Influence of the adhesive bonding protocol on the inflammatory cellular response and gelatinolytic activity in human teeth

The aim of this paper was to assess the biocompatibility and expression of metalloproteinase -9 of two dentin bonding agents in human teeth, using different methods of dentin pre-treatment and different time intervals. Deep Class I cavities were prepared on the occlusal surface of 18 sound molars. Restorations were performed with XP BondTM (Dentsply) and Futurabond® DC (VOCO). After 30, 90 or 120 days, the teeth were extracted and processed for histological and immunohistochemical assessment. In XP Bond group, was observed a moderate inflammatory infiltrate and immunoreactivity of metalloproteinase -9 in all storage intervals. In Futurabond DC group, a slight inflammatory infiltrate was found in the first intervals. After 120 days, the inflammatory infiltrate was either slight or absent, while the tissue morphology remained normal. The immunoreactivity of metalloproteinase -9 was considered moderate, except for two specimen after 120 days, that demonstrated no immunoreactivity. Therefore, etch-and-rinse adhesives exhibited a more intense pulpal response than self-etching agents. When performing adhesive restorations in medium and deep cavities in human teeth, the use of a self-etching adhesive strategy should be considered in view of the lower induction of inflammatory and proteolytic activity. Further studies of molecular events should be conducted, taking into consideration the possible inflammatory pulp reactions that can contribute to the success of restorative procedures.


Introduction
The vitality of dental pulp following restorative procedures depends on the degree of the inflammatory cell response, which is directly proportional to the remaining dentin thickness (RDT) (de Souza Costa et al. 2014, Murray et al. 2002, mechanical injury during tooth preparation and microleakage (Bagis et al. 2007). The pulpal inflammatory response can also be modified by dental materials, such as the dentin bonding agent (DBA), its components (Costa et al. 2021), and the technique used (etch-and-rinse or self-etching) (Alves & Sobral 2015, Elias et al. 2015.
Current dentin bonding agents follow either "etch-and-rinse" or "self-etching" protocols. Etch-and-rinse bonding agents contain phosphoric acid to pretreat dental tissues, followed by rinsing and the application of an adhesive. Self-etching bonding agents does not require pretreat dental tissues, but it contains acidic monomers, which simultaneously demineralize and infiltrate during the application (da Silva et al. 2014, Ozer & Blatz 2013, leading to the interposition of a (resinimpregnated) smear along the dentin surface (Suyama et al. 2013, Arrais & Giannini 2002. The composition of the material, the concentration and chemistry of the components and the tissue responsiveness influence the interaction between dental materials and tooth tissue (Ferracane et al. 2010). Substances commonly found in both DBA protocols have a toxic effect, i.e., Bis-GMA (bisphenol-A glycidyl methacrylate), TEGMA (triethylene glycol dimethacrylate), HEMA (hydroxyethyl methacrylate) and UDMA (urethane dimethacrylate) (da Silva et al. 2014, Nowicka et al. 2012, Tran-Hung et al. 2012. These substances can diffuse through dentin, reaching the pulp tissue, affecting the inflammatory cellular response and the biocompatibility of the restorative material (Machado et al. 2016, Elias et al. 2015, Shafiei et al. 2014, Bianchi et al. 2013, Nowicka et al. 2012, Kostoryz et al. 2003, Huang & Chang 2002.
When faced with an aggressor agent, an inflammatory process is initially triggered in the dentin-pulp complex.
Subsequently, there is a tendency to resolve this inflammatory process in the absence of bacterial infection (Leonardi et al. 2011). If the offending stimulus is high intensity, pulp necrosis may occur. Chemical materials applied to the dentin-pulp complex cause an inflammatory response (Leonardi et al. 2011). This response should be as minimal as possible, in order to avoid irreversible damage to the pulp tissue (de Souza Costa et al. 2014, Hebling et al. 2010. The biological response is extremely important in terms of maintaining the vitality of the dentin-pulp complex and should be carefully considered during clinical procedures (Hebling et al. 2010).
The inflammatory process involves inflammatory mediators (Bagis et al. 2007), dental fluid, odontoblasts, lymphocytes, neuropeptides, chemokines and inflammatory cytokines, which lead to a decrease in pH (Leonardi et al. 2011) and an increase in proteolytic activity, especially, in matrix metalloproteinases (MMPs) (Boelen et al. 2019). The MMPs are enzymes capable of degrading all components of the extracellular matrix (Sabatini & Pashley 2014). MMPs are believed to be involved in physiological processes, such as dentin maturation during aging, but also in pathological processes such as caries, periodontal disease, and pulpal inflammation (Boelen et al. 2019;Shimada et al. 2009). The MMPs can be activated by the drop in pH consequent to the treatment of the surface with primer and adhesive or to the biochemistry of the carious process, leading to the degradation of collagen fibrils and failure in the adhesion process .
It is still necessary to validate in vivo the activity of the collagenolytic and gelatinolytic endogenases that are stimulated by the acid (from acid attack of etch-and-rinse adhesives and acid monomers of self-etching adhesives), and it results in degradation of the hybrid layer (Hebling et al. 2005), comparing the two adhesive techniques over time. MMP-9 is the metalloproteinase predominantly found in the dentin matrix (Sabatini & Pashley 2014) and seems to be closely linked to pulpal pathological processes (Accorsi-Mendonça et al. 2013) and can serve as a biological marker related to the cellular inflammatory response.
It is believed that a group of activities at a molecular level are responsible for the structural degradation of tissue.
Concerning the pulp tissue, an understanding of the inflammation stages of this tissue is essential, together with knowledge of the molecular events that can contribute to the maintenance of the vitality of the pulp, in order to preserve dental elements during restorative procedures. Therefore , the aim of the present study was to assess the biocompatibility of two dentin bonding agents in human teeth cavities based on the inflammatory cellular response (cellular response and pulp tissue organization) and the immunoexpression of MMP-9 of pulp tissue, using different methods of dentin pre-treatment (etch-and-rinse and selfetching) and different time intervals.

