Azilsartan Reduced TNF-α and IL-1β Levels, Increased IL-10 Levels and Upregulated VEGF, FGF, KGF, and TGF-α in an Oral Mucositis Model

Oral mucositis (OM) is a common complication of treatments for head and neck cancer, particularly radiotherapy with or without chemotherapy. OM is characterised by oral erythema, ulceration, and pain. The aim of this study was to evaluate the effect of azilsartan (AZT), an angiotensin II receptor antagonist, on 5-fluorouracil (5-FU)-induced oral mucositis (OM) in Syrian hamsters. OM was induced by the intraperitoneal administration of 5-FU on experimental days 1 (60mg/Kg) and 2 (40mg/Kg). Animals were pretreated with oral AZT (1, 5, or 10 mg/kg) or vehicle 30 min before 5-FU injection and daily until day 10. Experimental treatment protocols were approved by the Animal Ethics Committee Use/CEUA (Number 28/2012) of the UFRN. Macroscopic analysis and cheek pouch samples were removed for histopathologic analysis. Myeloperoxidase (MPO), Malonyldialdehyde (MDA), interleukin-1 beta (IL-1β), interleukin-10 (IL-10), and tumour necrosis factor-alpha (TNF-α) were analysed by Enzyme Linked Immuno Sorbent Assay (ELISA). Vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), keratinocyte growth factor (KGF), and transforming growth factor (TGF)-α were measured by immunohistochemistry. Analysis of variance followed by Bonferroni’s test was used to calculate the means of intergroup differences (p ≤ 0.05). Treatment with 1 mg/kg AZT reduced levels MPO (p<0.01), MDA (p<0.5) and histological inflammatory cell infiltration, and increased the presence of granulation tissue. AZT treatment at 1 mg/kg reduced the TNF-α (p<0.05) and IL-1β (p<0.05) levels, increased the cheek pouch levels of IL-10 (p<0.01), and upregulated VEGF, FGF, KGF, and TGF-α. Administration of AZT at higher doses (5 and 10 mg/kg) did not significantly reverse the OM. AZT at a dose of 1 mg/kg prevented the mucosal damage and inflammation associated with 5-FU-induced OM, increasing granulation and tissue repair.


Introduction
Oral mucositis (OM) is a common complication of treatments for head and neck cancer, particularly radiotherapy with or without chemotherapy. OM is characterised by oral erythema, ulceration, and pain. The condition can predispose patients with neutropenia to septicaemia [1,2].
There are five phases in OM pathogenesis. The initiation phase involves the initial injury to cells by radiotherapy and/or chemotherapy. This injury may be induced directly via DNA damage or (more commonly) indirectly via reactive oxygen species. The consequent activation of various enzymes and transcription factors eventually leads to the upregulation of genes coding for inflammatory cytokines, such as tumour necrosis factor (TNF)-α, interleukin (IL)-1β, and IL-6, which target the submucosa and basal epithelium. The resulting inflammation and tissue damage lead to ulceration and subsequent bacterial colonisation, further feeding a vicious cycle of inflammatory cytokine-mediated damage. The final healing phase involves signalling via the extracellular matrix, resulting in epithelial proliferation, epithelialisation, and reestablishment of the mucosal barrier [3].
Different therapeutic approaches for cancer treatment-induced OM have been reported, including intensive oral hygiene care [4], antimicrobial agents [5], anti-inflammatory agents [6], cytokines, growth factors [7], and topical agents, such as laser therapy [8,9] or medicinal plants [10,11]. To date, however, no single intervention has been able to prevent and treat OM; combinations of treatments acting on different phases of OM must be used. Moreover, it is still unclear which strategies reduce OM, as no evidence supports any treatment as having superior efficiency and efficacy [12].
The mucosal immune response, including tolerogenic prevention of inflammatory reactions and the secretion of antigen-nonspecific suppressor cytokines (e.g., IL-10) [13]. Various signalling pathways have the ability to increase keratinocyte migration and proliferation. These pathways include epidermal growth factor (EGF) family members, such as transforming growth factor-alpha (TGF-α) [14]. In addition, other growth factors are involved in granulation tissue formation, such as vascular endothelial growth factor (VEGF) [15] and fibroblast growth factor (FGF)-2 [16].
Our group has studied angiotensin receptor blockers (ARBs) because these drugs have been shown to interfere with pathways that mediate inflammation in an experimental animal model [17][18][19][20]. The purpose of the study reported in this paper was to investigate the anti-inflammatory activity of azilsartan (AZT) in an experimental model of OM.

