Effect of Different Matrix Metalloproteinase Inhibitors on Microtensile Bond Strength of an Etch-and-Rinse and a Self-etching Adhesive to Dentin

Matrix metalloproteinases (MMPs) and cysteine cathepsins in dentin and dentinal fluid contribute to the enzymatic degradation of the adhesive hybrid layer and thus to the reduction of bond durability over time. pH changes caused by acid etching or acidic monomers and the adhesive resin monomers themselves can modulate the activation and expression of MMPs and cysteine cathepsins, resulting in an increased digestion of collagen within the hybrid layer.^1,2 

Different strategies have aimed to improve the bond durability by applying enzyme inhibitors as a pretreatment before resin infiltration or by admixing enzyme inhibitors to primers. It has been shown that chlorhexidine digluconate (CHX) can inhibit³  MMP-2, −8, and −9 and of cysteine cathepsins,⁴  preserving bond strength of etch-and-rinse and self-etching adhesives over time.^5-7 

As the application of 2% CHX on phosphoric acid-etched dentin is the only application procedure that has been tested clinically and shown to prevent bond strength loss in vivo,^8-10  this procedure can be considered as a kind of gold standard for maintaining hybrid layer stability. However, in recent studies, other potential MMP inhibitors, such as galardin, metal ions, or green tea catechins, were tested, which were suggested to be as or even more effective than CHX.^11,12  Similar to CHX, galardin as a synthetic inhibitor of MMP-1, −2, −3, −8, and −9 is acting as a zinc chelator.¹³  Green tea catechins, mainly epigallocatechin-gallate (EGCG), change the secondary structure of collagenases by hydrogen bonding and hydrophobic interactions.¹⁴  As well, binding of metal ions might cause conformational changes that inactivate their catalytic function.¹⁵ 

However, to date, the efficacy of these MMP inhibitors to prevent dentin bond strength loss over time has not been determined. Thus, the aim of the present study was to investigate the microtensile bond strength (microTBS) of an etch-and-rinse (Optibond FL) and a self-etching adhesive (Clearfil SE Bond), both of which have been shown repeatedly to be excellent adhesives in their respective class, after pretreatment with different MMP inhibitors.

This study tested the null hypothesis that the microTBS of an etch-and-rinse and a self-etching adhesive after 9 months of storage in artificial saliva is not affected by pretreatment with water-based solutions containing 2% CHX, 0.05% green tea extract, 1 mM FeSO₄, or 0.2 mM galardin.

Forty extracted, sound human molars were selected for this study. The teeth were stored in 0.01% (w/v) thymol at 4°C and were used within four weeks after extraction. Extracted teeth were collected as anonymous by-products of regular therapy. Because of that, our Medical Ethical Board stated that the performed research did not fall under the regulations of the “Act on Medical Research Involving Human Subjects” (METc 2009.305).

The teeth were embedded in circular molds with self-curing acrylic resin (Paladur, Heraeus Kulzer, Germany). A flat, mid-coronal dentin surface was prepared by means of a water-cooled, low-speed diamond saw (Isomet 1000, Buehler Ltd, Lake Bluff, IL). Polishing with wet 600-grit SiC paper (Water Proof Silicon Carbide Paper, Stuers, Erkrat, Germany) created a standardized smear layer. The dentin surfaces were verified for the absence of pulp chamber exposition using a stereomicroscope (Stemi 2000, Carl Zeiss, Feldbach, Switzerland).

The teeth were allocated randomly to test (each n=3) and control (n=5) groups of the two adhesives.

Bonding of dentin surfaces was performed using an etch-and-rinse (Optibond FL, Kerr, Scafati, Italy) or a self-etching adhesive (Clearfil SE Bond, Kuraray, Tokyo, Japan). The two adhesives were applied according to the manufacturer's instructions (Table 1) with or without the additional application of one of the different MMP inhibitor solutions: 1) CHX (2%), 2) green tea extract (0.05%), ferrous sulfate (1 mM), or 4) galardin (0.2 mM; Table 2).

To analyze the initial microTBS and the bond strength after aging (microTBS after 9 months), two groups of each adhesive were not treated with any MMP inhibitor. The remaining groups were treated with one of the MMP inhibitors and tested only after aging, as the immediate bond strength between the untreated controls and test groups was shown to be not different.^11,20 

The application of the MMP inhibitor solutions was performed with a microbrush for 60 seconds under a slight rubbing motion. The excess was removed using an absorbing paper, leaving the dentin surface moist.

