Adipose Mesenchymal Stem Cells Combined with Platelet Rich Plasma Accelerate Diabetic Wound Healing by Modulating Notch Pathway

Background Diabetic foot ulceration is a serious chronic complication of diabetes mellitus characterized by high disability, mortality and morbidity. Platelet-rich plasma (PRP) has been widely used for diabetic wound healing due to its high content of growth factors. However, its application is limited due to rapid degradation of growth factors. The present study aimed to evaluate the ecacy of combined adipose derived mesenchymal stem cells (ADSCs) and PRP therapy in promoting diabetic wound healing in relation to the Notch signaling pathway. Methods Albino rats were allocated into 6 groups (control, sham, diabetic, PRP-treated, ADSCs-treated and PRP+ADSCs-treated groups). The effect of individual and combined therapy was evaluated by assessing wound closure rate, epidermal thickness, dermal collagen and angiogenesis. Moreover, gene and protein expression of key elements of Notch signaling pathway (Notch1, Delta like canonical Notch ligand 4 (DLL4), Hairy Enhancer of Split-1 (Hes1), Hey1, Jagged-1), gene expression of angiogenic marker (Vascular endothelial growth factor & stromal cell-derived factor 1) and epidermal stem cells (EPSCs) related gene (ß1 Integrin) were assessed. Results Our data showed a strong wound-healing effect of PRP+ADSCs compared to their individual use after 7 and 14 days. Combined therapy caused marked increase in area percentage of collagen, epidermal thickness and angiogenesis. Moreover, Notch signaling was signicantly down-regulated, EPSCs proliferation and recruitment was enhanced compared to other treated groups and diabetic group. Conclusions These data demonstrated that PRP and ADSCs combined therapy signicantly accelerated healing of diabetic wounds induced experimentally in rats via modulating Notch pathway, promoting angiogenesis and EPSCs proliferation.


Introduction
Diabetes mellitus (DM) is a worldwide health problem affecting approximately 9.3% of the global population with its prevalence expected to rise by 25% in 2030 and 51% in 2045 (1). Diabetic foot ulceration (DFU) is one of the most common chronic diabetic complications leading to signi cant medical, economic, and social burdens. It is estimated that every 30 seconds, a complicated diabetic lower limb is lost worldwide as 15% to 25% of diabetic patients have a risk to develop a foot ulcer throughout their whole lifetime (2).
The strong regenerative & healing capabilities of skin are intimately linked to the existence of skin stem cells. The skin stem cells (predominantly epidermal stem cells (EPSCs) and hair follicle stem cells) are considered as important sources of cells for skin healing, regeneration, and metabolism and are located in the basal layer of the epidermis and also, in the hair follicle bulge which contains the most potent EPSCs (3). Typically, the more the remaining skin stem cells upon the wound surface, the particularly faster the curing rate, and the particularly less the scar formation. Normal cutaneous tissue has a diversity of stem cells with multipotent potentiality leading to, theoretically, any wound can easily rely on skin stem cells to reach physiological repair. Nevertheless, in profound skin wounds, the remaining skin stem cells cannot experience normal differentiation capacity and complete the anatomical structure and functional skin repair according to the preprogrammed pathways. Therefore, the healing progression may be disrupted, ultimately developing scar tissue devoid of hair follicle and sweat glands (4). This indicates that the process of wound healing is associated with interaction among cells, complex regulation of the extracellular matrix, and various paracrine elements (5).
Provided the complexity of the multifactorial and multicellular processes of wound healing, it is believable that a therapeutic approach targeting different signaling pathways that control cellular processes signi cant for wound healing would likely serve as a considerable solution for DFU therapy.
Notch signaling pathway is an evolutionarily preserved signaling mechanism with an extremely pleiotropic action. Notch signaling is essential for cell-fate determination. Besides, it also plays a critical role in regulating proliferation, angiogenesis, and apoptosis/survival, processes that are intensely disturbed in diabetic wounds (6). Notch signaling is actually a cell-cell interaction mechanism triggered as a result of the interaction between membrane-bound Notch receptors (Notch 1-4) and their particular ligands (Delta-like 1, 3, 4, and Jagged 1-2) on juxtaposed cells. This interaction induces γ-secretasemediated cleavage and translocation of the Notch intracellular domain (ICD) into the nucleus, where it forms a transcriptional activation complex inducing the expression of downstream target genes, such as Hairy Enhancer of Split-1 (Hes1) and Hey1 (7). This signaling pathway performs essential tasks during development and throughout the regulation of adult tissue homeostasis. Besides, it plays a vital role in the postnatal physiology of the skin as well as in normal wound healing through the positive regulation of cell migration, angiogenesis, and in ammation (8).
Several innovative treatment options intended for DFU management have been discovered, like bioengineered skin substitutes, negative pressure dressings, and hyperbaric oxygen therapy. Nevertheless, the typically effective therapies are remaining inadequate. Therefore, it is essential to use more successful and e cient therapies like the use of autologous biologics, such as mesenchymal stem cell (MSC)-based therapies and platelet-rich plasma (PRP), which hold considerable promise to improve tissue regeneration in addition to chronic wound care management strategies (9). PRP can be acquired throughout an autologous manner, via centrifugation of the patient's blood leading to a plasma fraction with a platelet concentration greater than that of the circulating blood. The therapeutic properties of PRP are mostly endorsed to the release of platelet growth factors after its activation. These group of growth factors includes epidermal growth factor (EGF), platelet-derived growth factor (PDGF), broblast growth factor (FGF), vascular endothelial growth factor (VEGF), and insulin-like growth factor (IGF-1, IGF2) which are recognized to favor tissue regeneration (10). On the other hand, adipose derived MSCs are adult multipotent stem cells with self-renewal potentiality, which can differentiate directly into different lineages and secrete paracrine elements starting the process of tissue regeneration. The plentiful supply of adipose tissue, simple isolation procedure, wide proliferative capabilities ex vivo, besides their capacity to secrete pro-angiogenic growth factors made them an ideal cell type to be used as a new therapy for the treatment of non-healed wounds. Moreover, the MSC secretome initiates healing simply by inducing a shift from pro-in ammatory to anti-in ammatory cytokine production at the injury site (11).
In the present study, we aimed to evaluate the therapeutic e ciency of PRP and ADSCs both individually and in combination in diabetic wound healing. Furthermore, we investigated the role of Notch signaling pathway in controlling the wound healing.