Animals
Male adult Syrian hamsters weighing 150 to 200 g were obtained from the vivarium of the Department of Biophysics and Pharmacology of the Federal University of Rio Grande Norte (UFRN) and Potiguar University (UNP), Brazil. Experimental and animal treatment protocols were approved by the Animal Ethics Committee/CEUA of the UFRN (no. 28/2012). All animals were housed in an animal room under standard laboratory conditions, at 22 ± 2°C with a 12-h/12-h light/dark cycle. Animals were fed pelleted food and water ad libitum. They were acclimatised for 7 days and fasted for 12 h before the experiments. All efforts were made to minimize the number of animals used and their suffering degree.

Induction of experimental OM
Hamsters were divided into six groups (n = 6/group). On days 1 and 2 of the experiment, 5-FU was administered intraperitoneally (i.p.) at 60 and 40 mg/kg, respectively, in accordance with an experimental OM model described by Medeiros et al. [21]. On day 4, animals were anaesthetised with 2% xylazine (10 mg/kg) and 10% ketamine (70 mg/kg). The cheek pouch mucosa was irritated by superficial scratching to potentiate OM. Irritation was performed twice, by dragging the tip of an 18-gauge needle across the everted cheek pouch in a linear manner, to reproduce the clinical signs of chronic irritation and to create a condition favourable for OM by mechanical trauma (MT). Animals were sacrificiedon day 10 by an overdose of anaesthesia with 2% thiopental (80 mg/kg, i.p.). After death a cardiac puncture was performed.

Experimental groups
Hamsters in the three experimental OM groups were treated with oral AZT (EDARBI, USA) at a dosage of 1, 5, or 10 mg/kg (groups AZT1/5-FU, AZT5/5-FU, and AZT10/5-FU, respectively). Oral AZT was administered 30 min before 5-FU injection, and then daily until the hamsters were sacrificied on day 10. Hamsters in the three control groups (Normal, MT, and 5-FU/ Saline groups) were treated with saline instead of AZT, with or without 5-FU and/or MT treatment. The Normal group was not treated with 5-FU or MT; the MT group was not treated with 5-FU, but received MT on day 4; and the 5-FU/Saline group received i.p. 5-FU on days 1 and 2 (60 and 40 mg/kg, respectively), and MT of the cheek pouches on day 4.

Macroscopic analysis of cheek pouch
Photographs were used for scoring lesions. For macroscopic analysis, inflammatory aspects, such as erythema, erosion, vasodilatation, epithelial ulcerations, and abscesses, were evaluated in a single-blind fashion and graded as follows [21]: Score 0, completely healthy cheek pouch with no erosion or vasodilatation; Score 1, presence of erythema, but no evidence of erosion in the cheek pouch; Score 2, severe erythema, vasodilation, and surface erosion; Score 3, formation of ulcers in one or more faces of the mucosa, but not affecting more than 25% of the surface area of the cheek pouch, as well as severe erythema and vasodilatation; Score 4, cumulative formation of ulcers of about 50% of the surface area of the cheek pouch; and Score 5, virtually complete ulceration of the cheek pouch mucosa, in which the fibrosis makes oral mucosa exposure difficult.

Histopathologic analysis
Specimens were fixed in 10% neutral buffered formalin, dehydrated, and embedded in paraffin. Sections measuring 5-μm thick were obtained for haematoxylin-eosin staining and were examined by light microscopy (×40 magnification). Inflammatory cell infiltration, vasodilatation, presence of haemorrhagic areas, oedema, ulcerations, and abscesses were determined in a single-blinded fashion and graded as follows [21]: Score 1, normal epithelium and connective tissue without vasodilatation, absence of or discreet cellular infiltration, and absence of haemorrhagic areas, ulcerations, or abscesses; Score 2, discreet vasodilatation or re-epithelisation areas, discreet inflammatory infiltration with mononuclear prevalence, and absence of haemorrhagic areas, oedema, ulcerations, or abscesses; Score 3, moderate vasodilatation, areas of hydropic epithelial degeneration, inflammatory infiltration with neutrophil prevalence, presence of haemorrhagic areas, oedema, and eventual ulceration, and absence of abscesses; and Score 4, severe vasodilatation, and inflammatory infiltration with neutrophil.

