Abstract
Oral mucositis is a reaction to chemoradiation therapy during cancer treatment. The aim of this study was to investigate the use of amniotic membrane as a biological dressing for oral mucositis lesions in rats. Sixty Wistar rats were divided into three groups ( n = 20): control, 5-fluoruracil (5-FU), 5-fluoruracil + amniotic membrane (5-FU + AM). Each group was subdivided ( n = 5) according to the time interval to sacrifice (3, 7, 14, and 21 days). Histology (haematoxylin–eosin staining) and immunocytochemistry (anti-rat antibodies CD4, CD8, VEGF, and PCNA) were evaluated. Immunocytochemistry results were analyzed using one-way ANOVA and Tukey tests. The amniotic membrane (5FU + AM) played an important role in cell proliferation (PCNA 3 days 27.08 ± 4.65, 7 days 27.90 ± 3.34) and especially in neovascularization (VEGF 3 days 23.00 ± 1.40, 7 days 26.00 ± 0.95) for all time intervals, when compared to 5-FU (PCNA 3 days 23.12 ± 1.61, 7 days 37.21 ± 1.20; VEGF 3 days 17.05 ± 1.51, 7 days 8.45 ± 1.35) and control (PCNA 3 days 29.99 ± 0.92, 7 days 16.33 ± 2.88; VEGF 3 days 13.65 ± 0.55, 7 days 15.70 ± 1.39). It was biocompatible, showing significant differences compared to the other groups in CD4 ( F = 40.72; P = 0.001) and CD8 ( F = 69.99, P = 0.001) staining together, only during the inflammation phase (7 days). Amniotic membrane presented biocompatibility and stimulated cell proliferation and neovascularization, functioning as a promising biological dressing.
Oral mucositis (OM) is a common, symptomatic, and regimen-limiting condition related to head and neck chemoradiation protocols. The clinical appearance of OM varies from redness of the intact mucosa to symptomatic ulcers. Pathologically, OM results in a thin epithelium and ulcers that are thought to be caused by inflammation and depletion of the basal layer of epithelial cells, with subsequent bacterial infection.
Studies have suggested that chemotherapy acts on the basal layer of epithelial cells, inducing a loss of self-renewal capacity. The pathobiology of OM can be divided into five stages: initiation, upregulation and message generation, amplification, ulceration, and healing. According to a model of mucositis, there are two main factors involved in its initiation by radiation or chemotherapy: the death of clonogenic cells and the generation of reactive oxygen species (ROS) by damaged cells. The initiation phase is a gatekeeper phase, and responding to or preventing it can minimize or prevent regimen-related injury.
According to the Multinational Association of Supportive Care in Cancer and the International Society for Oral Oncology (MASCC/ISOO), the therapies available for OM involve nutritional support, pain control, oral hygiene, and the management of dry mouth and bleeding.
The amniotic membrane (AM) is the innermost layer of the placenta. It has a thin epithelium and a non-vascular stroma that contain growth factors, cytokines, and other active substances. Although the mechanisms involved in the biological benefits of the AM are not yet completely understood, the AM is known to provide support for cell growth and adhesion. The AM has been used as a biological dressing in wound repair, vestibuloplasty, and alveoloplasty with favourable results. However, we could identify no studies investigating the benefits of AM in chemotherapy or radiation-induced mucositis.
The aim of the present study was to evaluate the effect of AM on wound repair of 5-fluorouracil (5-FU)-induced mucositis in rats. We used immunocytochemistry to detect proliferating cell nuclear antigen (PCNA), vascular endothelial growth factor (VEGF), CD4, and CD8 to evaluate cell and vascular proliferation. In this study, we found that the use of AM led to important tissue improvements; it acted as a physical barrier and scaffold during wound repair, and accelerated the healing process.
Materials and methods
Animals
All animal procedures were performed in accordance with the recommendations of the Brazilian College of Animal Experimentation (COBEA) and Federal Law Number 11794. The experiment was approved by the local ethics committee on research involving animals. All rats received food ad libitum .
Prior to the experiment, animals were treated with a single dose of the anthelminthic albendazole (Zentel 150 mg/kg; GlaxoSmithKline Brasil, Rio de Janeiro, RJ, Brazil) and multivitamins (Vita Gold; Tortuga Companhia Zootécnica Agrária, Mairinque, SP, Brazil), 40 drops per litre in drinking water for 15 days.
To obtain AM, 15 pregnant rats ( Rattus norvegicus var. albinus ; Wistar) were maintained in cages (five per cage) for a gestation period of 21 days.