Experimental Animals
Adult male albino rats (180-200 g), 6 weeks of age, were purchased from the Experimental Animal Unit, Faculty of Veterinary Medicine, Benha University, Egypt. The rats were bred and maintained in an airconditioned animal house under speci c pathogen-free conditions. All animals were housed in clean cages and given a standard diet and clean water ad libitum. The rats were subjected to a normal light/dark cycle (12-h light-dark cycle starting at 8:00 AM) and room temperature (23 ±3 °C and allowed free access to chow and water. This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health (NIH publication No. 85-23, revised 2011). All protocols were approved by the institutional review board for animal experiments of the Faculty of Medicine, Benha University, Egypt (BUFM 3 January 2018).

Adipose Derived Stem Cell (ADSCs) Preparation
Adipose tissues from the abdominal wall of rats were obtained and then placed into a labeled sterile tube containing 15 mL of a phosphate buffered solution (PBS; Gibco/Invitrogen, Grand Island, NY, USA). Enzymatic digestion was performed using 0.075% collagenase II (SERVA Electrophoresis GmbH, Heidelberg, Germany) in Hank's Balanced Salt Solution for 60 min at 37 °C with shaking. Digested tissue was ltered and centrifuged, and erythrocytes were removed by treatment with an erythrocyte lysis buffer.
The cells were transferred to tissue culture asks with Dulbecco Modi ed Eagle Medium (DMEM, Gibco/ BRL, Grand Island, New York, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco/BRL) and after an attachment period of 24 h, non-adherent cells were removed by a PBS wash. Attached cells were cultured in DMEM supplemented with 10% FBS, 1% penicillin-streptomycin (Gibco/BRL), and 1.25 mg/L amphotericin B (Gibco/BRL), and expanded in vitro. At 80-90% con uence, cultures were washed twice with PBS and the cells were trypsinized with 0.25% trypsin in 1 mM EDTA (Gibco/BRL) for 5 min at 37 °C.
After centrifugation, cells were resuspended in serum-supplemented medium and incubated in 50 cm 2 culture ask (Falcon). The resulting cultures were referred to as rst-passage cultures and expanded in vitro until passage three (12).

Immunophenotypic Characterization of Differentiated ADSCs
ADSCs were initially characterized by their adhesiveness, fusiform morphology, and by detection of established surface markers of rat ADSCs by ow cytometry. Following isolation ADSCs were passaged, viable cell counts established, and aliquoted individually at 1 x10 6 cells/mL per tube. ADSCs were then incubated with 10 µL of directly conjugated monoclonal antibodies; CD45 PE (rabbit monoclonal; [EP322Y] (ab200315), CD90 PE (mouse monoclonal Antibody (HIS51), bioscience, # 14-0900-81) and CD 105 PE (rabbit polyclonal antibody, CENTER E395; SAB1306487 Sigma Aldrich) at 4 °C in the dark for 20 minutes; matched isotype controls were included for control purposes. Following incubation, 2 mL of PBS containing 2% FCS solution was added to each tube followed by centrifugation for 5 min at 2500 rpm, discarding of the supernatant, and resuspending in 500 μL PBS containing 2% FCS. Cell analysis was performed using CYTOMICS FC 500 Flow Cytometer (Beckman Coulter, Brea, CA, USA) and CXP software version 2.2 (14).