Myeloperoxidase (MPO) assay
The extent of neutrophil accumulation in Check pouch tissue samples was measured by assaying MPO activity. Cheek pouch mucosa (6 samples per group) was harvested as described above and stored at -70°C until required for assay. After homogenisation and centrifugation (2000 × g for 20 min), the MPO activity in these samples (in units of MPO/mg tissue) was determined by a previously described colorimetric method [22].

Malonyldialdehyde (MDA) assay
Malonyldialdehyde (MDA) is an end product of lipid peroxidation. To quantify the increase in free radicals in check pouch tissue sample, MDA content was measured via the assay described by Esterbauer and Cheeseman [23]. Check pouch tissue samples (6 samples per group) were suspended in buffer Tris HCl 1:5 (w/v) and minced with scissors for 15 sec on an ice-cold plate. The resulting suspension was homogenised for 2 min with an automatic Potter homogenizer and centrifuged at 2500 × g at 4°C for 10 min. The supernatants were assayed to determine MDA content. The results are expressed as nanomoles of MDA per gram of tissue.

Biochemical, white blood cell (WBC) and bacteraemia analyses
On day 10, animals were killed by an overdose of anaesthesia with 2% thiopental (80 mg/kg, i.p.). Blood samples were collected by heart puncture for subsequent biochemical, WBC, and bacteraemia analyses.
For biochemical analyses, serum was obtained by centrifuging total blood without anticoagulants at 2,500 rpm for 15 min. Serum levels of alanine amino transferase (ALT), aspartate amino transferase (AST), creatinine, and urea were determined by using standardised diagnostic kits (LABTEST) and spectrophotometry.
For WBC analysis, 20 μl of total blood were added to 380 μl of Turk solution. Total and differential counts of leukocytes (Number of leukocytes/mm³) were determined by standard manual procedures using light microscopy [26]. For bacteraemia analysis, 10 μl of total blood were diluted tenfold in brain-heart infusion media (BHI). Bacterial growth was analysed after 24 to 48 h at 37°C by visual analysis of the turbidity of the culture medium. A turbid medium indicates bacteraemia (+), and a non-turbid medium suggests absence of bacteria in the blood. Organisms that grew on BHI were streaked for isolation on 5% sheep blood agar plates, MacConkey agar plates, and mannitol-salt-agar plates. Organisms that grew were identified by standard methods, including Gram stain morphology, DNAse, catalase, coagulase, motility, and sugar metabolism methods.

Statistical analysis
Data are presented as the mean ± standard error of the mean or as the median (range), when appropriate. Analysis of variance (ANOVA) followed by Bonferroni's test was used to calculate the means. The Kruskal-Wallis test followed by Dunn's test was used to compare medians (GraphPad Prism 5.0 Software, La Jolla, CA, USA). A p-value < 0.05 indicated a statistically significant difference. Analysis of MPO, MDA, IL-1β, IL-10 and TNF-α assay were performed in triplicate.

Effects of AZT on OM
Treatment of hamsters with 5-FU followed by MT of the cheek pouch (5-FU/Saline group) caused lesions within 10 days, as evidenced by erythema, hyperaemia, haemorrhagic areas, and extensive ulceration. Clinical dates were confirmed by histopathologic analysis, which revealed severe vascular ingurgitation and vasodilatation, accentuated inflammatory infiltration with the prevalence of neutrophils, oedema, and extensive ulceration and abscesses (Figs. 1 and 2, Table 1). On day 10, the median (range) for the macroscopic score was 5 (4.5-5) and for the histologic score was 4 (4-4).
Compared to the 5-FU/Saline group, the MT group evidenced less severe clinical lesions, with discrete erythema but no ulcers. Clinical dates were confirmed by histopathologic analysis, which showed reduced levels of cellular inflammation, oedema, and haemorrhage, although granulation tissue was still observed (Figs. 1 and 2, Table 1). On day 10, the median (range) for the macroscopic score was 2 (1.5-3) and for the histologic score was 2 (2-2) ( # p< 0.5 and ## p < 0.01 vs. 5-FU/Saline group).