For experimental mucositis, 60 male rats ( Rattus norvegicus var. albinus ; Wistar), 90 days old, were divided randomly into three groups: (1) a control group in which an oral ulcer was induced with 50% acetic acid on the buccal fornix on the lower incisors ( n = 20); (2) a 5-FU group, treated with 5-FU to induce OM + 50% acetic acid to induce an ulcer ( n = 20); (3) 5-FU + AM group, treated with 5-FU to induce OM + 50% acetic acid to induce an ulcer and then treated with an AM biological dressing ( n = 20). Each group was subdivided ( n = 5) according to the time interval to sacrifice after the induction of mucositis (3, 7, 14, and 21 days).
Amniotic membrane harvesting
Fertile female and male rats were maintained in cages overnight. In the morning, female rats were submitted to microscopic evaluation of cervical smears. The female rats that presented smears indicating a mating event (15 rats) were housed in separate cages for 21 days. Hysterectomies were then performed to obtain the AM.
On day 21 of gestation, the females were anaesthetized with a mixture of ketamine (Dopalen 5 mg/kg; Agribrands do Brasil Ind. e Com. Ltda, Paulínia, SP, Brazil) and xylazine (Anasedan 10 mg/kg; Agribrands do Brasil Ind. e Com. Ltda), and caesarean surgery was performed. After removing the placenta, the AM was separated aseptically from the chorion and washed with a saline solution and phosphate buffer (pH 7.4). The AM was stored in liquid nitrogen immersed in lactated Ringer’s solution (Equiplex, Aparecida de Goiânia, GO, Brazil) containing 10 mM HEPES (2.38 g/l) and 2 ml of penicillin (10,000 IU/ml) and streptomycin (10 mg/ml) (Cultlab, Campinas, SP, Brazil) until use. For application as a biological dressing, the AM was washed in phosphate buffer (pH 7.4) and incorporated into an orabase ointment.
Induction of oral mucositis
Rats were treated with three intercalated doses of 5-FU 30 mg/kg/day (Fauldfluor 2.5 g/50 ml; Libbs Farmaceutica Ltda, Embu das Artes, SP, Brazil) via the intramuscular route, to induce OM (days 0, 2, and 4).
After a week, clinical signs of ‘chemotherapy’ appeared: hair loss, diarrhoea, weight loss, and signals of OM (intact mucosal redness). We then defined the region to be treated. We placed a 9-mm 2 filter paper immersed in 50% acetic acid (C 2 H 3 COOH; 10 μl) on the fornix region for 60 s, in accordance with Fujisawa et al. During treatment with 50% acetic acid, animals were anaesthetized with ketamine 5 mg/kg and xylazine 10 mg/kg ( Fig. 1 A and B) .
Treatment with amniotic membrane
Once the study area was defined, rats from the 5-FU + AM group were anaesthetized (ketamine 5 mg/kg and xylazine 10 mg/kg) and the areas of mucositis (buccal fornix of the mandibular incisors) were treated with AM incorporated into the orabase ointment ( Fig. 1 C and D); this resulted in a multilayer dressing that was applied only to the area of the ulcer. The inner borders of the lower lip were immediately sutured using an absorbable thread (Vicryl 6–0; Ethicon, Johnson & Johnson, São José dos Campos, SP, Brazil). After application of the AM, all rats had their nails cut and were fed with ground food until the end of the experiment. These modifications helped to keep the AM in position until euthanasia.
The clinical aspect of the area of mucositis was evaluated throughout the entire experiment. Details are reported in the Results section together with data on the other clinical aspects as regard the effects of chemotherapy and membrane treatment. We did not measure the ulcer diameter.
Euthanasia
Five animals in each group were euthanized at time intervals of 3, 7, 14, and 21 days after the study area was delimited with 50% acetic acid (ulcer induction). They were anaesthetized with ketamine 10 mg/kg and xylazine 20 mg/kg and decapitated using a guillotine. The entire lower lip was fixed in 4% paraformaldehyde (pH 7.4) for 48 h, washed, and put into 70% ethanol. It was subsequently embedded in Paraplast (Sigma–Aldrich Co. LLC, St. Louis, MO, USA).
Microscopic analysis (histology)
After embedding in Paraplast, 5-μm histological sections were cut, placed on glass slides, and stained with haematoxylin and eosin (HE). The HE-stained section slides were described for their histological aspects.