Isolation of human platelet-rich plasma (PRP)
Whole blood was collected from rats into acid citrate dextrose solution A (ACD-A) anticoagulant at a ratio of 1 mL ACD-A: 9 mL blood. To separate platelets from erythrocytes and leukocytes in plasma, 40 mL of this mixture was put into a 50-mL centrifuge tube and centrifuged at 160 × g for 10 min, and then the separated plasma containing platelets was transferred to a new centrifuge tube and centrifuged at 250 × g for another 15 min. Most of the supernatant plasma was discarded, before the platelet pellet was resuspended in the residual plasma to obtain 4 mL PRP, then activated with 10% CaCl 2 to become PRP gel to prevent its leakage from the wound (16).

Induction of DM
Type I diabetes was induced in overnight fasted rats by a single intraperitoneal (IP) injection of freshly prepared Streptozotocin (STZ powder was obtained from Sigma-Aldrich Chemical Co., St. Louis, MO, USA; 60 mg/kg, dissolved in 0.1M cold citrate buffer, pH 4.5). After STZ injection, rats acquired drinking water containing sucrose (15 g/L) for 48 h, to lessen the early death due to insulin discharge from partially injured pancreatic islets. Seventy-two hours later, rats were checked for hyperglycemia, and those with fasting blood sugar more than 250 mg/dL were included in the study. Diabetic rats received long-acting insulin (2-4 U/rat) via subcutaneous injection to maintain blood glucose levels in a desirable range (350 mg/dL) and to avoid subsequent development of ketonuria (17). Animals were maintained in a diabetic state for 6 weeks before the start of the wound-healing experiment.

Wound Model
The Galiano's murine healing model was used (18) as this model minimizes rodent wound contractions and therefore mimics the wound healing processes occurring in humans including granulation tissue formation and reepithelialization. Rats were anesthetized with iso urane gas (SEDICO, Egypt) inhalation (2.5% in 500 ml/min of air), and surgeries were performed under standard sterile conditions. Two circular, full-thickness 5mm diameter cutaneous wounds were in ected on the back of each rat, and sterile donutshaped silicone splints with a diameter two times of the wound were xed to the surrounding wound edge with an adhesive lm (3M™ Steri-Strip™ Skin Closures, 3M Science, Egypt) and interrupted 6-0 silk thread sutures to prevent skin retraction. The wounds were then covered with semi occlusive dressing (3M Tegaderm®, Egypt). During all the experiments, rats daily received intraperitoneal injection of buprenorphine (0.1 mg/kg/day).

Wound Closure Analysis
Wound closures were blinded quanti ed through the measure of the wound reepithelialization at day 3, day 7, day 10 and day 14 post-surgery, through a macroscopic analysis of the lesions on the back of rats. A disposable 10-centimeter medical paper wound measuring ruler was used to measure the wound size. The wound closure rate at day X postsurgery was calculated as the percentage of the wound area at day X compared with that postoperative day 0 (9).

Experimental Design and Treatment Protocol
The experimental design is shown in Figure 1. Ninety-eight male rats were randomly divided into six groups as follows: Group I (control group; n = 21): Rats were fed a regular chow diet for 6 weeks. The rats were divided equally into three subgroups of 7 rats each: Subgroup Ia: The rats were left without any intervention.
Subgroup Ib: The rats were injected intraperitoneal with a single dose of 0.25 mL/kg body weight sodium citrate buffer (vehicle for STZ).
Subgroup Ic: Rats were injected with 100 μl saline solution + CaCl2 on the back of each rat.
Subgroup Id: Rats were injected with sterile phosphate-buffered saline solution on the back of each rat.
Group II (Sham operation group; n = 14): Rats were fed a regular chow diet for 6 weeks, then the wound was in ected at the back of each rat and immediately after injury the wound base &edges were injected with 100 μl saline.
Group III (Diabetic group; n=14): After six weeks of DM induction, the wound was in ected on the back of each rat and immediately after injury the wound base &edges were injected with 100 μl saline.
Group IV (DM+PRP group; n= 14): After six weeks of DM induction, the wound was in ected on the back of each rat and immediately after that the wound base & edges were injected with 4 mL PRP activated with 10% CaCl 2 .
Group V (DM+ADSCs group; n= 14): After six weeks of DM induction, the wound was in ected on the back of each rat and immediately after that the wound base & edges were injected with 100 μl of saline solution containing 2 × 10 6 ADSCs.
Group VI: (DM+ADSCs +PRP group; n= 14): After six weeks of DM induction, the wound was in ected on the back of each rat and immediately after that the wound base & edges were injected with 100 μl of saline solution containing 2 × 10 6 ADSCs-in combination with 4 mL PRP activated with 10% CaCl 2 .