MPO and MDA levels
The MPO activity was measured in the cheek pouch as an indicator of neutrophil infiltration. Levels of MPO in the Normal and AZT1/5-FU groups were decreased (p <0.01 and p < 0.05, respectively) compared to levels in the 5-FU/Saline group (Fig. 4A). Levels of MPO in the MT group were increased (p < 0.05) compared to levels in the Normal group (Fig. 4A). Levels of MDA in the Normal and AZT1/5-FU groups were decreased (p < 0.05) compared to levels in the 5-FU/Saline group (Fig. 4B).

Effects of AZT on systemic changes: Biochemical, WBC and bacteraemia analyses
To determine the effect of AZT on systemic changes, biochemical, WBC, and bacteraemia analyses were performed. Levels of hepatic enzymes (ALT and AST) and markers of renal function (urea and creatinine) did not show a significant difference between the groups ( Table 2). On day 10, animals in the 5-FU/Saline group presented significantly increased leukocyte levels compared to the Normal control group (p < 0.001; Table 1). Table 1 shows the results of the bacteraemia analysis for animals in the various groups on day 10. Animals in the Normal control group did not exhibit bacteraemia (-). The 5-FU/Saline group had the highest proportion of animals with bacteraemia (Staphylococcus xylosus was found in 4/6 samples). The AZT1/5-FU group did not present any sample with bacteraemia. Bacteraemia (+) samples were found in the AZT5/5-FU and AZT10/5-FU groups, and different bacterial species were identified.

