Bond stability of conventional adhesive system with MMP inhibitors to superficial and deep dentin

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

Contents lists available at ScienceDirect

Journal of the Mechanical Behavior of Biomedical Materials journal homepage: www.elsevier.com/locate/jmbbm

Bond stability of conventional adhesive system with MMP inhibitors to superficial and deep dentin

T

Fabiana Souza Simmer, Eduardo Moreira da Silva, Rafaela da Silva Gonçalves Bezerra, Maria Elisa da Silva Nunes Gomes Miranda, Jaime Dutra Noronha Filho, Cristiane Mariote Amaral∗ Analytical Laboratory of Restorative Biomaterials, School of Dentistry, Universidade Federal Fluminense, Niterói, Rio de Janeiro, Brazil

A R T I C LE I N FO

A B S T R A C T

Keywords: Dentin bond stability Sorption Solubility Microshear bond strength MMP inhibitor

Purpose: To evaluate the microshear bond strength (μSBS) to deep (DD) or superficial (SD) dentin (μSBS) overtime, nanoleakage (AG%), degree of conversion (DC%), water sorption (WSp), and solubility (WSl) of an adhesive system [Adper Single Bond 2(ASB)] containing matrix metalloproteinases (MMP) inhibitors [GM1489 (ASB-GM), Batimastat (ASB-BAT), or Chlorhexidine diacetate (ASB-CHX)]. ASB without inhibitor was used as control (CONTROL). Materials and methods: WSp and WSL were calculated based on ISO4049. DC% was analyzed using FT-IR spectroscopy. Dentin discs were used for μSBS evaluation. For AG%, resin-dentin beams were analyzed under scanning electronic microscopy. Data were analyzed using three-way ANOVA (AG% and μSBS) or ANOVA (DC%, WSp, WSl) and Tukey’s HSD test. Results: ASB-CHX presented the lowest DC%, lowest WSp, and highest WSl. ASB-GM reached the highest immediate μSBS in SD, only different from ASB-CHX. In DD, ASB-BAT and ASB-GM had the highest μSBS, statistically different from ASB-CHX. After twelve months, ASB-GM and ASB-BAT presented higher μSBS in SD when compared to CONTROL and ASB-CHX. In DD, ASB-GM reached the highest value, which was statistically different from CONTROL and ASB-CHX. CONTROL at both dentin depths and ASB-CHX at DD did not maintain bond stability. In SD after 12 months, ASB-BAT and ASB-GM decreased AG%. In DD, only ASB-GM reduced AG%. Conclusion: The ASB containing Batimastat and GM1489 maintained resin-dentin bond stability after 12 months for both dentin depths, without jeopardizing WSp, WSl, or DC% of the adhesive system. The ASB-GM presented greater μSBS after 12 months when compared with ASB. Clinical significance: Batimastat and GM1489 could be suitable for inclusion as an MMP-inhibitor into Single Bond to improve the bond stability to superficial and deep dentin, without jeopardize the physic-mechanical proprieties.

1. Introduction Morphological and structural differences between superficial and deep dentin can influence directly the quality of the hybrid layer formed by adhesive systems(Zhang et al., 2014). The bonding to deep dentin can be impaired due to low contents of intertubular dentin and collagen fibrils, which plays an important role during adhesion. (Marshall-Jr et al., 1997; Pashley and Carvalho, 1997). Moreover, the number of tubules is higher near the pulp chamber, which increase

dentin’s intrinsic wetness, and this moisture has been frequently considered a barrier to effective bonding(Marshall-Jr et al., 1997). Previous study also showed that bonding to deep dentin is more susceptible to degradation than in superficial dentin after aging (Zhang et al., 2014). The degradation of the hybrid layer has been considered the main limitation to bonding stability(Breschi et al., 2008; De-Munck et al., 2005). The hydrolysis of resin components caused by sorption is one of the mechanisms involved in this process, and the hybrid layer promoted by contemporary hydrophilic dentin adhesives behave as semi-

Abbreviations: microshear bond strength, μSBS; deep dentin, DD; superficial dentin, SD; nanoleakage, AG%; degree of conversion, DC%; water sorption, WSp; solubility, WSl; Adper Single Bond 2, ASB; matrix metalloproteinases, MMP ∗ Corresponding author. Universidade Federal Fluminense / Faculdade de Odontologia, Rua Mário Santos Braga, nº 30 - Campus Valonguinho, Centro, Niterói, RJ, CEP 24020-140 -, Brazil. E-mail addresses: [email protected]ff.br, [email protected] (C.M. Amaral). https://doi.org/10.1016/j.jmbbm.2019.103402 Received 18 March 2019; Received in revised form 13 August 2019; Accepted 19 August 2019 Available online 22 August 2019 1751-6161/ © 2019 Elsevier Ltd. All rights reserved.

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

Table 1 Materials used in this study. Material

Composition (in weight) / Characteristics

Manufacture (Batch #)

BisGMA (10–20%), HEMA (5–15%), ethanol (25–35%), water (< 5%), glycerol 1,3dimethacrylate (5–10%), UDMA (1–5%) copolymer of acrylic and itaconic acids (5–10%), silane treated silica (10–20%) Bis-EMA, Bis-GMA, TEGDMA, UDMA, zirconia / silica with 0.01–3,50 μm and medium size 0,6 μm and 60% by volume of the filler particles 37% phosphoric acid

3M ESPE, St Paul, MN, USA; (N672600BR)

Adhesive System

Adper Single Bond 2 (ASB2)a

Resin Composite Etch

Filtek Z250

Inhibitor

Batimastat

Inhibitor

GM1489

EMD Chemicals, Inc. San Diego, CA, EUA (D00123262)

Inhibitor

Chlorhexidine diacetate

Sigma Aldrich, St Louis, EUA (C1520000)

Acid etch gel

Bis-GMA = bisphenol A diglycidyl methacrylate; HEMA = 2-hydroxyethyl methacrylate; TEGDMA = triethylene glycol dimethacrylate; UDMA = urethane dimethacrylate. a also known as Adper Single Bond Plus or Adper Scotchbond 1XT.

