Effect of red propolis on hamster cheek pouch angiogenesis in a new sponge implant model 1

Purpose: To evaluate the effects of red propolis on cheek pouch angiogenesis in a hamster new model sponge implant. Methods: Forty eight animals divided into eight groups. (Groups I-IV), the animals were treated for 15 days before and 10 days after sponge implantation. (Groups V-VIII), the animals were treated for 10 days after sponge implantation (GI and GV: red propolis 100 mg/kg, GII and GVI: celecoxib 20 mg/kg, GIII and GVII: 1% gum arabic 5 mL/kg, GIV and GVIII: distilled water 5 mL/kg). On the 11th day of implantation, the animals were anesthetized for stereoscopic microscopic imaging and morphometric quantification of angiogenesis (SQAN), followed by histopathological evaluation (H&E). Results: In the SQAN analysis, no significant difference was found between the groups. However, on histology, propolis was found reduce the population of mastocytes in the qualitative analyses (p = 0,013) in the quantitative analyses to reduce the number of blood vessels (p = 0,007), and increase the macrophage count (p = 0,001). Conclusion: Red propolis inhibited inflammatory angiogenesis when administered before andcontinuously after sponge implant, and was shown to have immunomodulating effects on inflammatory cells (mastocytes and macrophages) in a new sponge implant hamster model.

the animals were treated for 15 days before and 10 days after sponge implantation.In the second arm (Groups V-VIII), the animals were treated for 10 days after sponge implantation (GI and GV=red propolis 100 mg/5 mL/kg, GII and GVI=celecoxib 20 mg/kg, GIII and GVII=1% gum arabic 5 mL/kg, GIV and GVIII=distilled water 5 mL/kg) (Figure 1).

Red propolis
Propolis in natura was acquired from a trusted supplier in Barra de Santo Antônio (Alagoas).Because propolis is water-insoluble, extraction was done with gum arabic, following the protocol described by Shulka, Bhaudaria and Jadon (2004).
In general, inflammatory processes are regulated by molecules capable of modulating vascular and cellular mechanismswhich promote angiogenesis 3 .Chronic inflammation results from persistent inflammatory stimuli (infectious, physical or chemical agents) triggering the release of inflammatory mediators 4 and the activation of endothelial cells and the immune system, leading to angiogenesis 5 .
Angiogenesis is the formation of new blood vessels from existing ones 6 .The regulation of angiogenesisrequires a finetuned balance between molecules stimulating and inhibiting angiogenesis 7 .Pathological angiogenesis occurs through deficient or excessive neovascularization 8 , one example that disorders is the inflammatory angiogenesis in psoriasis 9 .
The purpose of this study was to evaluate the effect of red propolis on cheek pouch angiogenesis in a new sponge implant hamster model.

■ Methods
The study protocol followed the guidelines of the Brazilian Society of Animal Experimentation (COBEA) and was approved by the Animal Research Ethics Committee (CEUA), Universidade Federal do Ceará (UFC) (nº 05/2016).
The study had two arms with 4 groups of 6 animals each.In the first arm (Groups I-IV),

Celecoxib (Celebra ® 200 mg, Pfizer)
The anti-inflammatory agent celecoxibe (C 17H14F3N3O2S) was diluted in sterilized distilled water and administered subcutaneously in the dorsocervical region at 20 mg/kg body weight.

Sponge
The implant consisted of a fragment of polypropylene sponge measuring 3 mm (diameter) by 1 mm (thickness).The fragment was cut with a manual hollow hole punch cutter (Elizabeth Graziano EPP) and sterilized in an autoclave.

Surgical procedure
Following anesthesia with intraperitoneal administration of ketamine (100 mg/kg body weight) and xylazine (10 mg/kg body weight), the cheek pouches were cleansed for food residues with sterile saline using a sterilized (needleless) plastic syringe and, when necessary, sterilized gauze.The left pouch was then everted and stretched with forceps and two fingers, forming a symmetrical epithelial fold with exposure of the medial face.Using an insulin needle, 0.1 mL saline was injected into the subepithelium to make room for the implant (Figure 2A).Using colibri forceps, a 1-mm opening was made at the site of the needle perforation (Figure 2B) into which the sponge fragment was implanted (Figure 2C).The implant site was gently compressed with the finger and, following hydration with 0.9% saline, the pouch was inverted back into the oral cavity.