After light-curing of the bonding, a composite buildup (CeramX Mono, A3, Dentsply, Konstanz, Germany) was made in five 1 mm increments, each light-cured for 40 seconds at 800 mW/cm² (bluephase, IvoclarVivadent, Schaan, Liechtenstein).

After 24-hour storage in water at 37°C, the teeth were sectioned perpendicular to the interface with a water-cooled diamond saw (Struers-Accutom 50, Struers, Denmark, and MOD 10, Struers, Denmark) to obtain rectangular beams of approximately 1 mm². The interfaces were precisely checked under a stereomicroscope to examine whether enamel remained. Sticks with remaining enamel were discarded, while all other sticks from one tooth were used. The beams of the control group were tested immediately (initial microTBS), while the beams of test groups were stored in artificial saliva²¹  at 37°C for 9 months. Thereby, all sticks from one tooth were stored in 3 mL of artificial saliva.

The artificial saliva contained 0.7 mmol/L CaCl₂, 0.2 mmol/L MgCl₂·6H₂O, 4.0 mmol/L KH₂PO₄, 30 mmol/L KCl, 0.3 mmol/L NaN₃, and HEPES buffer (all reagents purchased by Merck, Darmstadt, Germany) and was renewed weekly.

The tensile load of the beams was tested in a universal testing machine (Z010, Zwick/Roell, Ulm, Germany) with a 200 N load cell (KAF-TC, A.S.T., Dresden, Germany) at a crosshead speed of 0.15 mm/min. Therefore, the nontrimmed beams were fixed to a sandblasted microtensile bond jig with cyanoacrylate glue (Model Repair II Blue, Dentsply-Sankin, Tochigi, Japan) and tested under tensile force in a top-bottom manner described previously.²²  The dimensions of the beams were measured with a digital caliper to calculate the exact cross-sectional area.

The bond strength values (MPa) were determined by dividing the imposed force (N) at the time of fracture by the bond area (mm²).

In addition, each beam was observed under a stereomicroscope (Wild M8, WILD HEERBRUGG, Heerbrugg, Switzerland) at 50× magnification to determine the mode of failure, classified as adhesive (A), cohesive in dentin (CD), composite (CC), or mixed (M).

The number of sticks and pretest failures in each group is given in Table 3.

The mean microTBS (±standard deviation) for each group was computed. Kolmogorov-Smirnov and Shapiro-Wilk tests were applied to check the assumption of normality. Normal distribution was found in all groups, and data were further analyzed by linear mixed-model regression, separately for each adhesive, to account for the fact that multiple beams were gained from the same tooth and, therefore, were not independent from each other.

Relative frequencies of adhesive, mixed, and cohesive failures were calculated and analyzed by repeated-measures analysis of variance (ANOVA) followed by Greenhouse-Geyser correction using the adhesive and the test groups as independent covariables. As no significant differences with respect to the adhesives were found, differences between the test groups were statistically analyzed by one-way ANOVAs and Scheffé post hoc tests separately for each failure type.

The level of significance was set at p<0.05.

The microTBS of the etch-and-rinse adhesive was significantly decreased after 9 months of storage in artificial saliva (p=0.01). Pretreatment with all MMP inhibitors except FeSO₄ revealed bond strength values similar to the initial microTBS and significantly higher compared with the aged group without pretreatment (9-month microTBS, p<0.013).

The microTBS of the self-etching adhesive declined only slightly but not significantly over time (p=0.49). Bond strength values after pretreatment with the different MMP inhibitors were not different from the initial microTBS nor from the 9-month microTBS data (p>0.359; Table 3).

The failure mode analyses revealed significant differences between the test groups for adhesive (p=0.003) and mixed (p=0.04) but not for cohesive failures (dentin: p=0.749; composite: p=0.528). Overall, the frequency of adhesive failures in groups without MMP inhibitors was significantly (p=0.008) increased by aging. Application of the MMP inhibitors decreased the number of adhesive failures, but this difference was not significant with respect to the untreated groups after 9 months of aging (p>0.34). However, the relative frequencies of adhesive failures after pretreatment with CHX and green tea extract were not significantly (p>0.07) different from the untreated groups tested immediately. Post hoc comparisons of mixed failures showed no significant differences between the test groups (p>0.061).

The null hypothesis that the microTBS after 9 months of storage in artificial saliva is not affected by pretreatment with different MMP inhibitors was corroborated for the self-etch adhesive but not for the etch-and-rinse adhesive. Pretreatment with all MMP inhibitors except FeSO₄ prevented microTBS loss of Optibond FL significantly and equally effectively.