Sampling
Rats in each group (except the control group) were equally subdivided into two subgroups (a & b) as follow: Rats in subgroup a were sacri ced after 7 days of wound in ection to assess the in ammatory phase of wound healing while rats in subgroup b were sacri ced after 14 days of wound induction to assess the proliferative phase of wound healing. In each subgroup the samples were taken from the wound site of the ulcerative tissue.
Half of the skin tissues were collected from rat to be evaluated by light microscopy with hematoxylin and eosin (H&E) and Masson's trichrome staining. An immunohistochemical evaluation for PCNA and CD31were also performed. The other half of skin fresh tissue specimens were kept frozen at −80 °C for later quantitative real-time polymerase chain reaction (qRT-PCR) to assess the gene expression of Notch1, Dll4, Jag1, Hes1, Hey1, VEGF, SDF-1&EPSCm (ß1 Integrin). Western blot analysis was also performed to assess the protein expression of Notch 1, Jag1 and Hes1.

Gene expression pro le
Total RNA was extracted from the skin specimens of treated and control rats using TRIzol (Invitrogen) according to the manufacturer's instructions. The concentration and purity of extracted RNA were measured by the Nano-Drop 2000C spectrophotometer (Thermo Scienti c, USA). At absorbance ratio A260/A280, RNA purity for all samples was > 1.9. The integrity of RNA was veri ed on 2% agarose gel using a gel electrophoresis image (Gel Doc. BioRad) (19). Complementary DNA (cDNA) was synthesized for the target genes using SensiFast cDNA synthesis kits (Sigma Bioline, UK) according to the manufacturer's instruction using a T100 Thermal Cycler (Bio-Rad, USA).

Gene
Sequences ( Quantitative PCR was performed using Maxima SYBR Green/ROX qPCR master mix (2x) (Thermo Scienti c, USA) (20). Primer pairs for selected target and reference genes (Notch 1, Dll4, Hes1, Hey1, Jag1, VEGF, SDF1, EPSCm and GAPDH) were purchased from Genwez (New Jersey, USA) ( Table 1). Each PCR reaction consisted of 500 ng per reaction of cDNA (except for NTC and cDNA control), 12.5 μl Maxima SYBR Green qPCR Master Mix (Maxima SYBR Green qPCR, ThermoFisher Scienti c), 0.3 μmol l−1 of each forward and reverse primer, 10 nmol l−1/100 Nm ROX Solution, nucleases-free water to a nal volume of 25 μl. The reaction was completed in AriaMx Real-Time PCR (Agilent Technologies, USA) using a two steps protocol: initial denaturation at 95 ºC for 10 min, then 40 cycles of denaturation at 95 ºC for 15 s followed by annealing/extension at 60 ºC for 60 s. A melting curve protocol was run at the end of the PCR by heating at 95 ºC for 30 s followed by a 65 ºC for 30 s and 95 ºC for 30 s. The expression levels of target genes were normalized to the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Relative gene expression ratios (RQ) between treated and control groups were calculated using the formula: RQ = 2 -ΔΔCt (21). The resultant cell lysates were agitated on ice for 30 minutes followed by centrifugation at 21,000g for 30 minutes to collect the supernatant. Protein concentration was estimated by the Folin-Lowry method using a spectrophotometer (Beckman Coulter Inc, Indianapolis, Indiana). The extracted protein was incubated in Laemmli buffer for 10 minutes at 95°C. Protein (50 mg loading) was resolved using 10% SDS (sodium dodecyl sulfate) polyacrylamide gel electrophoresis and transferred onto polyvinylidene di uoride membrane (Amersham Biosciences, Bucks, United Kingdom). The blot was blocked with 5% nonfat dry milk (NFDM) in 1% TBS with Tween 20 overnight at 4°C. The membrane was incubated with anti-Notch 1, Jag1 and Hes1and GAPDH antibodies (1:500; Ab3209, Millipore) at room temperature for 2 hours followed by incubation with goat anti-rabbit horseradish peroxidase-conjugated secondary antibody (1:5000, Millipore) for 2 hours at room temperature.
Detection was performed with Super Signal West Femto substrate (Thermo Scienti c, Waltham, Massachusetts) on photographic lms (Eastman Kodak Co, Rochester, New York). The later blot was stripped using stripping buffer (62.5 mmol Tris, 2% DS, 100 mmol β-Mercaptoethanol) for 10 minutes at 60°C to detect housekeeping protein. GAPDH was used as housekeeping protein and detected using MAB1501 (Millipore) at a 1:5000 dilutions. After washing twice with 1X TBST, densitometric analysis of the immunoblots was performed using Image analysis software on the Chemi Doc MP imaging system (Version 3) produced by Bio-Rad (Hercules, CA).