Cheek pouch angiogenesis
On the 11th day of sponge implantation, the animals were anesthetized with ketamine (100 mg/kg body weight) and xylazine (20 mg/kg body weight) and submitted to a macroscopic cheek pouch evaluation.To do so, the left pouch was everted and cleansed for food residues with sterile saline.After verifying the presence of the implant and of possible purulent discharge, the base of the pouch was resected with a harmonic scalpel, prioritizing cauterization over resection.The resected pouch segment was fastened with insulin needles on a rubber surface and washed with saline to prepare the tissue for photography.Once photographed, the pouch segment was tied to acetate plates with nylon line and preserved in plastic vials containing 10% buffered formaldehyde.

Image acquisition and processing
The panoramic images of the cheek pouch were acquired at 16x magnification (Figure 3).The camera (PixelView ® PV-TV304P, Prolink Microsystems Corp. Taiwan) was coupled to a surgical microscope (D.F.Vasconcellos M90, São Paulo) and connected to a microcomputer.

Quantification of angiogenesis
The microphotographs were processed with a system (SQAN v. 1.00, 2005) 10 built specifically to quantify angiogensis (Figure 4).The system provides information on three variables: area of neovascularization, total vascular length, and number of blood vessels.The neovascular response was then quantified in the panoramic images (magnification: x16).

Staining with hematoxylin-eosin
The resected cheek pouch segments were cleaved, embedded in paraffin, processed, sliced and stained with H&E.

Statistical analysis
Continuous and discrete quantitative variables were submitted to the Kolmogorov-Smirnov test for normality of distribution.The descriptive statistics included mean values and standard deviation (parametric variables) or median, minimum and maximum values and interquartile intervals (nonparametric variables).The four groups in each study arm were compared pairwise with ANOVA associated with Tukey's multiple comparison test to detect differences between groups (parametric variables), or with the Kruskal-Wallis test followed by Dunn's multiple comparison test (nonparametric variables).Analyses and graphs of angiogenesis data were made with the software GraphPad Prism v. 5.00 (GraphPad Software, San Diego, California, USA).All tests were two-tailed, and the level of statistical significance was set at 5% (p < 0.05).
The groups were also compared with Student's t test or ANOVA with regard to inflammation, followed by Bonferroni's test associated with Fisher's multiple comparion test or Pearson's chi-square test.Inflammation levels were submitted to the Kolmogorov-Smirnov test of normality, expressed as mean values and standard error of the mean, and compared with Student's t test or ANOVA followed by Bonferroni's post-test (parametric variables).Analyses of inflammation data were made with IBM SPSS Statistics v. 20.0 for Windows, with a confidence interval of 95%.

Quantification of angiogenesis (SQAN)
No significant difference was observed within the groups of each study arm (I, II, III and IV, and V, VI, VII and VIII) with regard to the amount of angiogenesis.The first arm of the experimente with the respective values expressed in pixels (F = 2.6712; p = 0.0752) and percentage (F = 2.6711; p = 0.0752) and the second arm of the experiment (treatment only) with the respective values expressed in pixels (F = 0.6255; p = 0.6069) and percentage (F=0.6255;p = 0.6069).

Quantification of inflammation (scoring system)
First arm of the experiment (prevention + treatment): Groups I, II III and IV Groups I, II, III and IV did not differ significantly with regard to the following histological parameters: microabscess (p = 0.407), macrophages (p = 0.299), lymphocytes (p = 0.268), giant cells (p = 0.210), foreign body granuloma (p = 0.210) and fibrosis (p = 0.469).However, reduced mastocyte levels were more often observed in Groups I and II than in the water control group (p = 0.013), while moderate (33.3%) and mild (83.3%) mastocyte infiltration was more common in Groups III and IV.

First arm of the experiment (prevention + treatment): Groups I, II III and IV
The number of lymphocytes (p = 0.720) and giant cells (p = 0.103) did not differ significantly between Groups I, II, III and IV, but the number of blood vessels was significantly smaller in the groups treated with red propolis and celecoxib than in the control group (p= 0.007) (Figure 5), the number of mastocytes was significantly smaller in the group treated with celecoxib than in the control group (p = 0.002) (Figure 6), and the number of macrophages was significantly greater in Group I than in Groups I, II and III (p = 0.001) (Figure 7) (Table 1).Second arm of the experiment (treatment only): Groups V, VI, VII and VIII Groups V, VI, VII and VIII did not differ significantly with regard to mastocytes (p = 0.165), macrophages (p = 0.690) or lymphocytes (p = 0.079), but the number of blood vessels was significantly greater in the group treated with celecoxib (p = 0.008) and the number of giant cells was significantly smaller in the groups treated with red propolis, celecoxib and gum arabic than in the water control group (p < 0.001) (Table 2).

