S. maritima L. (S. maritima or Sm) is a common shrub that grows widely in the northeast region of Brazil, near the seacoasts. It belongs to Plantaginaceae family, tribe Gratioleae Benth (Albach et al., 2005). Dr. F. S. Cavalcanti and Prof. E. P. Nunes identified the plant and a voucher specimen (# 38483) was registered at Prisco Bezerra Herbarium, Federal University of Ceará, Fortaleza, Brazil. For this study, the extract was obtained from fresh leaves of Sm collected during the flowering stage along Fleixeiras Beach, Ceará, Brazil (da Silva et al., 2014).

The experimental procedures and treatments performed on animals were approved by the Animal Research Ethics Committee of Federal University of Ceará (Permit number: 08/2013) in accordance with the guidelines from the Brazilian Society of Laboratory Animal Science - SBCAL/COBEA). Female Wistar rats, 10-weeks-old and weighting 200 ± 20 g were housed in appropriated cages of polypropylene and maintained on a 12 h-12 h light-dark cycles with a constant room temperature of 25°C and received water and food ad libitum.

For the initial experiment, 30 animals were randomly divided into five specific groups (n = 6 in each group): unchallenged group (naive), three experimental periodontitis-challenged groups (EP) receiving oral gavage (per os) in different concentrations of S. maritima (0.2, 1, or 5 mg/kg) or 0.9% saline solution + DMSO (vehicle). Experimental Periodonttis (EP) was performed as previously described (Bezerra et al., 2002). Briefly, the animals were anesthetized with ketamine and xylazine (90:10 mg/kg, i.p.) and a sterilized nylon suture thread (3.0 Nylpoint, Ceará, Brazil) was placed around the second left maxillary molar. To facilitate suture placement, a guide through the medial and distal interproximal spaces was made using a 5.0 suture needle (Point Suture, Ceará, Brazil). Groups received vehicle and S. maritima (per os) 1 h before periodontitis induction and once daily, during 11 days. After this time, blood samples were obtained and an overdose of ketamine and xylazine (300:30 mg/kg; i.p.) were used to euthanize the animals. The maxillae were removed to analyze the total bone resorption area. Hystopathological analyses was performed with another animals following the same above designed groups (n = 30).

Sm at a dose of 5 mg/kg was found to be the most effective dose against alveolar bone loss (ABL), and therefore this dose was selected for the experiments. Subsequent series of experiments were conducted with the following groups: non-ligated (naive - animals not subjected to EP); ligature only (vehicle), and ligature plus treatment with S. maritima 5 mg/kg diluted in 0.9% saline solution + DMSO (S. maritima 5) to quantification of gingival levels of cytokines TNF-α, IL-1β, CINC-1 and IL-10 (n = 18), antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) (n = 18) and qRT-PCR analysis of TNF-α, IL-1β, RANK, and iNOS (n = 18). All these assays were performed using the gingival tissues from surrounding maxillary left molars (the half of the maxillae with ligature).

After the period of EP the maxillae were removed, divided in half and fixed in buffered formalin (10%) for 24 h. Following, the maxillae were defleshed and kept in 8% sodium hypochlorite for 4 h (Pimentel et al., 2012). After that, the specimens were washed in running water and stained with methylene blue (1%) to differentiate bone from teeth. Then, hemi-maxillae were fixed in a piece of wax with their occlusal planes parallel to the ground and long axes perpendicular to the camera and photographed with a 6.1-megapixel digital camera (Canon® 60D). ImageJ® software (National Institute of Health, Bethesda, MD, USA) were used for measurement of ABL, as described previously (Kuhr et al., 2004). The buccal area (mm²) corresponding to the exposed roots and coronary surface of the molars was calculated and obtained and subtracted from the correspondent area (mm²) of the normal right hemi-maxillae.

After definition of the most effective dose, one sample of each group was scanned using a high resolution microcomputed tomography (micro-CT) system (SkyScan 1174; Bruker-microCT, Kontich, Belgic, 50 kV and 800 μA) with a 0.5 mm aluminium filter and 15% beam hardening correction and ring artifacts reduction. We used 180 degrees of rotation and exposure range of 1 degree, time of scanning 38 min. Each specimen was scanned with acquisition of images every 0.7°, filed in TIFF format, with resolution of 19.7 μm and saved on a hard disk. After reconstruction of the images (NRecon v1.6.9; Bruker-microCT) 3D models were created with help of CTAn software v.1.12 program (Bruker-microCT) in accordance with the recommended guidelines.

