Detection of dentin antigenic fractions by salivary immunoglobulin G in patients undergoing orthodontic treatment

This study aimed to partially characterize the antigenic fractions of human dentin extract, and to evaluate the anti-dentin antibodies levels in patients undergoing orthodontic treatment. Two dentin extract fractions (FI and FIIa) were used to analyze the saliva samples from three groups of subjects. One group had 13 subjects who presented radiographic signs of mild to moderate root resorption 12 months after starting orthodontic treatment (T12); saliva samples taken prior to treatment served as controls (T0). The other groups had 10 subjects with radiographic signs of mild to severe root resorption up to 48 months after orthodontic therapy (PT); 10 individuals not undergoing orthodontic treatment were selected as controls (ST). Western blot analysis revealed the presence of dentinal fractions of approximately 35 to 70 kDa in (T12) and (PT) but not in the control samples. In immunoenzymatic assays, the anti-FI salivary IgG levels were significantly higher in T12 and PT groups than in the controls (P< 0.05). Our results demonstrate that dentinal fractions may be presented to the immune system and detected by salivary IgG. The levels of anti-dentin antibodies may remain elevated even years after finishing of orthodontic therapy.


INTRODUCTION
External apical root resorption (EARR) is an immunopathological manifestation of inflammatory origin that presents high prevalence (39 to 99%) in patients undergoing orthodontic treatment (Apajalahti and Peltola, 2007;Dudic et al., 2008).This process has consequence the irreversible shortening from apex root.Although in most cases the root resorption is mild and minor clinical importance, moderate (lesser than 2 mm) to severe *Corresponding author.E-mail: itanoeiko@hotmail.com.
Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License (greater than 2 mm) root resorption has been reported to occur at a frequency of 10 to 20% (Brin et al., 2003;Marques et al., 2010), representing a serious problem for orthodontists.The combination of several mechanical and biological risk factors, reported in the literature as: extensive tooth movement, movement type, orthodontic force magnitude, duration and type of force, individual susceptibility, anatomical factors, genetic influence, systemic factors (Sameshima and Sinclair, 2001;Al-Qawasmi et al., 2003;Segal et al., 2004;Artun et al., 2009), may predispose EARR.However, the destruction of the cementoblastic layer, that protect the tooth root, is the local factor required for the initiation of (Consolaro, 2005).Although the mechanisms of induction and regulation of this pathology are not yet completely understood, some results have indicated the potential of dentin to induce inflammatory and immune response in the root resorption process.Dentin contains numerous non-collagenous proteins (NCP) and signaling molecules usually hidden, trapped in their mineralized matrix.These dentin constituents, that seem to be multifunctional, can be exposed and released in consequence of local injury as from orthodontic movement (Butler et al, 1998;Silva et al., 2004).
The exposure of the dentin in vivo is, fre-quently, accompanied by inflammatory cell infiltration on periodontal tissues and by signals and cytokines production, for activation and differentiation of clasts, responsible by resorption of root surface (Bassaran et al., 2006).It is possible that clasts dissolve the mineralized matrix, endocytose, transport and continuously release dentin molecules in the tooth/periodontium microenvironment during the resorption process (Nesbitt 1997).
These components may act as chemotactic factor (Ogata et al., 1997) and activation factor of inflammatory cells (Lara et al., 2003) and resorptive cells (Nakagawa et al., 2000).Additionally, previous studies already have established the immunogenic potential of dentin and confirmed the presence of anti-dentin antibodies (autoantibodies), by using a model of root resorption in dogs and mice (King and Courts, 1989;NG et al., 1990;Wheeler and Stroup, 1993).Subsequent studies investigated the cellular response in root resorption process (Lara et al., 2003) and the presence of serum IgG and IgM levels was shown in patients with apical root resorption resulting in dental trauma (Hidalgo et al., 2005).And more recently, Ramos et al. (2011), used saliva samples to determine the levels of secretory IgA anti-human dentin extract.Because there are few studies in literature about the antigenic components of the dentin in EARR, the present research, described here, aimed to partially characterize human dentin antigenic fractions detected through the saliva and additionally to evaluate the anti-dentin antibodies profile in patients undergoing orthodontic treatment.The subjects in this study had mild to severe root resorption and were evaluated during or post-treatment.This analysis could help in further studies to obtain early diagnosis, development of preventive methods and understanding of the immunopathological mechanisms involved in root resorption.