■ Discussion
Synthetic implants (such as sponge) may be used to study different components of inflammatory and angiogenic processes 11 .In addition to studies on inflammatory angiogenesis, such implants are increasingly employed in tissue engineering research due to the interrelatedness of biocompatibility, angiogenesis and inflammation 12 .
The study of cheek pouch inflammatory angiogenesis induced by sponge implants relies on the use of new and reproducible models.In this study, we developed a model simulating a clinical setting of inflammation and angiogenesis caused by the presence of a foreign body (sponge) in hamster cheek pouch epithelium.
The hamster cheek pouch can be accessed easily by eversion, with little or no risk of trauma.The epithelial wall of the pouch is translucid; this facilitates visualization and helps avoid damage to blood vessels and potential antigenic contact through the blood.After inversion back into the oral cavity, the pouch wall accommodates the graft, protecting it against external trauma, making postoperative bandages unnecessary 13 , despite the risk of trauma associated with food storage.
Nevertheless, complications may occur at the implant site.Some of the animals in Groups I, III, V, VI, VII and VIII displayed sponge extrusion, pus, ulcerations, infiltration, and adhesions (they were excluded from the analysis).In general, the presence of food residues in the pouch increases the risk of contamination.
The choice of the 11th postoperative day for the evaluation of angiogenesis was based on a pilot experiment in which absorbable and non-absorbable sponge implants were studied macroscopically every other day.Absorbable material was found to be inadequate since the short time of absorption (up to 15 days) compromised the standardization of the inflammatory stimulus (foreign body) and subsequent evaluation of inflammatory angiogenesis.This problem was avoided by using non-absorbable material.
In our model, the test substances were administered subcutaneously.Alternatively, the sponge implant might have been soaked in the test substance, a method which remains to be tested.
The inflammation induced by the presence of the implant promotes changes like vasodilation, increased vascular permeability, and increased leukocyte adhesion and migration through the walls of the blood vessels.Neutrophils are the most abundant leukocytes in acute inflammation; when activated, they produce reactive oxygen species (free radicals), in addition to monocytes, lymphocytes and macrophages, all of which are characteristic of chronic inflammation and produce free radicals 14 .
When the level of reactive oxygen species overwhelms the antioxidant defense system, excess free radicals oxidate the membrane lipids, causing damage to nucleic acids-the molecular basis of several inflammatory conditions 15 .In this study, pouch tissues stained with H&E revealed increased levels of neutrophils, lymphocytes, mastocytes, macrophages and giant cells associated with inflammation.
The relationship between inflammatory angiogenesis and the production of free radicals by inflammatory cells highlights the need for the development of antioxidant compounds to prevent oxidative stress, which is etiopathogenic of a range of inflammatory conditions 16 .
The first study on the effect of propolis on inflammatory angiogenesis was published in 1999, suggesting propolis attenuates inflammation by inhibiting cyclooxygenase and lipoxygenase 18 .Another study explained the anti-inflammatory effect of propolis as resulting from anti-angiogenic mechanisms 19 .Yet others have reported anti-angiogenic effects with green propolis 20 .
However, in the morphometric quantification of angiogeneis (SQAN), which included an analysis of panoramic stereoscopic images of the vascular area around the implant and an analysis of quadrant mean scores, animals treated with propolis and controls displayed statistically similar results.In other words, in this analysis the use of red propolis did not inhibit angiogenesis.
The SQAN analysis of the tissues of animals treated with celecoxib revealed a small but non-significant level of inhibition in the vascular area around the implant when compared to controls.Thus, no significant inhibitory effect could be demonstrated for celecoxib.
The results of the SQAN analysis of the tissues of animals treated with gum arabic 1% were similar to the results for red propolis and celecoxib.In short, in this analysis none of the three substances significantly inhibited angiogenesis.
Mechanically extracted from the tree species Acacia senegal, gum arabic is known for its antioxidant properties 21 .Since red propolis was extracted using a solution of gum arabic at 1%, a positive control group treated with gum arabic was included in the study.