The maxillae with ligature were fixed in 10% neutral-buffer formalin (10%) and demineralized in formic acid (10%). The specimens went through the process of dehydration, paraffin embedment, and after were sectioned in a mesio-distal plane for hematoxylin and eosin (HE) staining. The area between the first and second molars where ligatures had been placed was sectioned in 4 μm thickness sections and evaluated under light microscopy considering the inflammatoryparameters such as cell influx and alveolar bone and cementum integrity. Previously standardized scores ranging from 0 to 3 were used (Lima et al., 2000; Leitão et al., 2005). Score 0 indicates absence of or only discrete presence of inflammatory cell infiltration cellular infiltration (restricted to the marginal gingiva, preserved alveolar process and cementum; score 1 represents moderate inflammatory cellular infiltration all over the insert gingiva, some minor alveolar process resorption and intact cementum; score 2 represents accentuated cellular infiltration in both gingiva and periodontal ligament, accentuated degradation of the alveolar process, and partial destruction of cementum; score 3 indicates accentuated cellular infiltration with complete resorption of the alveolar process and intense destruction of cementum. Two examiners who were masked to the identity of samples performed histologic evaluation.

The plasma concentration of bone alkaline phosphatase was obtained from blood samples collected from the retro-orbital plexus on the 11th day using the thermo-activation method, as previously described (Whitby and Moss, 1975). The samples were heated up to 56°C for 10 min. Serum levels of BALP were calculated by subtracting the concentration of the heated alkaline phosphatase in serum from the concentration of the total alkaline phosphatase (TALP) in serum. The analysis was performed according to the manufacturer's instructions (Labest, Lagoa Santa, MG, Brazil).

The stored gingival tissues were used to determine the concentrations of TNF-α, IL-1β, CINC-1 and IL-10 using specific commercially available kits (DuoSet Elisa kit, R&D Systems Inc., MN, USA). The inflammatory mediators' levels were determined by enzyme-linked immunosorbent assay (ELISA) using respective standard curves. Results were shown as picogram/ml (pg/ml). All kits were used according manufacturer's instructions.

The gingivae removed on 11th day were used to evaluate the effect of S. maritima 5 mg/kg on the oxidative stress. The SOD activity was assayed according the protocol previously described (Beauchamp and Fridovich, 1971). In a dark room, the gingival samples were homogenized in 20 μl of ice-cold phosphate buffer at 15,000 G for 20 min. The supernatants were mixed with a solution comprised of phosphate buffer (50 nM), EDTA (100 nM) and L-methionine (19.5 mM) in a pH of 7.8. Then, 150 ml of a solution of riboflavin (10 nM) and nitro NBT (750 nM) was added and the mixture was exposed to light (20 W) for 15 min. The absorbance of the samples was measured at 560 nm. The results are expressed as grams of SOD per ml.

The measurement of O₂ production rate and H₂O in proportion of H₂O₂ was calculated to obtain CAT activity. Briefly, 20 μl of gingival homogenate was mixed with a solution comprised of 3% H₂O₂ and Tris-HCl EDTA buffer (5 nM, pH 8.0). The absorbance was measure immediately and 6 min after preparing the samples at a 230 nm wavelength (Maehly and Chance, 1954).

Total RNA was extracted from gingivae tissues using the TRIzol reagent (Invitrogen, São Paulo, Brazil). The RNA concentration was estimated by reading the absorbance at 260 nm and was checked for purity at 280 nm in a spectrophotometer (Amersham Biosciences, Cambridge, England). For each sample, the RNA concentrations used to synthesize cDNA were adjusted to 1,000 ng/mL. Before the reverse transcription reaction, samples of RNA were incubated for 5 min at 70°C and then cooled in ice. The reverse transcription was performed in a total volume of 20 μL composed of 10 μL of sample RNA, 4 μL reverse transcriptase buffer (Invitrogen), 8 units RNAsin, 150 units of reverse transcriptase Superscript III, 0.036U random primers, 10 mM DTT and 0.5 mM of each dNTP (Invitrogen). The mixture was incubated at 42°C for 1 h, subsequently at 80°C for 5 min and finally stored at −20°C. The negative control was prepared under the same conditions but without addition of reverse transcriptase.

Quantitative real-time polymerase chain reaction (qRT-PCR) was performed in triplicate to determine the gingival levels of mRNA for TNF-α, IL1-β, RANK and iNOS. Each reaction contained 10 μL of SYBR® Green Master Mix (Applied Biosystems, Warrington, UK), 7,3 μL of ultra-pure water, 1 μL of cDNA and 0.5 μM of each primer and was performed in StepOne Real-Time PCR (Applied Biosystems, Warrington, UK) thermocycler. The thermal cycling profile for the first round of qRT-PCR was initial denaturation and activation of the polymerase for 10 min at 95°C, followed by 40 cycles of 15 s at 95°C, 30 s at 58°C, and 30 s at 72°C. The final extension was for 10 min at 72°C. The primers were designed by using the PrimerQuest® Tool https://www.idtdna.com/Primerquest/Home/Index. The primers used in this study are shown in Table 1. The glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as endogenous control for normalization of messenger RNA expression. The specificity of each primer pair was confirmed by melting curve analysis of qRT-PCR products. Relative quantifications of mRNA were carried out using the comparative threshold (CT) cycle method. The delta-delta-Ct method was used to transform the Ct values into normalized relative expression levels (Livak and Schmittgen, 2001).