3M ESPE, St Paul, MN, USA; (N682592) Dentsply, Rio de Janeiro, RJ, Brazil; (L1400228H) EMD Millipore Corp, Billerica, MA, EUA (D00166749)

Bis-EMA = ethoxylatedbisphenol

A

glycol

dimethacrylate;

which could favor its chelation potential(da-Silva et al., 2015). Although rarely studied in adhesive dentistry, both Batimastat and GM1489, have shown promise in maintaining the resin-dentin bond stability(da-Silva et al., 2015) and in the inhibition of dentin MMPs (Almahdy et al., 2012). Considering the different distribution of MMPs and their gelatinolytic potential, as well as the morphological differences and different patterns of hybrid layer degradation in superficial and deep dentin, it is important to investigate the effectiveness of different MMP inhibitors at different dentin depths in order to increase the bonding stability. Therefore, the aim of this study was to evaluate the effect of different MMP inhibitors (Batimastat, GM1489 and Chlorhexidine diacetate) incorporated into a conventional adhesive system on the stability of bonding to superficial and deep dentin and to characterize these adhesive systems. The null hypotheses were: (1) The presence of different MMP inhibitors did not decrease the sorption, solubility and degree of conversion of the adhesive system, (2) The different inhibitors did not increase the dentin bond strength and did not decrease nanoleakage (immediately and after 12 months of storage), and (3) The deep dentin did not present lower bond strength than superficial dentin, independent of inhibitor used.

permeable membranes permitting water movement throughout the bonded interface, even after polymerization(De-Munck et al., 2005; Liu et al., 2011; Malacarne et al., 2006). The incomplete resin infiltration, especially in etch-and-rinse adhesive systems, can also lead to exposed collagen fibrils (Armstronga et al., 2001; Pashley et al., 2011) that stay susceptible to degradation by endogenous collagenolytic enzymes (Pashley et al., 2004; Zhang and Kern, 2009). Furthermore, the acidic environment induced by adhesive systems may activate matrix metalloproteinases (MMPs) present in dentin. Once activated, MMPs can accelerate the degradation of collagen fibrils (Devito-Moraes et al., 2016; Visse and Nagase, 2003). Mineralized dentin contains MMPs −2, −3,-8, and −9(Mazzoni et al., 2011, 2012; Tjaderhane et al., 2015; Visse and Nagase, 2003). However, it has been shown that the distribution of MMP-2 and MMP-9 vary according to dentin depth. The concentration of these MMPs and their gelatinolytic potential decrease gradually from deep to superficial dentin. This activity variation can lead to different patterns of hybrid layer degradation according to the dentin depth(Niu et al., 2011). The use of synthetic MMP inhibitors has been proposed to improve bonding stability(Acharya et al., 2004; Almahdy et al., 2012;Breschi et al., 2010a; , Breschi et al., 2010b Liu et al., 2011; Thompson et al., 2012). The mechanism of inhibition can occur through pseudopeptides that copy structural components of MMP substrates and act as competitive, reversible inhibitors(Acharya et al., 2004). Chlorhexidine is the most common MMP inhibitor studied and can reduce dentin-resin degradation when used as a pre-treatment(Brackett et al., 2009) or when added to experimental(da-Silva et al., 2015; Yiu et al., 2012) and commercial(Stanislawczuk et al., 2014; Zhou et al., 2009) adhesive systems, while also decreasing the gelatinolytic activity of MMPs (Breschi et al., 2010a). Batimastat is a synthetic low-molecular-weight peptide-like analogue of the collagen substrate that has been used as MMMP inhibitor in cancer treatment because of its ability of inhibiting tumor growth(Almahdy et al., 2012). GM1489 is a potent broad-range MMP inhibitor that has functional groups capable of combining with specific sites of MMPs. It presents a complex heterocyclic structure,

2. Materials and Methods For this in vitro study, three different MMP inhibitors were incorporated into a commercial adhesive system, Adper Single Bond 2 (ASB): Batismatat, GM1489, and Chlorhexidine Diacetate, forming the groups ASB-BAT, ASB-GM, and ASB-CHX, respectively. The control group used the conventional adhesive system without an inhibitor (CONTROL). ASB was selected because it is one of most common adhesive system used in clinical practice and it was used previously with inclusion of inhibitors(Almahdy et al., 2012). The inhibitors, Batimastat and GM1489, were added at a concentration of 5 μM(Almahdy et al., 2012; da-Silva et al., 2015) (0.0002% in weight) and chlorhexidine diacetate was added at a concentration of 2% in weight(da-Silva et al., 2015; Yiu et al., 2012). All inhibitors were weighed using an analytical 2

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

Fig. 1. Schematic illustration of the μSBS test.

and they were repeatedly weighed at 24 h intervals until a constant mass was attained (m2) The specimens were again put in a desiccator containing fresh silica gel, at 37 °C (Q316B15, Quimis, Rio de Janeiro, Brazil), and weighed daily until a constant mass obtained (m3). WSp and WSl in μg/mm³ were calculated using the following formulae:

balance (AUW 220D, Shimadzu, Tokyo, Japan) and mixed into ASB2 using a dual centrifuge (150.1 FVZ Speed Mixer DAC, Flack Tek Inc., Herrliberg, Germany). The pH of each adhesive formulation was measured using a pH meter (S220 Seven Compact pH/ION, Mettler Toledo; Columbus, OH, USA). The LED unit (Radii Cal, SDI, Bayswater, Victoria - Australia) was used for light cure adhesive system and resin composites in all methodologies. The irradiance of LED unit (approximately 650 mW/cm2) was verified by radiometer (LED Radiometer, SDI, Bayswater, Victoria - Australia) every ten specimens. Filtek Z250 was used as restorative material for microshear bond strength and nanoleakage tests. The composition of these materials is shown in Table 1.

WSp =

m2 − m3 V

WSI =

m1 − m3 V

where m1 is the disk mass (mg) after drying, m2 is the disk mass (mg) at the equilibrium uptake (maximum sorption), m3 is the mass (mg) of the re-dried disk, and V is the disk volume (mm³).

