49], [50]. In each of these phases, there is a need of different cells and factors, which should help to regenerate and repair the traumatized tissue. The main objectives of sufficient healing are related to these steps and should finally lead to a fast epithelialization without major scarring. There are also many different influences such as surgical stress, immunosuppression, and reduced nutrition, which slow down or even arrest the physiological healing process. However, many different substances, such as growth factors, cytokines, and antibodies to specific receptors, have been evaluated to ameliorate skin wound healing [50]. As there are many new substances that are described to have a very wide spectrum of activity, the so-called pleiotropic agents, OXY also is described to have an influence on skin healing. OXY treatment, for example, led to significant smaller wound sizes of full dermal thickness punch wounds in female hamsters in comparison to the control treatment [12]. Furthermore, the application of an OXY antagonist showed opposite results with less healing velocity of the wounds [12]. Interestingly, the healing performance during immobilization as well as social stress was also associated with less cortisol concentrations, suggesting that social interactions promote skin wound healing as well [12]. Other studies assessed OXY in the context of burn injuries. Herein, it could be demonstrated that OXY is able to ameliorate skin regeneration as given by a fast epithelial recovery with mostly intact dermal structures [13], [51]. One major reason for these findings might also be the fact of better social interactions, as group housed rats as well as rats that have been given environmental enrichment presented with overall better healing performances most likely associated with increased OXY and decreased cortisol levels [12], [13], [51]. On the contrary, it could further be demonstrated that the application of OXY resulted in less remote-to-injury organ damage [51], [52], [53]. It has also been suggested that OXY might be involved in the modulation of the immune process and the inflammatory reaction via OXY receptors located in the thymus [54]. This might be due to the fact of less concentrations of tumor necrosis factor-α and interleukin-6, a decreased rate of apoptosis, diminished neutrophil recruitment, and less lipid peroxidation in the different organs under investigation [51]. In this context, OXY has been described to act as an anti-inflammatory agent via the increase of corticosterone levels even after a single OXY injection [7], [14], [15], [55] as well as an antioxidant via the regulation of the oxidant-antioxidant status [52], [53], [56], [57]. Furthermore, the anti-inflammatory OXY action was recently been reported to be a direct neutrophil-dependent mechanism, ameliorating the oxidative organ damage [52], [58]. In our study, the wounds of HD OXY animals revealed a two-fold lower number of infiltrating white blood cells compared to controls or the other OXY-treated group. Although the data have not been statistically significant different, it still shows the decreased effect of leukocytes within the granulation tissue of HD OXY-treated animals in comparison to control animals, indicating that decreased numbers of leukocytes might further contribute to better, at least not worsened, skin regeneration by a reduced inflammatory reaction [59], [60]. However, a clear mechanism on how OXY might lead to beneficial skin regeneration is still not clarified and should be newly discussed as the results of our in vivo model as well as the respective in vitro/ex vivo analysis could not show major differences in neither epithelialization, neovascularization, and wound cellularity nor inflammation.

However, some findings need to be discussed in more detail. Circular vessels of HD ATO showed an increase in vessel diameter on day 6 of the study. This seems to be a single difference in the study with no further consequence as circular vessels of HD ATO-treated animals did not show any differences on day 3, 9, or 12. Furthermore, circular HD ATO vessels after day 6 also decrease in their diameters until day 12, indicating vessel maturation [24]. Of interest, circular vessels are described as being directly at the wound edge and by this at the forefront of the microvascular regeneration [2], [21]. This so-called inner ring of heterogeneous vessels is characterized by large and irregular diameters, which markedly change over time as can also be seen in the reducing FMD from days 3 to 12 until they will completely disappear, when the wound is closed and the microvascular network has been regenerated [61]. OXY previously showed to have an important role in angiogenesis [61], [62]. Here, OXY induced the proliferation by binding to OXY receptors on human vascular endothelial cells [62], [63]. Furthermore, OXY stimulates the motility of immortalized human dermal microvascular and breast cancer-derived endothelial cells [64]. It can therefore be suggested that OXY may act as an endocrine/paracrine regulatory factor that contributes to angiogenesis [61]. This could further be supported by the finding that OXY receptor signaling promoted the angiogenic patterns of human umbilical vein endothelial cells (HUVECs) via Gli1-induced transcription of hypoxia-inducible factor-1α [62]. The lower FMD values for radial vessels on day 9, however, are in contrast to the described proangiogenic role of OXY, as OXY as well as ATO seem to slow down the regeneration of the microvascular network compared to the control group. However, as also shown in the studies of Sorg et al., radial vessels are subject to certain changes but not within the same range than circular vessels [2], [21], [24]. Therefore, this difference must be seen as a singular reduction of the FMD on day 9, which, however, seems still to be in the range of physiological differences over time [21].

The differences in the immunostaining for microvascular density showed significantly higher values in HD ATO-treated animals compared to HD OXY and LD ATO, which is again an interesting but contradictory point to what is known so far. Although there is no long-lasting difference in the OXY to ATO groups in the microvascular analysis over time (diameter, RBCV, and FMD), a marked difference could be detected especially of the HD ATO group in the CD31 staining. Especially, the difference to the LD ATO seems to be interesting, with the overall lowest microvascular density compared to the other groups. This is again in contrast to what is known on the use of ATO. Zhu et al. could demonstrate that the use of ATO (10 μM) significantly suppressed the angiogenic properties of HUVECs and clearly showed inhibitory function in comparison to OXY in their study [62]. Zhu et al. therefore concluded the proangiogenic role for OXY and its receptor [62]. However, this could not be detected in our study with HD ATO treatment.

Cellular behavior differed sometimes between fibroblasts and keratinocytes with regard to OXY treatment in our study. These results, however, are not surprising as both cell types are of different origin, showing different behavior patterns for migration and secretion. In the case of fibroblasts, there seems to be an effect that is comparable to the control group that was treated by 10% FCS only. Interestingly, the exposure to bFGF was not able to increase the proliferation rate of fibroblasts during the observation time frame of 240 min. However, the values of the bFGF and 10% FCS group are in accordance with what could be shown previously [65]. The impairment of keratinocytes in the wound scratch assays at 72 h is interesting. It can be seen throughout the three time points of evaluation that the treatment with OXY 100 showed decreased keratinocyte migration, which then got significant at 72 h. The differences range about 10%, which therefore seems to be negligible even when there is a statistical significant difference.

As many different groups reported about reduced stress levels, it might be likely that the change of glucocorticoid receptors or the cortisol levels as well as the overall reduction of stress-induced radical formation could be accounted for the assessed results [15]. Furthermore, insulin-like growth factor (IGF)-1 might also be a candidate for the mediation of OXY effects, as OXY-treated rats with musculocutaneous flaps showed better flap survival and had significantly elevated IGF-1 plasma levels [43]. Due to this knowledge, OXY seems to play an important role in skin wound healing issues. However, until now, the used transmitter system seems not to be clarified in detail and it might be proposed that it is associated with the stress response of the organism to various stimuli [7], [13], [40].

In conclusion, this study showed that OXY has no significant effects on physiological skin wound healing in vivo. Although other groups showed beneficial results, OXY seems at least not to be harmful in ordinary skin wound healing. Furthermore, the current knowledge on the favorable effects of good social interaction on healing processes solely associated with OXY might be newly challenged and deserves further research work.

Click here for additional data file.