Methodology
This was a longitudinal, descriptive, and observational study. A clinical or experimental phase was followed by a laboratory phase for histopathological analysis (Costa et al. 2021).

Clinical procedures
This study was conducted using the same clinical methodology used in the study of Costa et al. (2021). The present study was approved by the Ethics Committee of the University of Pernambuco (protocol No. 297/09). Eighteen patients (age range 20-40 years) with an indication for third molar extraction were selected and randomly divided into nine groups (n = 3/group). All of the participants signed a statement of informed consent.
The third molar was anesthetized and occlusal cavities were prepared with a high-speed handpiece (under watercooling) and a # 245 carbide bur (replaced after each preparation to avoid excessive heating). The cavity was approximately 1 -2 mm deep from the roof of the pulp chamber. The patients were randomly assigned to groups in which they would receive either the etch-and-rinse or the self-etching dentin bonding agent. This assignment was performed using a single-blinded envelope method. Research, Society andDevelopment, v. 10, n. 9, e56610918552, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i9.18552 4 Two bonding agents were used: XP Bonda etch-and-rinse adhesive, which involves an etching technique with 37% phosphoric acid, prior to the application of the adhesive from a single bottle containing both the primer and the adhesive; and Futurabond DCa self-etching, one-step adhesive, which contains acid, primer and adhesive in a single bottle. In each group, the bonding procedures were carried out in accordance with the manufacturers' instructions (Table 1). Cavities were restored using the microhybrid composite resin Filtek Z250 A3 (3M ESPE Dental Products, St. Paul, MN, USA). Table 1: Specifications of the materials tested in the present study.
After the restorative procedure, a new randomization (single-blinded envelope method) was performed for the analysis at different time intervals. For each time interval (30, 90 and 120 days), teeth extractions were performed and three specimens were taken for each material tested. The specimens were then immersed in a solution of 10% formaldehyde.

Laboratory procedures
During the laboratory phase, semi-serial sections (thickness of 5 μm for hematoxylin-eosin -H.E.-staining and 3 μm for immunostaining) were obtained in microtome (Leica RM 2125, Leica Microsystems, Nussloch GmbH, Germany). Immunohistochemical (I.H.C.) staining was performed using streptavidin-biotin optimized peroxidase, an LSAB amplification system (LSAB -Dako, Glostrup, Denmark) and the monoclonal antibody anti-MMP-9, with diaminobenzidine (DAB) used as chromogen. The primary antibody is specified in Table 2. The H.E. sections were assessed for two histologic features: the inflammatory cellular response and tissue disorganization. The degree of pulp inflammation and the organization of the pulp tissue were analyzed in accordance with the criteria proposed by de Souza Costa et al. (2006), as described in Table 3. The I.H.C. sections were examined considering the following histologic features: the type of immunopositive cell; the tissue location (focal or diffuse) and the intensity of the immunostaining. The immunostaining pattern was classified as positive (+) or negative (-). Positive cases were also classified quantitatively, according to the intensity of the marking: weak (+), moderate (++) and intense (+++). Subsequently, a marking score as a percentage of positive cells was assigned: score 0: no positive cell; score 1: 1-25% positive cells; score 2: 25% -50% positive cells; and score 3: > 50% of positive cells.
The negative control involved slides on which the primary antibody had been replaced by immunoglobulin G (IgG).
The positive control used breast carcinoma for MMP-9.