2.1. Sorption (WSp) and solubility (WSl) The solvent from the adhesive systems with incorporated inhibitors was allowed to evaporate. The adhesive systems were dispensed onto to a container (5.0 cm in diameter and 1.0 cm in depth) on an analytical balance with a precision of 0.01 mg (AUW 220D, Shimadzu, Tokyo, Japan) that was protected from ambient light to prevent premature polymerization. The initial mass was recorded and the specimens remained on the analytical balance until reaching mass equilibrium. Disk-shaped specimens were prepared using an aluminum mold (1 mm thick and 6 mm in diameter). A micropipette was used to dispense the adhesive systems directly into the mold. After filling the mold to excess, air bubbles were carefully removed using a hypodermic needle. After a polyester strip and glass slide were placed on top of the mold, the disks were light cured for 40 s(da-Silva et al., 2015). The disks were removed from the mold and the bottom surface was light cured for 40 s. During light-curing, the tip of LED unit was in close contact with glass slide. The top and bottom surfaces of all disks were manually polished using 1200- and 4000-grit SiC abrasive paper (Arotec, Cotia, SP, Brazil) to eliminate any surface irregularities. Six disks for each group (n = 6) were produced for the CONTROL, ASB-BAT, ASB-GM, and ASB-CHX groups. WSp and WSl were determined based on the ISO 4049 Standard (2000) as previously described(da-Silva et al., 2015). The disks were placed in a desiccator containing dehydrated silica gel and were weight daily using an analytical balance with a precision of 0.01 mg (AUW 220D, Shimadzu, Tokyo, Japan) until a constant mass was attained (m1) The thickness and diameter of each disk were measured using a digital caliper (MPI/E−101, Mitutoyo, Tokyo, Japan), and the volume (V) was calculated in mm³. The specimens were then individually placed in sealed glass vials containing 10 ml of distillate water at 37 °C

2.2. Degree of conversion (DC%) Standardized quantities of each adhesive system (0.6 μl) were dispensed onto an ATR crystal of a FT-IR spectrometer (Alpha-P/Platinum ATR Module, Bruker Optics GmbH, Ettlingen, Germany) and the spectra between 1600 and 1700 cm−1 were recorded with 120 scans at a resolution of 4 cm−1. Afterwards, the adhesive systems were light-cured for 40 s and the spectra were immediately recorded (n = 5). The DC% was calculated as the ratio between the integrated area of absorption bands of the aliphatic C]C bond (1638 cm−1) to that of aromatic C]C bond (1608 cm−1), used as an internal standard, which were obtained from the cured and uncured films, using the following equation:

Rpolymerized ⎞ ⎤ ⎡ DC% = 100 × ⎢1 − ⎜⎛ ⎟ ⎥ R ⎝ unpolymerized ⎠ ⎦ ⎣ where R = integrated area at 1638 cm−1 ̸ integrated area at 1608 cm−1. 2.3. Microshear bond strength test (μSBT) The μSBS test is illustrated in Fig. 1. Sixty-four non-carious molars (Ethical Committee Approval HUAP 1.972.896) were obtained from patients with age ranging from 18 to 35 years old. Teeth were disinfected in 0.5% chloramine (Dewald, 1997; Mobarak et al., 2010) for 7 days, stored in distilled water and used within 3 months after extraction. Two sections perpendicular to the long axis of the tooth were made using a cutting machine (IsoMet 1000, Buëhler, Lake Bluff, IL, 3

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

USA). The first section removed the occlusal enamel and the second section was made near to the top of pulp chamber, removing the root portion and the cervical third of the crown. Therefore, dentin disks were obtained. Retentions were made in the peripheral enamel of the discs, which were embedded in colorless polystyrene resin, leaving the dentin surfaces exposed. The superficial dentin surfaces were wet ground in a polishing machine (DPU-10, Struers, Copenhagen, Denmark) using 400 grit SiC for removing of residual enamel and 600 grit SiC papers were used for 1 min to standardize the smear layer. In this way, the superficial dentin was characterized by total removal of enamel and minimal wear of dentin. To standard the deep dentin, the dentin above of pulp chamber were wet ground in a polishing machine using 400 grit SiC to plan the surface and 600 grit SiC papers were used for 1 min to standardize the smear layer. Then, discs with 2.0 mm thickness were produced and verified with digital caliper (MPI/E−101, Mytutoyo; Tokyo, Japan), with the top surface corresponding to superficial dentin and the bottom surface corresponding to deep dentin. The specimens (containing superficial dentin and deep dentin) were randomly divided into 8 groups (n = 8), according to the different MMP inhibitors (NON INHIBITOR, ASB-BAT, ASB-GM, and ASB-CHX) and storage time (24 h and 12 months). The dentin surfaces were etched with 37% phosphoric acid for 15 s, rinsed with distilled water for 30 s and blot dried with absorbent paper. The adhesive system was applied following the manufacturer's instructions: two consecutive layers of the adhesive system were actively applied, followed by air-drying for 5 s each. 4 to 6 micro-bore tygon tubes (internal diameter = 0.8 mm, height = 1 mm; Embramed, Rio de Janeiro, RJ, Brazil) were placed on the treated dentin surface and then the adhesive system was light-cured for 20s, with the tip of LED unit in close contact with tygon tubes. Composite resin (Filtek Z250, 3M ESPE, St. Paul, MN, USA) was inserted into the tygon tubes and light-cured for 20s, with tip of LED unit positioned 1 mm from tygon tubes). After 1 h, the PVC tubes were carefully removed using a scalpel and the resin composite cylinder was checked to identify the presence of air bubbles or interfacial gaps. The specimens were stored in distilled water at 37 °C for 24h or 12 months and then subjected to μSBS testing (DL, 2000, EMIC, São José dos Pinhais, SP, Brazil) with a guillotine of 1 mm of thickness and load cell of 50N at a crosshead speed of 1.0 mm/min. During the adhesive procedure and the microshear bond strength test, the opposite side of dentin discs was protected with wet tissue papers to prevent dentin dehydration. Failure modes were evaluated with a stereomicroscope at 40x magnification (SZ40, Olympus, Tokyo, Japan) and classified as: adhesive (failures at the adhesive interface), cohesive (failures occurring mainly within dentin or resin composite), or mixed (mixture of adhesive and cohesive failure within the same fractured surface).