In the present study, the MMP inhibitor solutions were applied in a separate step in addition to the three-step or two-step application procedure of the etch-and-rinse or the self-etching adhesive, respectively. This approach might increase the overall chair time for a composite restoration and might be less feasible for self-etching adhesives because of the synchronous demineralization and infiltration during treatment. However, the application time of the MMP inhibitor solutions was limited to 60 seconds, which seems realizable under clinical conditions. Moreover, the incorporation of MMP inhibitors into dental adhesives might affect the mechanical properties of the products by decreasing the degree of conversion and the E-modulus.²³ 

The initial microTBS found for Optibond FL and Clearfil SE Bond correlates with the results found previously.^24,25  Both adhesives have consistently shown favorable bonding performance in different protocols for microTBS measurement. As the bonding performance of Optibond FL and Clearfil SE Bond was shown to be hardly affected by the cavity configuration²⁶  and dentin location,²⁵  the setup of the present study was simplified by using flat dentin surfaces. Flat dentin surfaces allowed for the use of multiple sticks per tooth; thus, the overall sample size could be increased. The sticks were aged by direct exposure to artificial saliva, which is a quicker aging strategy than aging intact bonded teeth.²⁷ 

Both adhesives exhibited lower microTBS values after aging, but this effect was significant only for the etch-and-rinse adhesive. Moreover, the fracture analysis revealed predominantly mixed adhesive-cohesive failure patterns with a clear tendency to fail more at the interface after 9 months of storage in the case in which no MMP inhibitor was applied. These results are in accordance with previous studies indicating a more rapid destruction of hybrid layers for three-step etch-and-rinse adhesives compared with mild two-step self-etching adhesives.^20,25  Although adhesive hydrophilicity and water sorption of adhesive interfaces are still considered the principal mechanisms of the resin-bond degradation, enzymatic degradation of the hybrid layer by MMPs contributes to the degradation process and loss of bond strength over time. The higher levels of MMP-2 and MMP-9 activity demonstrated for etch-and-rinse compared with self-etching adhesives^2,20  might therefore partly explain why the bond durability of Optibond FL is affected to a higher extent than that of the Clearfil SE Bond. However, it is likely that with longer aging times, the effect of enzymatic degradation might become more apparent for Clearfil SE Bond as well.²⁰ 

For both adhesives, application of the MMP inhibitors resulted in microTBS values similar to the initial bond strength. It is likely that this effect is related to the reduction of enzymatic dentin degradation. In a previous study, the degradation of demineralized dentin after application of 0.012% CHX, 400 μM EGCG, and 1 mM FeSO₄ was directly determined by assaying hydroxyproline. All test agents reduced enzymatic collagen degradation distinctly but were not significantly different from each other.¹² 

As the catechin concentration in the green tea extract solution is about 30%,²⁸  the inhibitory doses reported for EGCG and ECG on MMP-2 and MMP-9 were distinctly exceeded.¹⁴  EGCG concentrations between 0.02% and 0.5% were shown to prevent bond strength loss of an etch-and-rinse adhesive over 6 months,²⁹  thereby being as effective as a 2% chlorhexidine solution. Because of the low toxicity and the anti-inflammatory potential, polyphenol catechins can be considered as very biocompatible, with the MMP inhibitor as a possible rewetting agent.

Pretreatment with the galardin solution was as effective as chlorhexidine in preserving bond strength over time, which supports the results of a previous study, in which 0.2 mM galardin solution was used as rewetting agent for an etch-and-rinse adhesive.¹¹  In contrast to these results, galardin-modified adhesives at lower concentration were shown to be unable to prevent bond strength loss after three months of storage time.³⁰ 

Although ferrous sulfate was applied in a concentration previously shown to inhibit MMP-2 and MMP-9 activity and to reduce erosive dentin loss probably by inhibiting enzymatic degradation of the collagenous dentin matrix,¹⁸  it was unable to inhibit bond strength loss significantly. The inhibitory action of ferrous sulfate on MMPs might be partly abrogated by the fact that precipitates formed on the dentin surface and in the dentin tubules^18,31  act as a mechanical barrier hampering penetration of the adhesive. Moreover, ferric sulfate might induce some dentin discoloration,¹⁸  which might affect the color of a composite restoration negatively.

The MMP inhibitors chlorhexidine, green tea extract, and galardin preserved the microTBS of Optibond FL but not of Clearfil SE Bond significantly when aged for 9 months in artificial saliva. In the case of ferric sulfate application, bond strength was not affected negatively over time. It remains to be determined whether these MMP inhibitors will prevent bond strength loss in long-term experiments.

The authors of this manuscript certify that they have no proprietary, financial, or other personal interest of any nature or kind in any product, service, and/or company that is presented in this article.