METHODOLOGY
All procedures involving human were performed after consent given by the subjects or by a parent/legal guardian and were approved by the Human Ethics Committee at State University of Londrina.

Subjects' selection
This study had three different groups (A, B, C).The A study group had 13 subjects (mean age 16 ± 4 years old) with nine females and four males.This group was analyzed over two phases: 12 months following the beginning of the fixed appliance therapy, at which time the patients presented with radiographic signs of mild to moderate root resorption (T12); these subjects taken prior to treatment served as controls (T0).The B study group had 10 subjects (mean age 21 ± 2 years old) with five females and five males.This group was analyzed approximately 48 months following orthodontic treatment, at which time they presented with radiographic signs of mild to severe root resorption (PT).The C study group had ten volunteers matched by gender and age without treatment and no radiographic evidence of root resorption were used as controls (ST).Inclusion criteria were: no previous orthodontic treatment and no radiographic evidence of root resorption.Exclusion criteria were: trauma to the primary or permanent dentition, autoimmune disease, chronic inflammatory disease, periodontal disease, periapical lesions, asthma, active caries or oral mucosa lesions, use steroidal or nonsteroidal anti-inflammatory drugs for at least one month before sampling occurred.

Saliva samples
Saliva samples were collected at T0, T12, ST and PT.These samples were obtained between 10:00 and 16:00 to avoid the effect of cardian circle in salivary IgG secretion into saliva.Unstimulated whole saliva samples (2 ml) were collected by expectoration into sterilized vials after the subjects had rinsed their mouth twice with water.Saliva samples were centrifuged at 12000 rpm for 10 min, and the supernatants were then stored at -20°C until use.

Antigen preparation
Extract of human dentin containing NCP from the dentin matrix was used as the antigen.Powder dentin was obtained using a modified technique described by Wheeler and Stroup (1993); the third molars were donated by patients for whom extraction was indicated.The dentin was removed using a drill with a high-speed bit.The powder obtained was diluted at demineralizing solution (guanidine-HCl 5 M, 10%, ethylenediamine tetraacetic acid pH 5.0 and 1 µM phenylmethylsulfonylfluoride) for 14 days at 4°C and then centrifuged two times at 3.218 rpm for 20 min.The supernatant was collected.The protein content was assessed by the Lowry method (1951).Dentin extract was stored at -80°C until use.

Radiographs
Periapical radiographs were obtained from subjects at T0, T12, ST and PT (70 kV, 10 mA, exposure time 0.7 s).The radiographs of the upper central incisors were taken by using the long cone paralleling technique.Kappa values for the intra-examiner variation ranged from 0.85 to 0.9.The most resorbed incisor was considered for analysis.Tooth length was measured from the incisal edge to the apex.The measurements were made with a pachymeter (0.02 mm precision; Mitutoyo Sul Americana, São Paulo, Brasil) placed parallel to the pulp canal.Image distortion was determined by comparing the image length of a radiopaque object (metal clips) placed on the film.Image distortion between T0 and T12 radiographs was determined by comparing crown length.The maximum acceptable distortion was 5%.For analysis of degree of resorption, upper central incisor was selected, in subjects after treatment with orthodontic appliances.The degree of root resorption was classified by using criteria described by Levander and Malmgren (1988): mild resorption or degree I-with irregular apical root contour; moderate resorption or degree II-lesser 2 severe resorption or degree III-2 mm to one-third of the original root length.The patients presented Class I or Class II malocclusion.

Gel filtration chromatography on Sephadex G-25 and Sephadex G-100
To desalinate the extract, crude dentin extract was fractionated in a Sephadex G-25 gel filtration column (Sigma, St. Louis, USA).The automatic fraction collector (FC 203B; Gilson, Middleton, USA) was maintained at 4°C.Only one fraction was concentrated and refractionated on a Sephadex G-100 column (Sigma, St. Louis, USA).The eluted fractions in phosphate buffered saline (PBS) were read in a spectrophotometer at 280 nm (Pharmacia Biotech, Sweden).The total protein content of each fraction was assessed by the Lowry method (1951).