directions, across the bonded interfaces (IsoMet 1000, Buëhler, Lake Bluff, IL, USA) producing beams with a cross-sectional area of approximately 1 mm2. After storing for 24 h, two beams of each dentin fragment received two layers of nail varnish, 1 mm from the bonding interface on both sides. Then, the beams were individually immersed in an aqueous ammoniacal silver nitrate solution (50% in weight; pH = 7.0) and kept in a dark environment for 24 h. Each specimen was rinsed in running water and then immersed in a photo-developing solution (Kodak, Rochester, NewYork, USA) under fluorescent light for 8 h, in order to obtain precipitation of the metallic ions of silver along the bonding interface. The surfaces were polished with silicon carbide paper of decreasing abrasiveness (600-, 1200- and 4000-gritSiC paper), ultrasonically cleaned (Ultrassom 750 USC – Quimis, Rio de Janeiro, Brazil) for 15 min and dried for 24 h in a desiccator with blue silica gel at 37 °C. After 12 months storage, the remaining two beams of each dentin fragment were prepared similarly. The resin/dentin interface was observed using scanning electron microscopy (SEM) (Phenom ProX, Phenom-World BV, Eindhoven, Holland), at an accelerating voltage of 15 kV, backscattered and a charge reduction sample holder. Three images were registered for each beam: one from each end (right and left end) and one from the center, at a magnification of 2000x. The amount of silver nitrate uptake in the hybrid layer was registered as a percentage of the total area observed, using an EDS - Energy dispersive X-Ray Analysis program (Phenom ProX, Phenom-World BV, Eindhoven, Holland). 2.5. Statistical analysis The obtained data were analyzed using Statgraphics Centurion XVI software (STATPOINT Technologies, Inc, Warrenton, VA, USA). For μSBS and for each dentin depth, the dentin disc was considered the experimental unit, with the mean of μSBS of all composite cylinders representing each dentin disc). Initially, the normal distribution of errors and the homogeneity of variances were checked using ShapiroWilk's and Levene's tests, respectively. Based on these preliminary analyses, WSp, WSl and DC% were analyzed by ANOVA and Tukey's HSD test for multiple comparisons. The μSBS and nanoleakage were analyzed by three-way ANOVA (inhibitor vs dentin depth vs storage time) and Tukey's HSD test for multiple comparisons. Failure mode data were submitted to chi-square test. Linear regression analyses (Person correlation) were performed between: DC% vs WSp, DC% vs WSl, WSl vs WSp, and μSBS vs nanoleakage. The analyses were performed at a significance of 5%. 3. Results The pH values found for each group were: CONTROL - 4.4; ASB-BAT - 4.4; ASB-GM - 4.7; ASB-CHX - 4.2.

2.4. Nanoleakage test An additional 16 M were sectioned along the longitudinal axis of the tooth in the mesio-distal direction was made using a cutting machine (IsoMet 1000, Buëhler, Lake Bluff, IL, USA), separating the tooth into two halves. One half had the superficial dentin exposed. The other half received a cut 0.5 mm from the pulp chamber, characterizing the deep dentin. The peripheral enamel was removed using a diamond bur (#4138, KG Sorensen, Cotia, SP, Brazil). The smear layer of dentin was standardized using 400-and 600-grit SiC papers (Arotec, Cotia, SP, Brazil) in a polishing machine (DPU 10, Struers, Dinamarca) for 1 min. The dentin fragments were randomly divided into 8 groups (n = 8), according to the different MMP inhibitors and storage time, and in two subgroups according to dentin depth. Dentin surfaces were treated as previously described and 5 increments of 1 mm thick resin composite (Filtek Z250, 3M Espe, St Paul, MN, USA) were horizontally added to the bonded surfaces and individually light cured for 20s. After storage in distilled water at 37 °C for 24 h, the teeth were longitudinally sectioned in both the mesio-distal and buccal-lingual

3.1. WSp, WSl and DC% The results of WSp, WSl and DC% are summarized in Table 2. Oneway ANOVA detected significant differences among groups for WSp Table 2 Means values and standard deviations of WSp (μg/mm3), WSl (μg/mm3) and DC% of each group. Group

WSp

CONTROL ASB-BAT ASB-GM ASB-CHX

91.4 86.7 88.4 77.2

± ± ± ±

5.8 b 3.1 b 5.8 b 5.7 a

WSl

DC%

2.5 ± 1.1 a 3.4 ± 2.1 ab 8.9 ± 0.9 ab 10.8 ± 5.0 b

91.0 94.5 93.0 85.9

± ± ± ±

1.0 ab 2.2 a 0.5 a 7.4 b

For each column values with the same lowercase letters are statistically similar (Tukey's HSD, α = 0.05). WSp = water sorption; WSl = solubility; DC% = degree of conversion. 4

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

Table 3 Means and standard deviations of μSBS (MPa) after each period of storage and dentin depth. Group

CONTROL ASB-BAT ASB-GM ASB-CHX

Superficial Dentin

Deep Dentin

Immediate

12 Months

32.2 33.3 34.1 24.9

20.2 29.3 36.0 17.8

± 4.2 ABa* ± 4.0 Aa ± 6.0 Aa ± 5.2 Ba

± ± ± ±

2.4 6.3 4.3 2.7

Bb Aa Aa§ Ba

Immediate

12 Months

22.5 29.5 28.8 19.7

13.6 21.5 25.9 10.2

± 5.2 ABa* ± 1.8 Aa ± 4.6 Aa ± 4.7 Ba

± ± ± ±

3.0 8.7 5.7 2.1

CBb ABa Aa§ Cb

For each column values with the same uppercase letters are statistically similar. For each row, values with the same lowercase letters are statistically similar within the same dentin depth. Values with same symbol (* and §) are statistically different within the same evaluation time (Tukey's HSD, α = 0.05).

significant (p = 0.0587). The interactions inhibitor vs. time (p = 0.0003) and dentin depth vs. time (0.0372) were also significant. Tukey’s HSD test showed that all groups presented similar nanoleakage at the immediate time and both dentin depths. In superficial dentin after 12 months of storage, ASB-BAT and ASB-GM presented significantly lower nanoleakage when compared to those obtained by CONTROL, but were not statistically different from ASB-CHX. In deep dentin after 12 months storage, ASB-GM presented significantly lower nanoleakage than CONTROL. However ASB-GM did not differ from ASB-BAT and ASB-CHX, which were statistically similar to CONTROL. Moreover, only in the group CONTROL in superficial dentin presented increased nanoleakage after 12 months storage. Representative SEM images of nanoleakage are shown in Figs. 3 and 4. Fig. 5 shows the results of linear regression analysis, which demonstrated a significant negative correlation (p = 0.00213) between μSBS and nanoleakage (r = 0.72).