Results
Analysis of cellular and morphological response and immunoexpression of MMP-9 was performed for all experimental groups.   Source: Authors.
Histological features for XP Bond bonding agent in each interval group can be exemplified in Figure 1. For the 30-, 90-and 120-day intervals in the XP Bond dentin bonding agent group, a moderate inflammatory infiltrate (score 2) was observed in three specimens ( Fig. 1(a)). Partial disorganization of the pulp tissue morphology and odontoblastic layer (score 1) was found for all specimens ( Fig. 1(b)/1(c)). Regarding MMP-9 immunoexpression, the three intervals presented moderate and focal expression for all specimens (Fig. 1(d)). Each time interval presented a specimen with localized expression in the odontoblastic layer, close to the congested vessels, and in the inflammatory cells in the center of the pulp. For the other two specimens of each time interval, the expression was evidenced in the odontoblastic layer and in the inflammatory infiltrate ( Fig. 1(e)/1(f)).  Research, Society and Development, v. 10, n. 9, e56610918552, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i9.18552 Histological features Futurabond DC bonding agent in each interval group can be exemplified in Figure 2. For the 30and 90-day intervals in the Futurabond DC dentin bonding agent group, all of the specimens exhibited a slight inflammatory infiltrate close to the odontoblastic layer and a disorganized odontoblastic layer, although the central portion of the dental pulp was normal (Fig. 2(a)/2(b)). After 120 days, normal tissue morphology was observed, with a slight (or absent) inflammatory infiltrate (Fig. 2(c)). Regarding the MMP-9 immunoexpression, all specimens within the 30 and 90-day intervals and one 120day interval specimen exhibited moderate and focal expression with predominant localization in the odontoblastic layer and inflammatory infiltrate (Fig. 2(d)/2(e)). In the other two specimens of the 120-day interval, no MMP-9 expression was observed ( Fig. 2(f)).