ELISA for the detection of anti-dentin salivary IgG
ELISA immunoplates (Techno Plastic Products, Zurich, Switzerland) were sensitized with 30 µg/mL FI or FIIa fractions in carbonate-bicarbonate buffer (Na 2 CO 3 1.59 g, NaHCO 3 2.93 g, distilled water, qsp 1000 ml, pH 9.6), incubated for 1 h at 37ºC and overnight at 4°C.The plate was washed four times with phosphatebuffered saline pH 7.2 (PBS), containing 0.05% Tween 20, and 0.5% skimmed milk.Then, the plates were blocked with PBS containing 0.5% Tween 20 and 5% skimmed milk for 1 h at room temperature.After washing, undiluted saliva samples were incubated at 37°C for 90 min.After washing, the conjugated goatanti-human IgG-peroxidase (A8775; Sigma-Aldrich, St. Louis, USA) was added at 1:4000 dilution, and incubated for 90 min at 37°C.It was added after washing the O-phenylenediamine substrate solution.The reaction was stopped after 15 min using 50 L/well of 4NH 2 SO 4 .Absorbances were read at 492 nm in a Multiskan EX reader (Lab Systems, Helsinki, Finland).Antibody levels were expressed as absorbance in optical density units (O.D.).

SDS-PAGE electrophoresis and Western blot
FI and FIIa were subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) on 5-15% gradient acrylamide gels and were transferred to nitrocellulose membranes (NCM).The NCM was blocked with PBS containing 0.5% Tween 20 and 5% skimmed milk for 60 min h at room temperature.The NCM were cut into individual strips containing fractions F or FIIa and a molecular weight standard.Each strip was incubated with pools of undiluted saliva samples (T0, T12, ST or PT) with 10 samples from each group, followed by the addition of 1: 1000 conjugated goat-antihuman IgG-peroxidase (A8775; Sigma-Aldrich, St. Louis, USA) 90 min at 37°C and overnight at 4°C.After washing tetramethyl benzidine (00-2019; Invitrogen, San Francisco, USA) was added.The reaction was stopped with distilled water.The controls were saliva samples from the T0 and ST subjects.

Data analysis
The following non-parametric tests were used: the Wilcoxon test for paired groups (T0 and T12) and the Mann-Whitney test for unpaired group (ST and PT).These tests were used to detect differences in the ELISA absorbances (antibody levels expressed as O.D.).The results were considered statistically significant when P<0.05.The Lilliefors and Shapiro-Wilks normality tests were used.

Gel filtration chromatography
Chromatography of dentin extract on a Sephadex G-25 column resulted in one main peak of light absorption at 280 nm.From this peak, two fractions were obtained: FI and FII (chart demonstrated in Figure 1A).The FII fraction was concentrated by lyophilization and was refractionated on a Sephadex G-100 column, resulting in a 280 nm spectrophotometric profile containing two peaks.Fractions corresponding to each peak were mixed, resulting in pool FIIa and FIIb (chart demonstrated in Figure 1B).Only fractions FI and FIIa were used in the following experiments.

Anti-FI and anti-FIIa salivary IgG levels detected by ELISA pre-treatment (T0) and after 12 months of orthodontic treatment (T12)
The anti-FI salivary IgG levels was higher at T12 than at T0 (P <0.05, Wilcoxon test), as is shown in Figure 2A.No statistically significant differences were observed in anti-FIIa salivary IgG levels between T0 and T12 (P>0.05,(Wilcoxon test), as is demonstrated in Figure 2B.

Anti-FI and anti-FIIa salivary IgG levels detected by ELISA in control donors (ST) and in patients approximately 48 months after orthodontic treatment (PT)
The anti-FI salivary IgG levels were higher in the PT group than in the ST group (P <0.05, Mann-Whitney test; Figure 2C), but no statistically significant differences in the anti-FIIa IgG levels were observed between these groups (P >0.05, Mann-Whitney test), as shown in Figure 2D.

FI and FIIa fraction analysis by Western blot
Western blot results show that the salivary IgG of the T12 or PT groups (but not the T0 or St groups) detected dentin antigens in the FI fraction (approximately 50 to 70kDa) and the FIIa fraction (approximately 35 kDa and 40 kDa).These data are shown in Figure 3A and 3B.

