TNF-alpha expression, evaluation of collagen, and TUNEL of Matricaria recutita L. extract and triamcinolone on oral ulcer in diabetic rats

ABSTRACT Diabetes mellitus (DM) is a disease associated with delayed wound healing of oral ulcers by increased expression of proinflammatory cytokines and cellular apoptosis. Objective to evaluate the influence of Tumor Necrosis Factor alpha (TNF-α) and apoptosis in rats with DM treated with chamomile extract or triamcinolone. Material and Methods Wistar male rats (210.0±4.2 g) were divided into five groups: negative control group (NCG) without diabetes; positive control group (PCG) with DM (alloxan, 45 mg/kg); and groups treated with chamomile extract (normoglycemic= NCG group and diabetic= DCG group) and with triamcinolone (TG). Traumatic ulcers were performed on all animals that received topical triamcinolone, chamomile extract or saline 12/12 hours for ten days. Results On days five and ten the animals were euthanized and the ulcers were analyzed by light microscopy, TUNEL assay, and immunohistochemically (TNF-α). The NCG (p=0.0062), PCG (p=0.0285), NCG (p=0.0041), and DCG (p<0.0001) groups were completely healed on the 10th day, however, there was no healing on the TG (p=0.5127) group. The TNF-α expression showed a significant reduction from the 5th to the 10th day in NCG (p=0.0266) and DCG (p=0.0062). In connective tissue, the TUNEL assay showed a significant reduction in the number of positive cells in NCG (p=0.0273) and CNG (p=0.0469) and in the epithelium only in CDG (p=0.0320). Conclusions Chamomile extract can optimize the healing of traumatic oral ulcers in diabetic rats through the reduction of apoptosis in the epithelium and TNF-α expression.


INTRODUCTION
Diabetes Mellitus (DM) is a chronic metabolic disease characterized by deficiency in insulin production or resistance to its action, resulting in hyperglycemia and metabolic alterations 4 . The incidence of DM is increasing in the world, and it is considered the biggest health problem in XXI century 18 . It is estimated that in 2025 there will be twice as many diabetic patients compared with the year 2000, totaling approximately 300 million affected individuals worldwide 20 .
Chronic hyperglycemia causes numerous events that promote structural changes in tissue. It is associated with delayed wound healing, increased susceptibility to infection, alterations in neutrophil activity, and reduction of chemotaxis, adhesion, phagocytosis and angiogenesis 4,7 . Clinically, the wound healing disorder manifests itself as hypertrophic scars or chronic unhealed wounds (ulcers), being ulcers the most prevalent problem in healing 18 .
In the oral cavity, traumatic ulcers are caused by mechanical trauma due to maladjusted dentures, orthodontic brackets, accidental bites, or iatrogenic factors. Typically, when the causal agent is removed, healing occurs spontaneously from one to two weeks; however, in a few cases, the ulcer can persist for longer periods of time. It can be extremely painful and interfere with eating and speaking 6 .
Corticosteroids are commonly prescribed for the treatment of painful symptoms of traumatic oral ulcers 6 ; however, conflicting results have been reported in literature regarding the effects of this therapeutic modality on the healing process. Glucocorticoids have potent anti-inflammatory and immunosuppressive effects 1 . Triamcinolone is commonly used in clinical dentistry because of its analgesic effects on oral ulcers. It has a potent antiscores of mucositis and pain in patients undertaking radiotherapy 12 . But corticoids used in treatment wound healing 1, 12 .
In contrast, the use of natural products in the treatment of ulcerated oral lesions has increased over the past several decades 23 . The plant chamomile, also known as Chamomilla recutita L. or Matricaria recutita L., which is a member of the Asteraceae family, is one of the most widely used medicinal plants 10 apigenin), terpenes and acetylated derivatives, antifungal, antioxidant, hypocholesterolemic, and sedative properties 9, 10,14,19,21,26,27 .
These drugs have been indiscriminately used in dental clinic for the treatment of persistent ulcerative lesions, but mechanisms of the diseases can be different in diabetic patients. Hyperglycemia wound closure and healing and interfere in collagen deposition 3,25 . Drugs used in the treatment of oral modifying these parameters. Thus, the aim of this Factor alpha (TNF-) and apoptosis in rats with DM treated with chamomile extract or triamcinolone.

Animals
This study was approved by the Ethics Committee for Animal Research (protocol no. 11/11), and was performed in accordance with the Ethical Principles in Animal Experimentation adopted by the Brazilian College of Animal Experimentation (COBEA).
We used male Wistar rats (Rattus norvegicus albinus, Rodentia mammalia) weighing 210.0±4.2 g (Mean±SD) that were provided by the Central Animal Facility of the Federal University of Ceará (UFC). The rats were kept in the Animal Laboratory of Experimental Oncology (LOE), given an initial examination for systemic health conditions and stored in boxes with sawdust. All animals received industrial feed (Bio-base ® , Águas Frias, SC, Brazil) accordingly and water ad libitum and were kept at room temperature with controlled humidity for a photoperiod of 12 hours.