Histological Analysis
At the end of the experiment, the rats were anesthetized by sodium thiopental (40 mg/kg IP) after 12 h of fasting. Then, vascular perfusion xation through the left ventricle was performed. The rats were xed on an operating table to take skin specimens. The half of the skin tissue which collected from rats in all groups, was xed in 10% buffered formol saline, embedded in para n, and sectioned at 4.0 μm. The sections were dehydrated with successive concentrations of ethanol and washed twice in distilled water.
The sections of the skin tissue at day 7 &14 were stained with hematoxylin and eosin (H&E) and with Masson's trichrome in accordance with the protocols of the manufacturer to detect the reepithelialization/ granulation tissue formation and collagen deposition, respectively. Finally, the histological sections were observed and analyzed under a microscope (Leica DMR 3000; Leica Microsystem) by two blinded experienced investigators (30).

Immunohistochemical staining
For detection of new vessel formation, the wound areas were analyzed using CD31 primary antibody (rabbit monoclonal primary antibody, SAB5500059, Sigma Aldrich, USA). For the detection of basal keratinocyte proliferation, the PCNA primary antibody was used (rabbit polyclonal Anti-Proliferating Cell Nuclear Antigen antibody, SAB2701819, Sigma-Aldrich, USA). For detection of adipose derived MSCs(ADSCs) in cutaneous tissues after injection in both treated groups group IV(DM+ADSCs) & group V (DM+ADSCs+PRP), the CD105 antibody (rabbit, polyclonal primary antibody, SAB1306487, Sigma Aldrich, USA) was used.
Brie y, after xing, embedding in para n, and dewaxing, the tissue sections were blocked in 3 % normal goat serum/0.3 % Triton X-100/ 0.1 % BSA (Sigma Aldrich) in PBS. The sections were then incubated for 24 hours at 4 °C with the primary antibody (primary) against CD31 & PCNA respectively (1:100 dilution), followed by goat anti-mouse IgG (secondary) for each primary antibody for 1 hour at room temperature.
After hematoxylin staining, tissue sections were washed and then dehydrated with ethanol, treated with dimethylbenzene, and sealed for microscopic analysis by two blinded experienced investigators (30).

Morphometric study
The mean area percentage of collagen ber deposition as indicated by Masson's trichrome staining, the mean area percentage of PCNA, the number of CD31 positive vessels were quanti ed to detect the neoangiogenesis and CD105 surface marker of ADSCs were measured in ve images from ve non overlapping elds from each rat of each group using the Image-Pro Plus program version 6.0 (Media Cybernetics Inc., Bethesda, Maryland, USA).

Statistical Analysis
Statistical analysis was performed using the statistical software package SPSS for Windows (Version 16.0; SPSS Inc., Chicago, IL, USA). GraphPad Prism 8.0.2 (GraphPad Software, SanDiego, CA, USA) was used for graphical representation. Differences between groups were evaluated using one-way Analysis of Variance (ANOVA) followed by Tukey test, and Kruskal Wallis followed by Mann-Whitney U test regarding parametric and non-parametric data, respectively. Data are expressed as mean ± standard error (SEM) (parametric) and median (non-parametric), P-value <0.05 was considered signi cant.

Con rmation of adipose derived MSC isolation
ADSCs were initially identi ed after 2 weeks' isolation in culture by an inverted microscope as spindle- Enhanced wound closure rate in ADSCS+PRP group via modulating Notch signaling pathway The mean wound closure rate at days 3, 7, 10, and 14 post wound in ection were 21.5 ± 2.0, 57.2 ± 1.3, 71.5 ± 0.9, and 90.9 ± 1.1% respectively in the sham group (group II) while, in the diabetic group (group III), the wound closure rates were signi cantly decreased (4.7 ± 0.6, 0.9 ± 0.5, 20.   (Figure 4h). At the 14 th day (group Vb), the epidermis appeared thicker, with more prominent keratohyalin granules and keratin. The papillary layer showed more collagen bers and blood capillaries, while the reticular layer had thick collagen bundles (Figure 4i).
PRP+ADSCs treated group at the 7 th day (group VIa) revealed intact thin skin. The epidermis showed all keratinocyte layers. Fine collagen bers and blood capillaries in the papillary layer, and thick collagen bundles in the reticular layer (Figure 4j). At the 14 th day (group VIb), the epidermis appeared thicker.
Abundant, ne, interlacing collagen bers and numerous blood capillaries were seen in the papillary layer and reticular layers showed thick collagen bundles and numerous hair follicles (Figure 4k).