Angiogenesis
was evaluated morphometrically, quantitatively and qualitatively, and on slides stained with H&E (scoring system), while the blood vessels were evaluated quantitatively only.In the quantitative analysis, the number of blood vessels was significantly smaller in the group treated preventively with red propolis than in the water control group (p= 0,007), indicating inhibition of angiogenesis.Interestingly, in a study using LDL gene knockout, red propolis was found to inhibit inflammatory angiogenesis related to atherosclerosis 22 .
In the morphometric analysis, propolis did not present statistical significance in the inhibition of angiogenesis.SQAN analysis is after the 11th day of implant when the graft becomes pseudo-encapsulated (fibrosis) thereby limiting intra-sponge angiogenesis.The SQAN evaluates the periesponja angiogenesis.It should be kept in mind that the latter relies on stereoscopic microscopy at 16x magnification, while the histological analysis (H&E) uses optical microscopy at x400 magnification.Indeed, the significant findings of this study were in the microvascular setting.Thus, the vessels quantified in the panoramic view do not represent angiogenic vessels.
Quantitatively, our finding of significant anti-angiogenic activity for red propolis at the microvascular level supports the notion that propolis has immunomodulating properties, stimulating or inhibiting immunological events, as the case may be.
This immunomodulating effect was evidenced in our qualitative analysis: in the first arm of the experiment (prevention + treatment), a significant reduction was observed in the mastocyte levels of the propolis group in relation to the control group (p = 0.013).The same was observed in the quantitative analysis: mastocyte levels in the celecoxib group were significantly reduced compared, respectively, to the water control group (p = 0.002).
Mastocytes are crucial to the development of inflammatory conditions as they alter vascular permeability by releasing vasoactive mediators and recruiting inflammatory cells 23 .The observed reduction in mastocyte levels matches the findings of a study on the effect of propolis on mastocytes in wound healing in the oral cavity of hamsters 24 .The relationship between mastocytes and vasculature and the high mastocyte levels observed in inflammatory conditions are associated with angiogenesis 25 .Thus, based on the results of this study, it seems likely the anti-angiogenic effect of propolis is mediated by mastocytes.
Derived from monocytes, macrophages are the most important phagocytes in chronic inflammation.The persistence of the inflammatory stimulus (e.g.sponge implant) chronifies the inflammatory process.The condition is further consolidated by the presence of angiogenesis and macrophages which in turn release inflammatory cytokines, thereby prolonging it 26 .
In the first arm of the experiment (prevention + treatment), the number of macrophages was significantly greater in tissues treated with red propolis (p = 0.001) in relation to Groups II, III and IV.
The observed increase in macrophages in the propolis group was unsurprising since propolis is known to promote macrophage activation.Phagocytosis plays a major role in organic defense, the production of free radicals, the mediation of inflammatory processes and the release of an array of substances, including enzymes, cytokines and components of the complement cascade.This may be accompanied by increased humoral and cellular immune response.Other researchers have shown that propolis can increase the phagocytic capacity of macrophages 27 .
Giant cells (clusters of macrophages) are a common finding in chronic inflammation, triggered by the presence of agents with low immune activity, such as non-absorbable materials 28 .The water control group and the propolis group differed significantly with regard to the number of giant cells (p≤ 0.001), suggesting red propolis has anti-inflammatory activity.
The number of vessels was significantly smaller in the group treated preventively with celecoxib than in the water control group (p = 0.007), indicating inhibition of angiogenesis, but in the second arm of the experiment (treatment only) the number of vessels was significantly greater than in the control group (p = 0.008).In other words, celecoxib can both inhibit and promote angiogenessis 29 .
In the morphometric analysis, celecoxib reduced angiogenesis very slightly (non-significantly).The results might have been significant, as shown in the literature, if a different dose and time of exposure had been used.Or perhaps two different mechanisms are involved, producing opposite effects (stimulatory and inhibitory).
In the qualitative analysis, the concentration of mastocytes was significantly smaller in the group treated preventively with celecoxib than in the water control group (p = 0.013), a finding subsequently confirmed in the quantitative analysis (p = 0.002).This is in agreement with the literature, which shows celecoxib to have immunomodulating effects 30 .
In the second arm of the experiment (treatment only), the number of giant cells differed significantly between the control group and the celecoxib group (p ≤ 0.001), suggesting celecoxib has anti-inflammatory properties.A similar difference was observed between the control group and the gum arabic group with regard to giant cell count (p ≤ 0.001), suggesting gum arabic has anti-inflammatory properties as well.

■ Conclusion
Red propolis inhibited inflammatory angiogenesis when administered both before and after sponge implantation, and was shown to have immunomodulating effects on inflammatory cells (mastocytes and macrophages) in a new sponge implant hamster model.

Figure 2 -
Figure 2 -Surgical procedure.A: Inoculation of saline to make room for the sponge implant.B: Creation of small opening with colibri forceps.C: Implantation of sponge.

Figure 3 -
Figure 3 -Eleven days after sponge implantation in hamster cheek pouch.Angiogenesis at different levels towards the sponge in the four quadrants.Panoramic stereocopic microphotograph at x16 magnification.

Figure 5 -
Figure 5 -Blood vessel count in the first arm of the experiment (prevention + treatment).*= p< 0.05 compared to Group I.; † † = p< 0.01 compared to Group II.

Table 1 -
Quantitative evaluation of inflammation in the first arm of the experiment (prevention + treatment) expressed as mean values and standard error of the mean.

Table 2 -
Quantitative evaluation of inflammation in the second arm of the experiment (treatment only) expressed as mean values and standard error of the mean.