Diabetes induction
The induction of diabetes was performed by injection of alloxan (45 mg/kg) in diluted sterile saline (0.9%) intravenously after mild sedation with ether. Two milliliters of blood were collected 48 h after the induction of diabetes from the retro-orbital plexus for determination of blood glucose. The animals were considered diabetic when the blood glucose was equal to or greater than 200 mg/dL 24 . blood was collected again for glucose measurement Animals with a blood glucose level of less than 200 mg/dL were excluded.

Experimental protocol to induce the ulcers
For the induction of ulcers, the animals were anesthetized with intraperitoneal ketamine (80 mg/kg) and xilazin (10 mg/kg). Antisepsis was performed with an oral solution of 0.12% chlorhexidine gluconate in cotton pellets. The ulceration in the left buccal mucosa was performed by abrasion with a number 15 scalpel blade, and a marker with an 8 mm diameter was used for standardization of the lesion area. The surgical technique was standardized for all animals and performed by the same operator 9 .

Groups and treatment
groups by lot: -Groups with Saline Treatment: Negative Control Group (normoglycemic rats) and a Positive Control Group (diabetic rats); -Groups with Chamomile Treatment: Chamomile Normoglycemic Group (normoglycemic rats treated with chamomile) and Chamomile Diabetic Group (diabetic rats treated with chamomile); -Groups with Triamcinolone Treatment: Triamcinolone Group (diabetic rats treated with Triamcinolone). topical application of sterile saline solution in the Negative Control Group and Positive Control Group, Omcilon-A, orabase ® , 1 mg/g (B-MS, São Paulo, SP, Brazil) in the Triamcinolone Group, or Ad-Muc ® 10% ointment (BIOLAB, São Paulo, SP, Brazil) in the Chamomile Group (diabetic and normoglycemic). The application of the drugs was performed with an individual sterile and disposable microbrush (KG Sorensen ® , Cotia, SP, Brazil). and ten days after ulcer induction.

Clinical evaluation
The animals were weighed and had the blood a 0.5 mm digital pachymeter (L=larger diameter; m=minor diameter) to calculate the area of the All measurements were performed by the same operator 8 . Weight, blood glucose, and ulcer diameter were evaluated in the 5 th and 10 th days.

Histological analysis
the specimens were macroscopically analyzed, subjected to dehydration in crescent alcoholic series, melted at 60º C. Then, the fragments were placed The fragments were sectioned to a thickness of 5 μm using a microtome (Leica Geosystems, Atlanta, Georgia, USA ® ), and histology was performed using routine coloration by hematoxylin-eosin (HE). The histopathological parameters were determined and scored from 0 to 4 according to previously published criteria (Figure 1

Collagenesis assay
The same fragments were sectioned to a thickness of 3 μm and analyzed by histology using picrosirius staining to evaluate collagen deposition. , Naperville, Illinois, USA). The photomicrographs were quantitatively evaluated using image analysis software (ImageJ ® , National Institutes of Health, Bethesda, Maryland, USA) after calibrating images by the Color Threshold command (Image>Adjust>Color Threshold) in the RGB function for colors Red (minimum 71, maximum 255), Green (minimum 0, maximum 69) and Blue (minimum 0, maximum 92). After calibration, images were converted to a color scale of 8-bits (Image>Type>8-bit) and binarized (Process> Binary>Make Binary) before the percentage of collagen in a given area, marked in black, was analyzed (Analyze>Analyze Particles). The mean of tree percentages was used as a sample unit 2 .

Immunohistochemistry assay
After deparaffinization and rehydration, immunohistochemical assay. Antigenic recuperation was performed by heat in citrate pH 6.0 solution. After cooling, the slides were submitted to peroxidase blocking with H 2 O 2 3% solution diluted in PBS (phosphate buffered saline) for 30 minutes.
The percentage of cells in connective tissue with cytoplasmatic or nuclear expression was grouped into scores: (0) no positive cells; (1 -mild) 1-33% of positive cells; (2 -moderate) 34-66% of positive cells; (3 -intense) 67-100% positive cells. The same score, obtained by two or more observers, 11 . TUNEL assay to antigen retrieval with Proteinase K (Dako ® , Carpinteria, CA, USA) (1:250) for 15 minutes and peroxidase blocking with H 2 O 2 3% solution diluted in PBS (phosphate buffered saline) for 5 minutes. After peroxidase blocking, the slides were incubated for 10 seconds with PBS buffer (ready to use) and with Tdt (1:2.5) for 90 minutes at room temperature. The tissue slides were rinsed with the wash buffer (1:15) for 10 minutes and after this the anti-digoxigenin conjugate (ready to use) was added for 30 minutes. The revelation system was done with DAB (Dako ® , Carpinteria, CA, USA). The tissue slides were stained with metil green for 10 minutes followed by dehydration, diaphanization, and mounting with coverslips.