Masson's trichrome results
In the control group (group I), dermal collagen was seen as ne interlacing bers in the papillary layer, and thick, irregular blue bundles in the reticular layer (Figure 5a). The sham group at the 7 th day (group IIa) showed ne collagen bers in the papillary layer and thick collagen bundles parallel to the surface in the reticular layer (Figure 5b). At the 14 th day (group IIb), abundant ne interlacing collagen bers were seen in the papillary layer and thick collagen bundles in the reticular layer (Figure 5c). The diabetic group (group III) revealed an evident decrease in collagen in both the papillary and reticular layers. At the 7 th & 14 th day (group IIIa &IIIb), both layers showed pale ne collagen bers with some areas de cient collagen deposition. Few areas showed thick collagen bundles at 14 th day (Figure 5d &5e). On the other hand, the treated groups (group IV, V&VI) showed a progressive increase in ne collagen ber deposition in the papillary layer with improvement in the organization of the thick collagen bundles which ran parallel to the surface in the reticular layer. The best collagen deposition and organization was observed in group VI compared to the control group (Figure 5f-5k).

Immunohistochemistry staining results
Immunohistochemical staining with anti-PCNA antibody was performed to assess cellular proliferation of the epidermis in the wound area. In group I, the basal layer of keratinocytes showed intense reaction in many cells (Figure 6a), while in group II, there was moderate reaction on 7th day (Figure 6b) and the reaction became intense on the 14th day in the basal layer of keratinocytes, comparable to that of the control group (Figure 6c). On the other hand, in group III, a weak reaction was observed in basal layer of keratinocytes at the 7th and 14th day (Figure 6d & 6e). In the treated groups (group IV, V &VI), the intensity of reaction in basal layer of keratinocytes increased from moderate to intense reaching a reaction comparable to the control at the14th day of group VI ( gure 6f-k).
Immunohistochemical detection of angiogenesis (new capillary formation) was performed using endothelial CD31 marker. In the control group (group I), a moderate CD31 expression was observed in the capillaries of the papillary layer of the dermis (Figure 7a). In the sham group (group II) CD31 expression was observed at the 7 th day which peaked at the 14 th day, (Figure 7b &7c). In the diabetic group (group III), a negative CD31 reaction was observed on both days (Figure 7d, e). On the other hand, the treated groups (groups IV, V, and VI) showed new vessel formation with CD31 immuno-expression at the 7 th day, and peaked at the 14 th day (Figure 6f-6k).

Morphometric study
In H&E stained sections, the mean epithelial thickness was 236.4±1.1 μm in the control group (group I), while in the sham group (group II) it was 219.8±1.0 μm on the 7 th day and 233.7±1.4 μm on the 14 th day.
In the diabetic group (group III), the signi cant decreased epithelial thickness was observed at both days To elucidate the role of Notch signaling pathway relevant to diabetic wound healing, the expression of Notch receptor (Notch1), two Notch ligands (Dll4 & Jag1) and two Notch target genes (Hes1 & Hey1) were detected. qPCR was used to assess the expression of Notch1 pathway-related genes in the diabetic wound.
As shown in Figure 8, Sham group (group II) at day 7 showed signi cant upregulation of Notch1 and its downstream genes including Notch1, Dll4, Jag1, Hes1 and Hey1 when compared to control group (group I) (p < 0.01). Nevertheless, Group II at day 14 showed non-signi cant upregulation of Notch1 and its downstream genes when compared to group I. In contrast, Notch1 and its downstream genes showed signi cant higher expression levels in diabetic group (group III) than those in group I (p < 0.01). The expression of Notch1 and its downstream genes, in all treated groups (IV, V and VI), was found to be signi cantly decreased compared to group III. Group VI at day 14 showed no signi cant changes in the expression of Notch1 pathway-related genes when compared to either group I or group II. On the other hand, group VI showed signi cant changes compared to group IV and group V both at day 7 and day 14 (p < 0.01).
Angiogenic gene (VEGF), epidermal stem cell-related genes (EPSCm and SDF-1) were signi cantly upregulated in Sham group (group II) compared with those in control group (group I) (p < 0.05 for VEGF and SDF-1 and p < 0.01 for EPSCm at day 7, p < 0.05 for the three genes at day 14). Diabetic group (group III) displayed signi cant downregulation of these genes compared to either group I or II (p < 0.01).
Nevertheless, these genes were unchanged between group III and group IV. Further, both group V and VI displayed signi cant up-regulation of these genes when compared to group III (P<0.01). Moreover, group VI showed signi cant changes compared to group IV and group V both at day 7 and day 14.
Effect of PRP and/or ADMSCs on protein levels of Notch1 pathway key elements in rat diabetic wound Western blot results of Notch1, Jag1 and Hes1 showed a similar trend to that of the gene expression analysis and con rmed it.
As shown in Figure 9, Sham group (group II) at day 7 showed signi cant upregulation of protein levels of Notch1 (p < 0.05), Hes1 (p < 0.01) and Jag 1 (p < 0.05) when compared to group I. However, at day 14 group II showed non-signi cant upregulation of the aforementioned proteins when compared to control group (group I).
Further, Notch1, Hes1 and Jag 1 showed signi cant higher expression levels in group III than those in group I (p < 0.05). The expression of Notch1 in all treated groups (IV, V and VI) was found to be signi cantly decreased compared to group III (p < 0.05). The expression of Hes 1, and Jag1 showed nonsigni cant decrease in group IV and V and signi cant decrease in group VI at day 7, and signi cant decrease in all treated groups at day 14 when compared to group III. Group VI at day 14 showed no signi cant changes in the expression of Notch1 pathway-related genes when compared to either group I or group II (p < 0.05).