(p = 0.0009), WSl (p = 0.0191) and DC% (p = 0.0041). Tukey’s HSD test showed that ASB-CHX presented the lowest WSp. ASB-BAT and ASB-GM showed similar WSp to CONTROL. With respect to WSl, ASBCHX presented significant higher WSl when compared to CONTROL, but with no difference from ASB-GM and ASB-BAT, which were similar to CONTROL. Regarding DC%, ASB-CHX presented lower DC% when compared to ASB-BAT and ASB-GM. All groups had similar DC% when compared to CONTROL. The results of the linear regression analysis showed no statistically significant relationship between DC% vs WSl (p = 0.3927), DC% vs WSp (p = 0.2460), or WSl vs WSp (p = 0.2509). 3.2. μSBS test The results of the μSBS test are shown in Table 3. Three-way ANOVA showed a statistical significance for the three factors: time (p = 0.0000), inhibitor (p = 0.0000) and dentin depth (p = 0.0000). The interaction inhibitor vs. time was also significant (p = 0.0006), while the other interactions were not significant. The immediate μSBS of ASB-BAT, ASB-GM, and ASB-CHX did not differ from CONTROL at both dentin depths. The immediate μSBS of ASB-GM and ASB-BAT was significantly higher than ASB-CHX for superficial and deep dentin. In superficial dentin after 12 months of storage, the groups CONTROL and ASB-CHX had lower μSBS when compared to ASB-BAT and ASB-GM. The μSBS of CONTROL was reduced after 12 months storage, while ASB-BAT, ASB-GM, and ASB-CHX were statistically equal at both periods of storage. In deep dentin, the immediate μSBS of ASB-BAT and ASB-GM was significantly higher when compared to ASB-CHX, but was similar to CONTROL. After 12 months of storage, ASB-GM had the highest value of μSBS, but with no statistical difference from ASB-BAT, which did not show a significant difference from CONTROL. CONTROL was similar to ASB-BAT and ASB-CHX in deep dentin after 12 months storage. When comparing the μSBS immediately and after 12 months of storage in deep dentin, ASB-BAT and ASB-GM maintained μSBS, while CONTROL and ASB-CHX presented reduced μSBS after 12 months storage. The results showed no significant differences between the μSBS of superficial and deep dentin, except for CONTROL at immediate period and for ASB-GM after 12 months of storage, which presented greater μSBS in superficial dentin than in deep dentin. The failure mode analysis after μSBS test is presented in Fig. 2.The failure mode was predominantly adhesive for all groups. Table 4 presents the results of chi-square test. Excepting for adhesive system in deep dentin after 12 months, where was found a significant association between the failure mode and the type of adhesive system (χ2 = 14.27 / p-value = 0.027), it was not found statistical significance for all other analyzed conditions (p > 0.05).

4. Discussion In this study, the lowest WSp was observed in the ASB-CHX group, while ASB-BAT, ASB-GM, and CONTROL presented similar WSp. Chlorhexidine diacetate has more polar groups (10) to establish hydrogen bonds with water molecules than Batimastat (7) and GM1489 (7)(Table 1).(Goncalves et al., 2008; Malacarne et al., 2006)Therefore, it was not expected that ASB-CHX would present the lowest WSP. However, these results are in agreement with a previous study (da-Silva et al., 2015), which also showed the lowest WSP for an experimental adhesive system with diacetate chlorhexidine when compared to the addition of Galardin or Batimastat(da-Silva et al., 2015). The authors described that it is possible that intramolecular hydrogen bonding among –NH groups present in the chlorhexidine diacetate molecule may have reduced the amount of –NH groups prone to hydrogen bonds with water, leading to the lowest WSP observed by ASB-CHX(da-Silva et al., 2015). On the other hand, ASB-CHX presented significantly greater WSl (10.81μg/mm3) when compared to CONTROL (2.4781μg/mm3). In another study(da-Silva et al., 2015) the incorporation of chlorhexidine diacetate, Batimastat, or GM1489 did not affect the WSl of an experimental adhesive system. However, this can be due to differences in composition of adhesive systems. In the present study, the inhibitors were inserted into a commercial adhesive system which is different from the other study, which used an experimental adhesive system. Studies have suggested that the incorporation of chlorhexidine could impair the polymerization process, leading to a higher amount of nonreacted monomers, which could leech from the polymer network (Anusavice et al., 2006; Riggs et al., 2000).Although the present results did not show a significant difference between DC% for the CONTROL and ASB-CHX groups, it is possible to note that ASB-CHX presented a lower value of DC%, which was statistically lower than ASB-GM and ASB-BAT. This reduced value of DC% presented by ASB-CHX may have contributed to the higher WSL of this group, reinforcing the findings of previous studies(Anusavice et al., 2006; Riggs et al., 2000).The first hypothesis was partially accepted, since ASB-CHX presented the lowest

3.3. Nanoleakage The nanoleakage results are summarized in Table 5. Three-way ANOVA showed a statistical significance for the factors inhibitor (p = 0.0002) and time (p = 0.0000). The factor dentin depth was not 5

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

Fig. 2. Failure mode (%) of each group after each period of storage.

diacetate used was higher than the minimal inhibitory concentration (0.0001% for MMP-2, 0.002% for MMP-9 and 0.01% for MMP-8), these findings indicate that Batimastat and GM1489 were more effective in inhibiting MMP when compared to diacetate chlorhexidine in deep dentin. Thus, the second hypothesis was partially accepted: the incorporation of GM1489 improved the μSBS and decreased nanoleakage after 12 months for both dentin depths, which was similar with the incorporation of Batimastat in superficial dentin. Moreover, dentin depth only affected the ability of Chlorhexidine diacetate to maintain resin-dentin bond stability. It is most likely that this behavior is related to the MMP concentration, which may be activated by the adhesive procedure, based on the dentin depth. Previous results(Niu et al., 2011) showed that the concentration of MMP-2 and MMP-9 was significantly higher in deep dentin (5.33 ng/mg and 3.18 ng/mg respectively) than in superficial dentin (3.27 ng/mg and 0.23 ng/mg respectively). Another study also showed that the concentration of all evaluated MMP (MMP-1, -2, -3, -8, and -9) was significantly higher in deep dentin than in superficial dentin(Wang et al., 2014). Therefore, the lower effectiveness of ASB-CHX in maintaining the bond stability in deep dentin when compared to ASB-BAT and ASB-GM may be related to the greater concentration of MMP in deep dentin, to its reduced potential to inhibit MMP-3(Khaddam et al., 2014)and the greater solubility and lower DC% of adhesive systems with Chlorhexidine diacetate incorporated, which may lead to failures of the hybrid layer. The inhibitory constants (Ki) for dentin MMP may express the degree of interaction between the inhibitor and the inhibited enzyme. The