Statistical analysis
The area of ulceration, weight loss, blood glucose change, and percentage of collagen in an area were analyzed by the t-test and ANOVA/Tukey test. The Mann-Whitney test and Kruskall-Wallis/Dunn test were used to histological scores. The relationship between the area of ulceration, weight loss and blood glucose variation in the four experimental groups was analyzed by Pearson correlation. The relationship between these parameters and the histological scores were analyzed by Spearman correlation. All analyses were performed with the software GraphPad Prism 5.0 ® (GraphPad Software Inc., San Diego, California, USA), with a p<0.05

Clinical evaluation
Dimensional analysis of the ulcer: All experimental groups showed a decrease in ulcer area from the 5 th day to the 10 th day, except the Triamcinolone Group. On the 10 th experimental day, only the group higher ulcer size compared with the Positive Control Group ( Figure 2, Table 1).
Weight loss: After the treatment, Negative Control, Positive Control, Chamomile Diabetic, and Triamcinolone Groups had weight loss relative to day zero. In Negative Control Group and Chamomile recovery of weight from the 5 th day to the 10 th day. The diabetic groups showed no recovery of weight.
loss compared with the Positive Control Group on the 5 th and 10 th days of the experiment. The Normoglycemic group from the 5 th day to the 10 th day (Table 1) (Table 1).

Histological analysis
Histological analysis: On the 5 th day of treatment, Negative Control Group, Positive Control Group, and Triamcinolone Group presented a median score of 4, characterized by the presence of an ulcer treated with chamomile extract (normoglycemic and diabetic) presented a median score of 3, with infiltrate (mononuclear cells and neutrophilic granulocytes) in the underlying connective tissue. On the 10 th experimental day, the Negative Control Group and the Chamomile Diabetic Group presented a median score of 1, represented by    (Figure 4, Table 2).

TNF-immunostaining
In connective tissue, the immunostaining for th day. All groups showed a high score (median=3) of TNFexpression by elevated number of connective tissue cells expressing TNF-in cytoplasm, except Chamomile Normoglycemic Group (median=1). On the 10 th day the Negative Control Group (median=1) and the Chamomile Diabetic Group scores (p=0.0266 and p=0.0062, respectively). Positive Control Group (median=3, p=0.3367), Chamomile Normoglycemic Group (p=0.1683), and Triamcinolone Group (median=3, p=1.0000) did not showed similar reduction, exhibiting a high number of TNF-positive connective tissue cells. The score of the Triamcinolone Group on the 10 th day was statistically higher than chamomile groups ( Figure 5, Table 3).

Apoptosis assay
Connective tissue: In connective tissue, the number of TUNEL positive cells was elevated. All groups showed a high number of cells in apoptosis (median=3). Only Negative Control Group (p=0.0273) and Chamomile Normoglycemic Group (p=0.0469) showed reduction in score of TUNEL staining on the 10 th day (median=1). On the 10 th day all animals of Positive Control Group and Triamcinolone Group exhibited more than 66% of TUNEL positive cells (median=3, minimum=3, maximum=3) with statistical difference between Negative Control Group. Chamomile groups did not showed difference in scores of TUNEL positive cells in relation to Negative Control Group ( Figure 6).
Epithelium: Similarly, all groups exhibited more than 66% of TUNEL positive epithelial cells on the 5 th day. On the 10 th day, only Chamomile Normoglycemic Group showed reduction in score of TUNEL positive epithelial cells (median=2, p=0.0320), but there is no differences regarding  Positive Control Group (median=3) and Negative Control Group (median=3). The Triamcinolone Group score (median=3) was higher than Chamomile Normoglycemic and diabetic groups on the 10 th day ( Figure 6).

Correlation analyses
area of the oral ulcer and weight variation in the Negative Control Group (p=0.002, r=-0.766) and the Positive Control Group (p=0.010, r=-0.763) because the weight loss was proportional to the area of the ulcer. In other groups, the correlation In the Positive Control Group (p=0.011, r=0.728), the correlation between weight variation weight loss proportional to glucose lowering.
Histological scores presented significant correlation directly with the area of the oral ulcer variation (Table 3).