Discussion
Wound healing is de nitely a complex process coordinated by numerous molecular events leading to closure of the wound with or without scar formation. Typically, the events that occur soon after a skin injury could be allocated into four overlapping phases: coagulation and hemostasis, in ammation, proliferation and remodeling. The proper and coordinated progress of such processes is essential to a normal and effective wound healing. Insu cient wound healing take place when one or more underlying molecular processes within the different phases are usually disrupted (32). Therefore, wound healing is a natural response, but in severe or chronic conditions, such as burns and diabetes, this process is insu cient to achieve effective repair.
Epidermal stem cells (EPSCs) are a multipotent cell type and are committed to the formation and differentiation of the functional epidermis (33). The microenvironment of stem cells, called "stem cell niches," performs a key role in regulating the stem cells proliferation, migration and differentiation throughout a network system of numerous interconnected signaling pathways. Among these pathways, the Notch signaling pathway which is essential constituents of stem cell "niches" which play a vital role in skin development and wound repair. After skin injury, cytokines concentration as well as the extracellular matrix (ECM) components are changed resulting in stem cell niche affection and Notch signaling pathway activation. Hence, the proliferation and differentiation of wound EPSCs are prompted, ultimately contributing to wound healing or scar formation (34).
The sham operated group of the current study proved that the activation of Notch1 pathway resulted in improved wound closure as evidenced by improved epidermis layer thickness, rejuvenated skin appendages along with more organized and regular collagen bers arrangement leading to intact epidermis and dermis, which were more or less normal in structure. Moreover, the EPSCs marker (ß Integrin), Notch1 and its ligands DLL4 & Jag1, with its downstream target genes Hes1 & Hey1 were signi cantly increased. Hes1 &Hey1 are considered as chief target genes in the Notch1 signaling pathway, and they play a vital role in maintaining the proliferation potential of EPSCs con rming that the Moreover, sham operated group showed a signi cant increased Jag 1 expression, which is the rst ligand of Notch receptor expressed in all skin layers. It plays a signi cant role in controlling the differentiation of EPSCs (37). Concomitant with these results, Chigurupati 2007 (38), veri ed that mice treated with the Notch ligand, Jagged, showed accelerated wound closure (as assessed by surface wound size) suggesting that these effects were mediated by the Notch pathway, so, Jag1 mediated the "dialogue" of Notch signaling in cutaneous tissue, controlling EPSC proliferation and differentiation as well as playing a role in wound healing and scar formation. This explains that, the Notch signaling pathway could affect the biological microenvironment ("niche") of EPSCs. This could be attributed to the role of Notch signaling pathway in the angiogenic process. Indeed, vascular endothelial cells express receptors Notch1, 4 and ligands Delta-like 1, 4 in addition to Jag1. Thus, Notch1 and Dll4 play vital role in angiogenic budding. The budding is directed by endothelial tip cells which express high amounts of Dll4.
For this purpose, Dll4 is placed at the protruding front directed towards to angiogenic signals (39). Such outcomes were supported by the results of the current study, as evidenced by the improved vascularity in the wound healing process of sham operated group, which is con rmed by the signi cant increase in CD31 immuno-expression indicating new blood vessels formation & by the signi cant increase in VEGF &STDF-1 gene expression. These results were explained by Chigurupati et al 2007 (38), who reported that Notch signaling affected several behaviors of vascular endothelial cells which are critical for angiogenesis. Angiogenesis includes endothelial cell migration into the surrounding tissue, cell proliferation, alignment and tube formation, recruitment of parenchymal cells and a return to quiescence.
So, activation of Notch in wound healing process enhanced vascular endothelial cell proliferation, migration and tube formation (40).