sorption and highest solubility. The presence of other inhibitors did not affect the physicochemical properties of the adhesive system. With regards to immediate μSBS, the current results revealed that the incorporation of MMP inhibitors in ASB2 did not affect the bond strength in either superficial or deep dentin. However, the immediate μSBS of ASB-CHX was significantly lower than ASB-GM and ASB-BAT in both superficial and deep dentin. It is possible that a chemical incompatibility occurred between chlorhexidine and ASB2. Another study observed a greater effectiveness of one adhesive system over another when chlorhexidine was incorporated into them(Stanislawczuk et al., 2014). After 12 months of storage, ASB-BAT and ASB-GM exhibited the highest μSBS in superficial dentin. In deep dentin, only ASB-GM was significantly better than CONTROL and ASB-CHX. These results showed that the addition of MMP inhibitors can improve the long-term dentin bonding. As ASB-CHX presented lower DC% and higher WSL than CONTROL, and none of these properties were affected for ASB-BAT and ASB-GM, it is possible that a decreased quality of hybrid layer created by ASB-CHX caused the lower μSBS for ASB-CHX when compared to ASB-BAT and ASB-GM after storage. With regards to the resin-dentin bond stability, all inhibitors were efficient in superficial dentin in maintaining dentin μSBS after storage, differently from the CONTROL group. On the other hand, only ASB-GM and ASB-BAT were efficient in maintaining dentin μSBS in deep dentin after storage. CONTROL group and ASB-CHX were not able to maintain the μSBS after storage. Although the concentration of chlorhexidine

Table 4 Results of chi sque for failure modes. Condition

(χ2)

p-value

Statistical significance

Superficial vs. Deep dentin (immediate) Superficial vs. Deep dentin (12 months) Immediate vs. 12 months Superficial dentin (Immediate vs. 12 months) Deep dentin (Immediate vs. 12 months) Adhesive systems (superficial dentin immediate) Adhesive systems (deep dentin immediate) Adhesive systems (superficial dentin 12 months) Adhesive systems (deep dentin 12 months)

1.96 1.79 3.43 1.10 1.88 6.21 4.44 1.70 14.27

0.38 0.41 0.18 0.58 0.39 0.40 0.62 0.95 0.02

n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. p < 0.05

n.s. = not significant. 6

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

Table 5 Means and standard deviations of nanoleakage (percentage of Ag) after each period of storage in distilled water. Group

CONTROL ASB-BAT ASB-GM ASB-CHX

Superficial Dentin

Deep Dentin

Immediate

12 Months

Immediate

12 Months

0.98 1.17 0.96 1.13

1.83 1.27 1.03 1.53

1.30 1.33 1.28 1.16

1.76 1.44 1.09 1.42

± ± ± ±

0.17Aa 0.26Aa 0.13Aa 0.19Aa

± ± ± ±

0.24Bb 0.12Aa 0.09Aa 0.04ABa

± ± ± ±

0.24Aa 0.25Aa 0.15Aa 0.21Aa

± ± ± ±

0.15Ba 0.23ABa 0.08Aa 0.10ABa

For each column values with the same uppercase letters are statistically similar. For each row, values with the same lowercase letters are statistically similar within the same dentin depth (Tukey's HSD, α = 0.05).

that the bond strength of superficial dentin was higher than in deep dentin (Pegado et al., 2010; Zhang et al., 2014) and other studies found similar bond strengths in superficial and deep dentin(Zheng et al., 2009). In this study, when inhibitors were inserted in the adhesive system, an increase in μSBS to deep dentin was observed, making the μSBS to deep dentin similar to superficial dentin (except for CONTROL group at immediate time and ASB-GM after twelve months of storage). The nanoleakage results revealed that there was no difference between superficial and deep dentin, independent of storage time and inhibitor used. In the immediate nanoleakage evaluation, the incorporation of MMP inhibitors did not affect the results for both dentin depths. In the superficial dentin after twelve months of storage, ASBBAT and ASB-GM exhibited the lowest nanoleakage (lowest resindentin degradation). Meanwhile, only ASB-GM was significantly better than CONTROL group in deep dentin. These findings are in accordance with those found in the μSBS tests, showing a superior performance from ASB-BAT and ASB-GM in superficial dentin and from ASB-GM in deep dentin. The results of nanoleakage presented significant correlation with results of μSBS (r = 0.72). The rationale for the lower nanoleakage in ASB-BAT and ASB-GM in superficial dentin and ASB-GM in deep dentin is probably the same as discussed for the μSBS results. All groups maintained similar nanoleakage after 12 months storage at both dentin depths, except for CONTROL group in superficial dentin. Although nanoleakage increased in deep dentin after storage, it was not

Ki of Batimastat is: MMP- 2 = 4 nM, MMP-3 = 20 nM and MMP9 = 4 nM and the Ki of GM1489 is: MMP-2 = 500 nM, MMP3 = 20 μM, MMP-8 = 100 nM and MMP-9 = 100 nM(da-Silva et al., 2015; www.merckmillipore.com/BR/pt/product/GM-1489 ). When comparing this information with the results found in the present study, it can be observed that, although Batimastat has no inhibitory effect against MMP-8, it was able to maintain bonding stability over twelve months, which is in accordance with a previous study(da-Silva et al., 2015). Thus, it is reasonable to suppose that the effect of MMP-8 on resin-dentin degradation may be less aggressive than other dentin MMP. However, other factors, such as the availability of the MMP inhibitors within the polymer chain, may be factors for dentin-resin bond stability. When comparing different dentin depths within each time of storage, the present study demonstrated that, although there is a tendency for greater values of bond strength in superficial dentin when compared to deep dentin, this difference was only significant for the CONTROL group at the immediate time and the ASB-GM group after twelve months of storage. The difference observed for ASB-GM after twelve months can be attributed mainly to it is greater performance in superficial dentin, since this group had the highest value (35.98 MPa). When comparing these results with those found in the literature(Pegado et al., 2010) it is possible to note that the bonding performance at different dentin depths is controversial. Some studies have reported