Discussion
This study observed that previous acid etching led to a greater inflammatory cellular response and expression of MMP-9. The hybridization technique may have a cytotoxic effect on pulp tissue after acid etching (de Souza Costa et al. 2014). The toxic effects of materials may risk cell function and at long last cause cell death (Ferracane et al. 2010). The response of the dentin-pulp complex to the application of a dentin bonding agent in deep cavities in human teeth has been previously analyzed, however, the in vivo biological effect comparing adhesive protocols has not been widely studied and deserves further attention. Büyükgüral and Cehreli (2008) reported that restorative procedures in deep cavities may cause irreversible pulp tissue reactions, particularly when the RDT is less than 0.5 mm. Ballal et al. (2017) found more MMP-9 in deep carious lesions than from in the shallow one. The intensity of the pulp reaction increased as the remaining dentin thickness decreased (Hebling et al. 1999). In the present study, the remaining dentin was approximately 2 mm. This quantity of remaining dentin can be considered a protective barrier for the pulp against injuries from the adhesives irritant agents (Alves & Sobral 2015).
The XP Bond group exhibited a slight inflammatory cellular response in most specimens. However, a moderate response was also observed. This increase in the inflammatory cell response may have occurred due to the pre-treatment with 37% phosphoric acid. Etching exposes collagen fibers that will be filled by the adhesive system at a later stage. It is known that the chemical diffusion gradient of the bonding agent prevents the complete infiltration of adhesives (Strobel & Hellwig 2015).
The insufficient penetration of the etch-and-rinse DBA monomers along the entire length of the demineralized dentin leaves unfilled gaps (da Silva et al. 2014). The presence of exposed collagen fibrils leads to proteolytic degradation and an increased pulp inflammatory response (Elias et al. 2015, Bianchi et al. 2013).
In addition to the increased proteolytic activity, acid etching increases the permeability of dentinal tubules, furthering the contact between the cytotoxic substances in dentin bonding agents and the pulp tissue (Shafiei et al. 2014), even in deep cavities where dentinal tubules are naturally larger. Monomers' leach may occur in the first 24 hours until a month after the restorative procedure and trigger pulp inflammation and MMP induction (Lovász et al. 2021).
Concerning the organization criteria of pulp tissue, the odontoblastic layer was found to be disorganized. Assuming that odontoblasts are the most peripheral pulp cells, they are more susceptible to aggressions arising from the application of dental materials (Elias et al. 2015, de Souza Costa et al. 2014. Therefore, it is believed that the XP Bond, through its components and the associated tooth surface treatment, caused stimuli that led to reversible cellular damage in the odontoblastic layer. This is a transitory event which may lead to some form of defense response (Murray et al. 2002).
The favorable results observed in the Futurabond DC group may be explained by the dentin pre-treatment method involved and the composition of the bond agent. The weak acid in the Futurabond DC bonding agent (Arrais & Giannini 2002) and the interposition of a (resin-impregnated) smear along the dentin surface results in what may be considered as a positive biological response, despite the fact that the DBA contains components considered to be irritants (Murray et al. 2002). The smear layer obliterates the dentinal tubules, reducing dentin permeability and limiting the toxicity of self-etching bonding agents on the pulp (de Souza Costa et al. 2014). De Munck et al. (2009 analyzed in vitro the MMP-2 and MMP-9 release after using DBA through zymography and observed endogenous enzymatic activity only for etch-and-rinse adhesive against mild self-etching adhesive. In the first stage of the analysis, tissue disorganization was observed along almost the entire length of the odontoblastic layer, with normal central pulp tissue morphology. Interestingly, after 120 days, the odontoblastic layer had been reorganized ("in palisade") and involved all of the dentine, with no inflammatory cells. The dental pulp has an elevated inherent regenerative potential (Tran-Hung et al. 2012). This could be explained by the reparative capacity of ectomesenchymal cells, which reactivate the condition of the affected pulp, thereby favoring the constant remodeling of the odontoblastic layer. This pulp response proves the reparative capacity of the pulp tissue (de Souza Costa et al. 2014, Hebling et al. 2010, mainly due to the decrease in dentin permeability owing to the deposition of reactionary dentin over time (Büyükgüral & Cehreli 2008).
O 'Boskey Jr and Panagakos (1998) proposed that inflammatory cytokines stimulate gelatinase levels in human pulp connective tissue cells, and MMPs were stimulated during the intense pulpal inflammatory process as in any other inflammatory process. MMP-9 has already been detected in both mineralized and demineralized dentin (Van Strijp et al. 2003). MMPs-9 are synthesized by the odontoblasts (Sulkala et al. 2002) and data present in the literature have showed the presence of MMP-9 in healthy and inflamed pulps. According to the results of the present study, we suggest the relation of the expression of the MMP-9 enzyme with the regions of the pulp tissue where there was disorganization of the odontoblasts layer and the inflammatory cells near the congestive vessels.
In the study by Gusman et al. (2002), the levels of MMPs (-1, -2, -3, and -9) were evaluated in human dental pulps under conditions of normality and inflammation, through enzyme linked immunosorbent assay (ELISA) and zymography.
According to the results, the high absolute quantification of MMP-9 in inflamed pulp and the positive correlation with gelatinolytic activity suggest that MMP-9 is important in the degradation of pulp tissue in inflammatory processes. Costa et al. (2021) evaluated the expression of MMP-9 after using a etch-and-rinse adhesive on human teeth and observed that the immunoreactivity of MMP-9 was considered intense in the most of specimens in the first storage intervals, before decreasing over time.
The expression of MMP-9 at the 30-and 90-day intervals for both Futurabond DC and XP Bond was considered moderate, located in the odontoblasts layer, and in inflammatory cells, thus the considerations are similar for both adhesive systems, denoting the influence of factors such as: surgical technique of cavity preparation, depth of preparation and restorative technique. For the 120-day interval, in two specimens of the Futurabond DC group, expression of MMP-9 was not visualized.
We believe that this result is related to the tissue normality found and for not having presented any inflammatory conditions.
In the groups of the Futurabond DC adhesive system, the capacity of pulp tissue to be repaired in the face of low intensity stimuli in medium depth cavities was clearly observed. In addition to the reaction caused by dentin adhesives, the influence of phosphoric acid on process is extremely important because the acids, when applied directly on the dentin, facilitates fluid movement , as well as the transport of substances that are part of the composition of the adhesive systems up to the pulp. However, in self-etching adhesive systems, such as Futurabond DC, the acids are weaker than the phosphoric acid (Perdigão 2007), which may lead to a lower inflammatory response and, consequently, lower proteolytic activity, as observed in the present study.

Conclusion
In the present study, the two methods of dentin pre-treatment ("etch-and-rinse" and "self-etch") generated different inflammatory cellular responses and could provide practical parameters for clinical decision making. Considering the limitations and the results of the present study, as well as the findings described in the literature and seeing the need to simplify the dental clinical practice, we believe that the self-etching adhesive systems are the best choice in thin RDT cavities. Further longitudinal in vivo studies are needed to investigate the expression of proteolytic molecules capable of degrading the hybrid layer and contribute to the decrease of the adhesive bond stability and the substances capable of inhibiting this activity.