DISCUSSION
Based on the evidence that anti-dentin antibodies can be detected in animal experimental models and in patients with root resorption (King and Courts, 1989;Ng et al., 1990;Wheeler and Stroup, 1993;Hidalgo et al., 2005;Ramos et al., 2011), it has been hypothesized that pathological root resorption may be associated with autoimmune responses against dentinal components.The development of autoimune disorders could be bound to the presence of autoreactive T lymphocytes in the thymus that have not been deleted (Weiner et al., 1995).An effective mechanism to prevent self-reactivity is the compartmentalization of certain self-antigen under normal conditions (Kindt et al., 2008).And the exposure of some hidden antigens, released after injury, makes them accessible to specific receptors of antigen-presenting cells (APCs) and adaptive immune cells, which can result in the production of autoantibodies (Janeway, 2007).The presence of anti-dentin autoantibodies in the saliva of control groups could be explained by the indexes of dental resorption described in the general population (Massler and Malone, 1954).However, the high level of specific autoantibodies observed in this study in T12 may be due to the release of dentinal antigens as consequence of orthodontic movement, compared to control, without treatment.When dentin antigen-IgG complexes are internalized by APCs (macrophages, dendritic cells) and presented to autoreactive effector T cells, an inflammatory response will occur.Then, the antigen is eliminated, the ratio of IgG to antigen increases, decreasing immune responses and reducing tissue damage, with a slope of the response towards tolerance (De Groot et al., 2008).Natural regulatory T cell (nTreg) CD4 + CD25 Hi FoxP3 + , is an important mechanism of effector T cell regulation, and may represent one of the critical forms of autoregulatory response to self-antigens.On the other hand, the course of resorption will depend on a complex interaction of secreted molecules, inflammatory and bone cells within the tooth surrounding tissues (Ne et al., 1999).
The results shown, also indicated a significantly higher level of anti-FI fraction salivary IgG in (PT) than subjects (ST).This fact could be due to the persistence of the stimulus and the high half-life of the memory cells that produce antibodies anti-dentinal components (Schittek and Rajewsky, 1990;Slifka, 1998).Thus, the induction of inflammatory events by dentin molecules released into microenviroment may contribute to the maintenance of this process (Silva et al., 2004).Western blot analysis of the FI fraction also detected another high molecular weight band that occurred in all the samples, including the control groups (data not shown).No significant difference in the FIIa fraction was observed between the groups and its control using an ELISA.However, western blot analysis detected dentinal antigens of approximately 35 and 40 kDa by saliva of the T12 and PT groups but not in the control groups.The concentration of FIIa fraction by lyophilization was necessary to detection in western blot, by salivary IgG.The results often contradictory in the literature about in vivo and in vitro dentin studies can be explained by the use of different extraction methods, source of dentin and methods of cell treatment.Based on data of this study, we hypothesized that not all dentinal proteins may be immunogenic and some of them may be tolerogenic.It is possible that in the case of EARR, specially, mild or moderate, anti-dentin natural autoantibodies have a homeostatic function in mediating removal of dentinal components released within periodontal ligament microenvironment (HANS et al., 2007).In conclusion, our results suggest that antigenic fractions of approximately 35 to 70 kDa may be released from dentin, presented to the immune system and detected by salivary IgG in orthodontic patients with mild to severe root resorption.The high levels of salivary IgG to fraction of human dentin following orthodontic treatment may remain elevated even years after completion of therapy.The use of saliva samples provide great advantages of adherence by patients and may be a route of choice for the development of preventive methods, early diagnosis and understanding of the immunopathological mechanisms involved in the root resorption process.Further studies, with a larger number of patients, should be performed to confirm these results.

Figure 2 .
Figure 2. Analysis of the levels of salivary IgG against FI or FIIa fractions of human dentin extract using ELISA assays.Statistically significant differences (P < 0.05) are marked with asterisks.(A) Salivary IgG against FI fraction of human dentin extract in T0 or T12.(B) Salivary IgG against FIIa fraction of human dentin extract in T0 or T12.(C) Salivary IgG against FI fraction of human dentin extract in patients PT or ST.(D) Salivary IgG against FIIa fraction of human dentin extract in patients in PT or ST.

Figure 3 .
Figure 3.Western blot results of fractions FI or FIIa using pooled saliva samples from patients.(A) T0 or T12, (B) PT or ST.