Fig. 3. Representative back-scattering SEM images of the resin-dentin interfaces in superficial dentin according to inhibitor and storage time. In immediate evaluation, all groups presented low nanoleakage: 3A- Control; 3B- ASB-BAT; 3C- ASB-GM; 3D- ASB-CHX. After 12 months of storage, groups ASB-BAT and ASB-GM presented lower nanoleakage than CONTROL: 3E- Control; 3F- ASB-BAT; 3G- ASB-GM; 3H- ASB-CHX. 7

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

Fig. 4. Representative back-scattering SEM images of the resin-dentin interfaces in deep dentin according to inhibitor and storage time. In immediate evaluation, nanoleakage of all groups was similar: 4A- Control; 4B- ASB-BAT; 4C- ASB-GM; 4D- ASB-CHX. After 12 months of storage, ASB-GM presented the lowest nanoleakage. 4E- Control; 4F- ASB-BAT; 4G- ASB-GM; 4H- ASB-CHX.

incorporation of MMP inhibitors in bond stability to superficial and deep dentin, it still presents some limitations because it is an in vitro study in which dentin slices were used instead intact or perfused teeth. These aspects should be considered in future investigations. 5. Conclusions Based on the present results, it can be concluded that the incorporation of Batimastat and GM1489 did not affect the DC%, WSp and WSl of ASB2, while the incorporation of chlorhexidine diacetate decreased WSp and increased WSl. The addition of MMP inhibitors to ASB2 did not jeopardize the immediate bond strength at any dentin depth. After twelve months, GM1489 was effective at improving the resin-dentin bond strength and decreases the nanoleakage at both dentin depths, while Batimastat was effective for superficial dentin only. The incorporation of MMP inhibitors equalized the immediate bond strength of the deep dentin to superficial dentin. Moreover, bond stability to superficial dentin was improved with the incorporation of MMP inhibitors. In deep dentin, only Batimastat and GM1489 maintained the resin-dentin bond strength.

Fig. 5. Linear regression of μSBS plotted against nanoleakage.

statistically significant. The quality of sealing against nanoleakage may be influenced by the heterogeneous character of the dentin structure, mainly due to tubule orientation(Yuan et al., 2007). This could have contributed to the most remarkable differences found in superficial dentin nanoleakage between immediate testing and after twelve months of storage. Based on the results of this study, the incorporation of GM1489 and Batimastat can be indicated because they maintain μSBS to superficial and deep dentin after storage, without jeopardizing the physical-mechanical properties. Additionally, chlorhexidine diacetate affected the physical-mechanical properties of the adhesive system and was not efficient at maintaining the μSBS to deep dentin. Furthermore, GM 1489 merits special mention because, in the most critical situation (deep dentin, after 12 months storage), ASB-GM was the only inhibitor with significantly greater μSBS and lower nanoleakage when compared to CONTROL group. Although the present study adds new aspects about the effect of

Conflicts of interest The authors declare that they have no conflict of interest. Funding This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) Finance code 001. References Acharya, M.R., Venitz, J., Figg, W.D., Sparreboom, A., 2004. Chemically modified tetracyclines as inhibitors of matrix metalloproteinases. Drug Resist. Updates 7, 195–208. Almahdy, A., Koller, G., Sauro, S., Bartsch, J.W., Sherriff, M., Watson, T.F., Banerjee, A.,

8

Journal of the Mechanical Behavior of Biomedical Materials 100 (2019) 103402

F.S. Simmer, et al.

layer detected with in situ zymography. J. Dent. Res. 91, 467–472. www.merckmillipore.com/br/Pt/Product/Gm-1489. Mobarak, E.H., El-Badrawy, W., Pashley, D.H., Jamjoom, H., 2010. Effect of pretest storage conditions of extracted teeth on their dentin bond strengths. J. Prosthet. Dent 104, 92–97. Niu, L.N., Zhang, L., Jiao, K., Li, F., Ding, Y.X., Wang, D.Y., Wang, M.Q., Tay, F.R., Chen, J.H., 2011. Localization of MMP-2, MMP-9, TIMP-1, and TIMP-2 in human coronal dentine. J. Dent. 39, 536–542. Pashley, D.H., Carvalho, R.M., 1997. Dentin permeability and dentin adhesion. J. Dent. 25, 355–372. Pashley, D.H., Tay, F.R., Yiu, C.K., Hashimoto, M., Breschi, L., Carvalho, R.M., Ito, S., 2004. Collagen degradation by host-derived enzymes during aging. J. Dent. Res. 83, 216–221. Pashley, D.H., Tay, F.R., Breschi, L., Tjaderhane, L., Carvalho, R.M., Carrilho, M., Tezvergil-Mutluay, A., 2011. State of the art etch-and-rinse adhesives. Dent. Mater. 27, 1–16. Pegado, R.E.F., Amaral, F.L.B., Florio, F.M., Basting, R.T., 2010. Effect of different bonding strategies on adhesion to deep and superficial permanent dentin. Eur. J. Dermatol. 4, 110–117. Riggs, P.D., Braden, M., Patel, M., 2000. Chlorhexidine release from room temperature polymerising methacrylate systems. Biomaterials 21, 345–351. Stanislawczuk, R., Pereira, F., Munoz, M.A., Luque, I., Farago, P.V., Reis, A., Loguercio, A.D., 2014. Effects of chlorhexidine-containing adhesives on the durability of resindentine interfaces. J. Dent. 42, 39–47. Thompson, J.M., Agee, K., Sidow, S.J., Mcnally, K., Lindsey, K., Borke, J., Elsalantry, M., Tay, F.R., Pashley, D.H., 2012. Inhibition of endogenous dentin matrix metalloproteinases by ethylenediaminetetraacetic acid. JOE 38, 62–65. Tjaderhane, L., Buzalaf, M.A., Carrilho, M., Chaussain, C., 2015. Matrix metalloproteinases and other matrix proteinases in relation to cariology: the era of 'dentin degradomics. Caries Res. 49, 193–208. Visse, R., Nagase, H., 2003. Matrix metalloproteinases and tissue inhibitors of metalloproteinases: structure, function, and biochemistry. Circ. Res. 92, 827–839. Wang, D., Zhang, L., Li, F., Ma, K., Chen, J., 2014. Localization and quantitative detection of matrix metalloproteinase in human coronal dentine. Zhonghua Kou Qiang Yi Xue Za Zhi 49, 688–692. Yiu, C.K., Hiraishi, N., Tay, F.R., King, N.M., 2012. Effect of chlorhexidine incorporation into dental adhesive resin on durability of resin-dentin bond. J. Adhesive Dent. 14, 355–362. Yuan, Y., Shimada, Y., Ichinose, S., Sadr, A., Tagami, J., 2007. Effects of dentin characteristics on interfacial nanoleakage. J. Dent. Res. 86, 1001–1006. Zhang, S., Kern, M., 2009. The role of host-derived dentinal matrix metalloproteinases in reducing dentin bonding of resin adhesives. Int. J. Oral Sci. 1, 163–176. Zhang, L., Wang, D.Y., Fan, J., Li, F., Chen, Y.J., Chen, J.H., 2014. Stability of bonds made to superficial vs. deep dentin, before and after thermocycling. Dent. Mater. 30, 1245–1251. Zheng, T.L., Huang, C., Zheng, Z.X., 2009. Influence of different dentin depths on microtensile bond strength of two dentin adhesive systems. Shang Hai Kou Qiang Yi Xue 18, 532–535. Zhou, J., Tan, J., Chen, L., Li, D., Tan, Y., 2009. The incorporation of chlorhexidine in a two-step self-etching adhesive preserves dentin bond in vitro. J. Dent. 37, 807–812.