Immunocytochemistry
For immunocytochemistry, 3-μm slices were placed on silane-coated slides. The amplification labelled streptavidin–biotin (LSAB) method was used for the following antibodies: anti-CD4 (Code C2255-71; USBiological, Salem, MA, USA) and anti-CD8 (Code C2259-36; USBiological) to evaluate the immunocompatibility of the AM, anti-PCNA (Code M0879; Dako Cytomation Norden A/S, Glostrup, Denmark) to evaluate cell proliferation, and anti-VEGF (Code SC-7269; Santa Cruz Biotechnology, Santa Cruz, CA, USA) to evaluate vascular proliferation. Antigenic retrieval was performed in Tris buffer using a microwave for anti-CD8, anti-PCNA, and anti-VEGF and in Tris–ethylenediaminetetraacetic acid (EDTA) buffer in a Pascal pressurized heat chamber (Dako Cytomation Norden A/S) for anti-CD4.
For immunohistochemical analysis, the area of mucositis was delimited using a drawing tool in the Axiovision 4.7 software program (Carl Zeiss Vision Imaging Systems; Carl Zeiss, Oberkochen, Germany, Germany). For cell counting, images were taken. We used a 63×/0.80 objective lens (Achroplan; Carl Zeiss) and a 10× ocular lens (W-PI; Carl Zeiss) in a light microscope (Axioskop 40; Carl Zeiss) coupled to a digital camera (AxioCam MRc5; Carl Zeiss). Eight random fields were chosen inside the OM area. The immune-positive cells were counted and a mean value was calculated.
Statistics
The quantitative results were submitted to analysis of variance (one-way ANOVA) and Tukey tests (Minitab; Minitab Inc., State College, PA, USA). The level of significance was set at P < 0.05 and results are presented as the mean ± standard deviation (SD).
Results
Clinical evaluation
We evaluated clinical aspects based on the observation of physical and mucosal aspects of the animals.
After 3 days of OM induction, the animals were prostrate and exhibited signs of diarrhoea, per ocular bleeding, low weight, and hair loss. During the next 18 days, the clinical parameters improved gradually, and on day 21, there was very little hair loss.
In the first 3 days after delimitation of the study area, all animals presented ulceration in the studied region, although the membrane group (5-FU + AM) did not exhibit fibrino-purulent membrane adherence of the lower lip to the mandibular gingiva as presented by the 5-FU and control group animals.
After 7 days, the membrane group still had a visible membrane covering the OM, but it appeared to be incorporated into the mucosa with no reddish area surrounding the ulcer as in the other groups. The control group had signs of minor inflammation at 7 days, but there was still a reddish area covered by a fibrino-purulent membrane. In group 5-FU, we noted acute inflammation surrounding the wound area that had necrotic points.
On days 14 and 21, all animals were better, more active, and exhibiting a renewed mucosa of the lower lip. The membrane group showed no sign of any tissue rejection.
It was difficult to keep the AM in place in all rats, even with ground food and cut nails. Therefore, for immunocytochemistry analysis, we considered only five rats per period (20 per group).
Microscopic analysis (histology)
On examination of the events involved in the pathophysiology of OM and histomorphological aspects of the mucosa, we divided the healing process of 5-FU-induced mucositis into four phases: inflammation, cell proliferation, tissue organization, and tissue repair. These phases were correlated with the observation time intervals (3, 7, 14, and 21 days).
Phase I, or the inflammatory phase (3 days), was characterized by extensive oedematous and necrotic areas associated with inflammatory cells, especially polymorphonuclear leukocytes, in all groups. The epithelium adjacent to the AM was atrophic, with hydropic degeneration and spongiosis. In the AM group, we observed adherent AM acting as a protective barrier against bacterial infection on day 3.
Phase II, or the cell proliferation phase (7 days), presented high cell proliferation with fibroblasts and fibrocytes, endothelial cells and neoformed vessels, and epithelial cells re-epithelializing the mucosal surface. The 5-FU + AM group had well-defined oral mucosa and the area of OM was partially re-epithelialized in most of the sections.
Phase III, or the tissue organization phase (14 days), was characterized by high numbers of cells, irregularly disposed collagen fibres, few vessels, and a re-epithelialized oral surface. We observed mononuclear inflammatory cells surrounding the connective tissue. The muscle layer had a normal aspect.
Phase IV, or the tissue repair phase (21 days), was characterized by tissue maturation, with regularly disposed, remodelled collagen fibres and very few blood vessels. In general, the oral mucosa had a normal aspect.
Although it was possible to standardize the periods into phases, it is important to highlight that the AM group showed milder inflammation and better morphology compared with the control group. At the same time, animals in the 5-FU group seemed to experience a delayed healing process due to intense necrosis. Figure 2 shows the histological characteristics of the groups on days 3 and 7.