Discussion
AZT is an ARB that was approved for use in 2011. At its maximal dose, AZT has superior efficacy to both olmesartan and valsartan at their maximal, approved doses, without increasing adverse events [27]. Studies have shown that AZT exhibits anti-inflammatory activity [18], confirming reports regarding the anti-inflammatory activities of ARBs [17,28,29].
The 5-FU/Saline group in this study showed a median macroscopic score of 5, with nearly complete ulceration of the cheek pouch mucosa and fibrosis that made exposure of the oral mucosa difficult. These findings differed significantly from those of the AZT1/5-FU group (median macroscopic score = 2), which evidenced erythema and vasodilation but no ulceration. The results were confirmed by histopathologic analysis, which revealed reduced infiltration of inflammatory cells and oedema. Treatment of animals with 5 mg/kg AZT (AZT5/5-FU group) or 10 mg/kg AZT (AZT10/5-FU group) did not reduce the 5-FU-induced lesions in the cheek pouch, as indicated by the macroscopic and histologic scores.
The histopathologic results confirmed the presence of granulation tissue in the AZT1/5-FU group. The macrophages in granulation tissue control the cellularity of wounds by inducing apoptosis and phagocytising various wound cells. Macrophages target neutrophils during the inflammatory phase of repair, as well as fibroblasts and endothelial cells (ECs) during the resolution of this phase [30]. Neutrophils become less prevalent as the wound matures, and macrophages emerge as the predominant inflammatory cell. Neutrophils undergo apoptosis in the wound and are recognised and ingested by macrophages [31]. We found that the AZT1/5-FU group showed a significant reduction in the number of neutrophils, as confirmed by the significant reduction of level MPO.
Wound healing is an evolutionarily conserved, complex, multicellular process that, in skin, aims at barrier restoration. This process involves the coordinated efforts of several cell types, including keratinocytes, fibroblasts, ECs, macrophages, and platelets. The migration, infiltration, proliferation, and differentiation of these cells culminate in inflammation, new tissue formation, and, ultimately, wound closure. This complex process is executed and regulated by an equally complex signalling network involving numerous growth factors, cytokines, and chemokines [32]. OM is characterised by an intense inflammatory reaction, caused by the effects of chemotherapeutic agents on the mucosa lamina propria cells, which results in the production of proinflammatory cytokines (e.g., IL-1β, IL-6, and TNFα) [33]. We demonstrated that oral treatment with 1 mg/kg AZT significantly (P < 0.05) reduced TNF-α and IL-1β levels in an experimental model of OM in hamsters. In agreement with our data, various articles have shown that AZT decreases IL-1β production [18].
IL-10 is the important cytokine with anti-inflammatory properties. In monocytes and macrophages, IL-10 diminishes the production of inflammatory mediators and inhibits antigen presentation, although it enhances their uptake of antigens [34]. In our study, the AZT1/5-FU group showed significantly increased levels of IL-10 and TGF-α compared to 5FU/saline group. Increased tissue levels of TGF-α can indicate macrophage migration to granulation tissue [35]. Macrophages initiate granulation tissue development and release various growth factors, including FGF, EGF, TGF, and platelet-derived growth factor (PDGF) [32]. Macrophages provide an ongoing source of cytokines to modulate inflammatory cell adhesion, cell migration, and fibroblast proliferation. Within hours of injury, re-epithelialisation is initiated, and    the release of EGF, TGF-α, and FGF acts to stimulate epithelial cell migration and proliferation [36]. TGF-α has the ability to increase keratinocyte migration and proliferation [14]. FGFs are produced by keratinocytes, fibroblasts, ECs, smooth muscle cells, chondrocytes, and mast cells. FGF-2 or FGF is increased in the acute wound and plays a role in granulation tissue remodelling [16]. FGF-7 or KGF-1, which may be important during the late stages of neovascularisation, upregulates VEGF [37], stimulates the proliferation and migration of keratinocytes, and plays an important role in re-epithelialisation [36]. In the present investigation, the AZT1/5-FU group evidenced increased tissue expression of growth factors (i.e., VEGF, FGF, KGF, and TGF-α) involved in the healing process and, especially, in the activation of keratinocytes and neovascularisation with collagen scaffold formation.
The use of AZT was tested in three doses; however, only the 1 mg/kg dose (AZT1/5-FU group) showed efficacy. At this dose, the inflammatory process was reduced, with a significant reduction in MPO levels, notable presence of granulation tissue, and increased levels of IL-10, VEGF, FGF, KGF, and TGF-α. The use of AZT showed a dose-independent effect, with preference for a lower dose in order to achieve the best clinical results by accelerating the healing process. The effect of AZT on systemic changes was tested through biochemical, WBC, and bacteraemia analyses. AZT proved to be a very safe drug; even associated with 5-FU (used to induce OM), AZT did not cause changes in the biochemical parameters. With the WBC and bacteraemia analyses, bacteraemia was found in four of six samples in the 5-FU/Saline group, which also showed significant leukocytosis. Although 5-FU treatment is expected to result in leukopenia, the observed leukocytosis was probably in response to bacterial lipopolysaccharide or host proinflammatory cytokines as a result of secondary infection. The AZT1/5-FU group no showed the presence of bacteria in samples. These findings strengthen the evidence for a protective role of AZT in 5-FU-induced OM.
A limitation of the present study is the use of an experimental model of oral mucositis that was established in hamsters. Given that major genetic differences exist between rodents and humans, the present results need to be confirmed in primates prior to the clinical application of azilsartan to humans.
In the present study, we show that azilsartan mediates a protective effect towards the inflammatory parameters that characterize 5-FU-induced OM. Many of the functional effects demonstrated are dependent on two key factors related to azilsartan as a molecule: 1) it has a high affinity and slow dissociation rate from AT1R, and 2) it exhibits inverse agonistic properties. These properties suggest that azilsartan represents a unique therapeutic agent for the potential treatment of a wide range of angiotensin II-dependent cardiometabolic diseases. These diseases include cardiac hypertrophy, unstable atherosclerotic plaques, cardiac fibrosis, insulin resistance, renoprotection and antiinflammatory effects [38,39]. In addition, the results obtained from a number of clinical studies indicate that angiotensin II (AII) receptor type 1 (ATR1)blocking drugs (ARBs) mediate anti-inflammatory effects [40][41][42].
In conclusion, the findings in this study demonstrate that AZT accelerates the healing process in an experimental model of 5-FU-induced OM, stimulating granulation tissue formation, fibroblast and keratinocyte migration, and collagen deposition. These data were confirmed by the increased levels of TGF-α, FGF, and KGF, as well as the upregulation of VEGF, which has been shown to promote angiogenesis in scar tissue. Consequently, use of AZT at a dose of 1 mg/kg accelerates the healing process by activating growth factors involved in angiogenesis and re-epithelialisation, in addition to increasing levels of the anti-inflammatory cytokine IL-10. Application of azilsartan-associated chemotherapy to the model of oral mucositis that was established did not induce an anti-inflammatory effect. It remains to be determined whether hypertensive cancer patients will benefit from an anti-inflammatory effect mediated by azilsartan-based chemotherapy