2012. Effects of MMP inhibitors incorporated within dental adhesives. J. Dent. Res. 91, 605–611. Anusavice, K.J., Zhang, N.Z., Shen, C., 2006. Controlled release of chlorhexidine from UDMA-TEGDMA resin. J. Dent. Res. 85, 950–954. Armstronga, S.R., Kellerb, J.C., Boye, D.B., 2001. The influence of water storage and Cfactor on the dentin-resin composite microtensile bond strength and debond pathway utilizing a filled and unfilled adhesive resin. Dent. Mater. 17, 268–276. Brackett, M.G., Tay, F.R., Brackett, W.W., Dib, A., Dipp, F.A., Mai, S., Pashley, D.H., 2009. In vivo chlorhexidine stabilization of hybrid layers of an acetone-based dentin adhesive. Oper. Dent. 34, 379–383. Breschi, L., Mazzoni, A., Ruggeri, A., Cadenaro, M., Di-Lenarda, R., Dorigo, E.D.S., 2008. Dental adhesion review: aging and stability of the bonded interface. Dent. Mater. 24, 90–101. Breschi, L., Mazzoni, A., Nato, F., Carrilho, M., Visintini, E., TjäDerhane, L., Ruggeri-Jr, A., Tay, F., Dorigo, E., Pashley, D., 2010a. Chlorhexidine stabilizes the adhesive interface: a 2 year in vitro study. Dent. Mater. 26, 320–325. Breschi, L., Mazzoni, A., Nato, F., Carrilho, M., Visintini, E., TjäDerhane, L., Ruggeri-Jr, A., Tay, F.R., Dorigo, E.D.S., Pashley, D.H., 2010b. Use of a specific MMP inhibitor (Galardin) for preservation of hybrid layer. Dent. Mater. 26, 571–578. Da-Silva, E.M., De-Sá-Rodrigues, C.U.F., Matos, M.P.O., De-Carvalho, T.R., Dos-Santos, G.B., Amaral, C.M., 2015. Experimental etch-and-rinse adhesive systems containing MMP-inhibitors: physicochemical characterization and resin-dentin bonding stability. J. Dent. 43, 1491–1497. De-Munck, J., Van-Landuyt, K., Peumans, M., Poitevin, A., Lambrechts, P., Braem, M., Meerbeek, B.V., 2005. A critical review of the durability of adhesion to tooth tissue: methods and results. J. Dent. Res. 84, 118–132. Devito-Moraes, A.G., Francci, C., Vidal, C.M., Scaffa, P.M., Nesadal, D., Yamasaki, L.C., Nicolau, J., Nascimento, F.D., Pashley, D.H., Carrilho, M.R., 2016. Phosphoric acid concentration affects dentinal MMPs activity. J. Dent. 53, 30–37. Dewald, J.P., 1997. The use of extracted teeth for in vitro bonding studies: a review of infection control considerations. Dent. Mater. 13, 74–81. Goncalves, L., Noronha-Filho, J.D., Guimaraes, J.G., Poskus, L.T., Da-Silva, E.M., 2008. Solubility, salivary sorption and degree of conversion of dimethacrylate-based polymeric matrixes. J. Biomed. Mater. Res. B Appl. Biomater. 85, 320–325. Khaddam, M., Salmon, B., Le-Denmat, D., Tjaderhane, L., Menashi, S., Chaussain, C., Rochefort, G.Y., Boukpessi, T., 2014. Grape seed extracts inhibit dentin matrix degradation by MMP-3. Front. Physiol. 5, 425. Liu, Y., Tjaderhane, L., Breschi, L., Mazzoni, A., Li, N., Mao, J., Pashley, D.H., Tay, F.R., 2011. Limitations in bonding to dentin and experimental strategies to prevent bond degradation. J. Dent. Res. 90, 953–968. Malacarne, J., Carvalho, R.M., De-Goes, M.F., Svizero, N., Pashley, D.H., Tay, F.R., Yiu, C.K., Carrilho, M.R., 2006. Water sorption/solubility of dental adhesive resins. Dent. Mater. 22, 973–980. Marshall-Jr, G.W., Marshall, S.J., Kinney, J.H., Balooch, M., 1997. The dentin substrate: structure and properties related to bonding. J. Dent. 25, 441–458. Mazzoni, A., Carrilho, M., Papa, V., TjäDerhane, L., Gobbi, P., Nucci, C., Di-Lenarda, R., Mazzotti, G., Tay, F.R., Pashley, D.H., Breschi, L., 2011. MMP-2 assay within the hybrid layer created by a two-step etch-and-rinse adhesive: biochemical and immunohistochemical analysis. J. Dent. 39, 470–477. Mazzoni, A., Nascimento, F.D., Carrilho, M., Tersariol, I., Papa, V., Tjaderhane, L., DiLenarda, R., Tay, F.R., Pashley, P.H., Breschi, L., 2012. MMP activity in the hybrid

9