Regarding the in ammatory phase of the wound healing, the in ammatory cell in ltration of sham operated group in the current study was increased during the rst week of wound healing con rmed by histological study then decreased after that in the proliferation phase of the wound healing. Consistent with these ndings, Kimbal et al. 2017 (41)  The expression of SDF-1, a vital factor for recruitment of EPCs is negatively controlled by Notch signaling. This provides a mechanism for how angiogenesis is regulated via Notch signaling. In diabetic wound healing, pathologically activated Notch pathway impairs EPCs incorporation into the wound site secondary to decreased SDF-1 expression. This leads to a signi cant defect that contributes to impaired wound healing in diabetes Caiado et al 2007 (44). In addition, chronic hyperglycemia leads to upregulation of DII4, activating both canonical and rapid non-canonical Notch1 pathways. Subsequently, a hyperglycemia-induced Dll4-Notch1 positive feedback loop has been recognized to contribute to pathogenic sustained Notch activation in diabetes (45). This is in concordance with the negative impact of Dll4-dependent Notch1 signaling on angiogenesis (8).
Therefore, Notch inhibition in diabetic wounds lead to improvement in EPSCs proliferation as well angiogenesis via facilitating the recruitment of EPCs. These results re ected the profound consequences of an increased Notch signaling for diabetic wounds (46). Subsequently, Notch1 signaling blockage both in vitro as well as in vivo through either genetic or pharmacological methods was found to enhance wound healing in diabetes, via numerous mechanisms central for wound healing as cellular proliferation, migration, and angiogenesis. This signi es that Notch1 signaling is a new potential therapeutic target for diabetic wound (47), (48), (49).
Interestingly, recent studies have demonstrated that cell therapy and growth factors enhance diabetic wound healing. These researches suggested that MSCs have the ability to differentiate into other different cell types within the injured tissue to stimulate repair and regeneration of skin. Also, PRP may provide a suitable microenvironment for MSCs to enhance proliferation and differentiation (9). Therefore, in the current study, diabetic wounds were treated with adipose derived MSCs (ADSCs) and PRP both individually and in combination to compare their therapeutic e cacy in diabetic wound healing and to assess their relation to Notch signaling pathway.
The current study revealed that Notch1 pathway-related genes were signi cantly downregulated to near the normal levels in ADSCs plus PRP group compared to the other treated groups and the diabetic group.
There In addition, results of the current study regarding angiogenesis clearly suggest that the ADSC and PRP therapy could induce a strong angiogenic effect in wound healing as CD31 immuno-expression as well as the gene expression of VEGF &STDF-1 were signi cantly increased in all treated groups with better results in ADSCs plus PRP group in comparison with diabetic wound group. This angiogenic response is critical for wound healing as it is necessary to increase oxygen and nutrient supply and to ensure the survival of keratinocytes. Moreover, it helps in sustaining the newly formed granulation tissue composed of a large number of blood vessels, which were typically functioning and contained red blood corpuscles. The SDF-1 secreted from MSCs induces the survival of vascular endothelial cells, promotes vascular branching, and pericytes recruitment. These paracrine effects of MSCs play essential roles in angiogenesis rather than their direct differentiation to endothelial cells and/or pericytes (51). In addition, some studies have revealed that angiogenic factors like VEGF, angiopoietins and hepatic growth factor, which are released from injured tissue assist the recruitment of MSCs to the wound site (53 They found that inhibiting c-secretase decreased the proliferation of human MSC but it did not modify the expression of the osteoblast markers. This information suggested the essential role of Notch signaling pathway in regulation of MSCs proliferation.     The H&E-stained skin sections revealed that PRP+ADSCs-treatment accelerate the diabetic wound healing: (a) group I showed normal epidermis (E) and dermis with its papillary(P) and reticular(R) layers. Masson's trichrome staining of cutaneous tissue was performed to assess dermal collagen different experimental groups (a-k), as there was ne interlacing bers (arrow) in the papillary (P) layer, and thick, irregular blue bundles (bold arrow parallel) to the surface in the reticular (R) layer. In group III there were areas of de cient collagen deposition are seen beneath the epidermis (asterisk). (l) Mean area percentage of dermal collagen. Data are expressed as median (maximum and minimum), ** signi cant compared to control group at p<0.01, * signi cant compared to control group at p<0.05, $$ signi cant compared to Sham group at p<0.01, # signi cant compared to diabetic group at p<0.05, ## signi cant compared to diabetic group at p<0.01, ££ signi cant compared to group IV and V at p<0.01, ~ signi cant compared to group V at p<0.05, ~~ signi cant compared to group IV at p<0.01.