Variation of body mass, organ weight alteration and biochemical and histopathological parameters were evaluated in animals treated daily during 11 consecutive days with a single dose of S. maritima 5 mg/kg and saline + DMSO. Peripheral blood samples were collected in order to obtain biochemical analysis of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels and creatinine (Labtest, Lagoa Santa, MG, Brazil).

After sacrificing the animals, the liver, kidney and heart were removed and weighed. The stomach was also removed for histological analysis. The organs were fixed with formalin and then dehydrated with increasing concentrations of ethanol and embedded in paraffin. The blocs were sliced in 5 μm thick sections, stained with hematoxylin-eosin (HE) and observed at light microscope (Leica DM 2000, Wetzlar, Germany).

Shapiro-Wilk normality test was performed to analyze the data. Results are presented as means ± standard error (SEM) or as medians when appropriate. ANOVA followed by Tukey test or Games-Howell test were used to compare means and Kruskal–Wallis and Dunn tests were used to compare medians. P < 0.05 was considered significant. Analyses were performed using IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY or GraphPad Prism 6 software, San Diego, CA, USA.

No significant ABL was observed in the naive group (Figures 1a,b). Data indicated that S. maritima 5 mg/kg was the most effective dose protecting against ABL. Gavage (per os) administration of S. maritima 5 mg/kg 1 h before the placement of the ligature and once daily for 11 days resulted in a significant (P < 0.05) inhibition of ABL (2.37 ± 0.53) (Figures 1a,bG,H), compared to the group that received the vehicle only (4.47 ± 0.22) (Figures 1a,bD,E).

Histopathologic analysis of animals subjected to EP (vehicle group) demonstrated accentuated inflammatory cell infiltration, breakdown of alveolar bone, collagen fiber derangement within the periodontal ligament, and resorption of cementum, receiving a median score of 3 (range 2 to 3) (Table 2). The periodontium of animals treated with S. maritima 5 mg/kg showed preservation of the alveolar process and cementum, reduction of the inflammatory cell infiltration, and partial preservation of collagen fibers of the periodontal ligament (Figures 1bC,F,I), receiving a median score of 1 (range 1 to 2) (Table 2). These values were statistically different (P < 0.05), compared with the vehicle group.

Experimental periodontitis decreased BALP serum levels and the treatment with S. maritima 5 mg/kg resulted in a significant increase in the BALP serum levels (P < 0.05), when compared with the vehicle group (Figure 2).

Periodontitis challenge was associated with significant (P < 0.05) increase of pro-inflammatory cytokines (TNF-α, IL-1β, and CINC-1) in gingival tissue; at the same time it was observed a significant (P < 0.05) decrease in gingival levels of IL-10, an anti-inflammatory cytokine, when compared to naive group. Treatment with Sm 5 mg/kg significantly (P < 0.05) decreased TNF-α and CINC-1 but no significant difference was observed in IL-1β gingival levels; also, S. maritima 5 mg/kg increased IL-10 levels, when compared to vehicle treated group (Figures 3A–D).

In the present study periodontitis induction resulted in a significant (P < 0.05) reduction of SOD and CAT levels in gingival tissue, when compared with naive group. S. maritima 5 mg/kg significantly (P < 0.05) increased both markers for oxidative stress in gingival tissue, when compared with vehicle group (Figures 4A,B).

The vehicle groups showed significant increase in TNF-α (Figure 5A), IL-1β (Figure 5B), RANK (Figure 5C) and iNOS (Figure 5D) mRNA expression, when compared with the naive control group. Administration of S. maritima 5 mg/kg significantly reduced mRNA expression in all parameters evaluated compared to vehicle group (Figure 5).

No signs of systemic illness, adverse pharmacological events or changes in behavior were observed throughout the experimental period. S. maritima 5 mg/kg or vehicle did not affect the animals' body mass or the wet weight of the livers, kidneys or hearts compared to the naive control throughout the study period. Gross necropsy findings did not show any abnormalities. A single-dose of S. maritima (5 mg/kg) or vehicle (saline + DMSO) over 11 consecutive days had no significant histological alterations in the hepatic, renal parenchyma, and cardiac tissue. The values obtained for ALT/AST and creatinine did not differ from the control group.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.