DISCUSSION
High levels of blood glucose affect oral wound healing negatively. Wound healing in an uncontrolled diabetic patient occurs more slowly compared with normoglycemic or controlled diabetic patients 13 . This delay in oral wound healing can cause chronicity of oral lesions in patients with DM 18 .
In this study, a wound healing deficit was observed in diabetic rats. On the 10 th experimental day, we observed characteristics compatible with tissue healing in the normoglycemic group (score=1), but the diabetic group showed of the underlying connective tissue (score=2).
in collagen deposition in the diabetic rats when compared with the normoglycemic rats at both time points. Reinforcing the histopathology analysis, the macroscopic evaluation showed that on the 10 th experimental day the oral ulcers were completely healed in the normoglycemic rats, but not in the diabetic animals.
Animals treated with chamomile extract showed the 5 th experimental day, unlike Negative Control, Positive Control, and Triamcinolone Groups. Additionally, on the 10 th for the Chamomile Group were similar to those for the Negative Control Group (normoglycemic), showing total epithelial healing and mild chronic inflammation. In the fibrogenesis evaluation, collagen deposition in the Chamomile Group was Group (diabetic) and did not differ from the Negative Control Group (normoglycemic).
been linked to compounds present in its extract derivatives 21,26,27 . The mechanism of action of direct inhibition of ciclooxigenase-2 and synthesis E2 28 . In this study, the chamomile group (DCG) was expression in diabetic rats that are likely to augment actions lead to a reduction in vascular and cellular alpha-bisabolol has been associated with promoter activity in the formation of granulation tissue during the process of wound repair 4 . Additionally, the reduction of apoptosis scores only in connective tissue of NCG may be associated that is proportional to TNF-activation 29 . NCG showed high reduction in TNF scores in connective tissue and have a potent antioxidant activity, leading to reduction of apoptosis only in connective tissue.
The increase in TNF-and oxidative stress due to elevated levels of glucose decreases cell viability and survival of various cell types, 17,30 . However, there was no cells in any treatment. Only Negative Control TNF-alpha expression, evaluation of collagen, and TUNEL of Matricaria recutita L. extract and triamcinolone on oral ulcer in diabetic rats Group had reduction in apoptosis in connective tissue. The reduction in TNF-expression occurred concomitantly with decreased epithelial cells apoptosis in CDG showing that it can modulate apoptosis of these cells when topically used. In a study with diabetic rats, oral administration of chamomile extract showed anti-hyperglycemic effects and was able to ameliorate the complications caused by hyperglycemia associated with DM 9, 14 . In our study, the animals treated with chamomile showed significant reduction of blood glucose variation from the 5 th to the 10 th experimental days.
However, despite reports of the anti-hyperglycemic effects of systemic chamomile, further researches chamomile was only topically applied to the oral ulcers.
In contrast, therapy with corticosteroids in traumatic lesions has negative effects on wound healing. Steroids affect the synthesis and maturation of collagen, alter the tensile strength of the wounds, inhibit cell function and proliferation, and decrease the antibacterial and phagocytic actions of some defense cells, resulting in a delay in wound healing 1,22 . In this study, the Triamcinolone Group exhibited an oral ulcer area Negative Control Group and the NCG on the 10 th experimental day. The triamcinolone administered topically delayed the repair process, which caused the 5 th and 10 th experimental days. This effect was not observed in other groups. The Triamcinolone Group exhibited less collagen deposition and higher weight loss than the Positive Control Group (diabetic rats), suggesting that this drug inhibits additional wound healing in the oral cavity in diabetic rats.
of blood glucose variation on both the 5 th and 10 th experimental days in animals treated with chamomile and triamcinolone when compared with the untreated diabetic rats. These data can be associated with the antinociceptive effects of both drugs 10 . Chamomile and triamcinolone can reduce nociception, leading to increased food intake and a consequent augment in blood glucose. The size of the ulcerative process in the oral cavity is proportional to the painful symptomatology, and topical drugs with anesthetic or analgesic effects diminish this relationship 16 . Thus, this model is supported by the analysis of the correlation between the area of the oral ulcer and weight loss, since the weight loss was proportional to the area of the ulcer only in the untreated animals (Positive Control Group and Negative Control Group).
The weight loss occurs due to diabetic anorexia, which is coupled with nociception and consequent dysphagia resulting from oral ulceration 3 . This combination may interfere with feeding and be responsible for the variation in blood glucose and diabetic animals 15 . The results of the present study showed weight loss in all of the experimental diabetic groups; however, the Triamcinolone Group had Control Group. Corticosteroids promote a complex metabolic process that results in malnutrition and catabolism. Consequently, this poor health condition of the animals may contribute to the delayed healing and the increase in weight loss 1 .
The corticosteroids, even administered topically, wound healing. The extract of Matricaria recutita L.
(chamomile) may be an important tool to optimize the healing process of oral traumatic ulcers in DM

CONCLUSION
Thus, chamomile extract can optimize the healing of traumatic oral ulcers in diabetic rats through the reduction of apoptosis in the epithelium and TNF-expression. On the other hand, steroid therapy had negative effects on the healing process of oral ulcers.