0.1% EGCG/DWB and 1%EGCG/DWB preserved dentin bond strength and reduced interfacial nanoleakage.
1%EGCG/DWB inhibited S. mutans growth and biofilm formation along the dentin–adhesive interface.
Adjunctive application of DMSO wet-bonding and EGCG achieves desirable dentin bonding performance and prevents secondary caries.
To determine whether dentin–adhesive interface stability would be improved by dimethyl sulfoxide (DMSO) wet-bonding and epigallocatechin-3-gallate (EGCG).
Etched dentin surfaces from sound third molars were randomly assigned to five groups according to different pretreatments: group 1, water wet-bonding (WWB); group 2, 50% (v/v) DMSO wet-bonding (DWB); groups 3–5, 0.01, 0.1, and 1 wt% EGCG-incorporated 50% (v/v) DMSO wet-bonding (0.01%, 0.1%, and 1%EGCG/DWB). Singlebond universal adhesive was applied to the pretreated dentin surfaces, and composite buildups were constructed. Microtensile bond strength (μTBS) and interfacial nanoleakage were respectively examined after 24 h water storage or 1-month collagenase ageing. In situ zymography and Streptococcus mutans ( S. mutans ) biofilm formation were also investigated.
After collagenase ageing, μTBS of groups 4 (0.1%EGCG/DWB) and 5 (1%EGCG/DWB) did not decrease ( p > 0.05) and was higher than that of the other three groups ( p < 0.05). Nanoleakage expression of groups 4 and 5 was less than that of the other three groups ( p < 0.05), regardless of collagenase ageing. Metalloproteinase activities within the hybrid layer in groups 4 and 5 were suppressed. Furthermore, pretreatment with 1%EGCG/DWB (group 5) efficiently inhibited S. mutans biofilm formation along the dentin–adhesive interface.
This study suggested that the synergistic action of DMSO wet-bonding and EGCG can effectively improve dentin–adhesive interface stability. This strategy provides clinicians with promising benefits to achieve desirable dentin bonding performance and to prevent secondary caries, thereby extending the longevity of adhesive restorations.
To minimize tooth preparation and acquire satisfactory aesthetic outcome, adhesive technique has become one of the greatest inventions in clinical dentistry in the 20th century. The durability of dentin bonding, however, remains limited despite its immediate efficiency [ ]. Insufficient bonding durability may lead to frequent replacement of restorations and extra cost [ ]. Therefore, effective measures to improve dentin bonding durability are urgently needed, thus extending the service life of adhesive restorations.
The decline of dentin bond strength is principally caused by the degradation of the hybrid layer at dentin–adhesive bonding interfaces [ ]. Evidence has proven that adhesive hydrolysis, enzymolysis from cysteine cathepsin and matrix metalloproteinases (MMPs), inadequate penetration of resin monomers, and secondary caries are potential threatens for hybrid layer degradation [ , ]. Strategies, including application of collagen cross-linkers, MMP inhibitors, ethanol wet-bonding, and biomimetic remineralization, have been designed to preserve the integrity of dentin–adhesive interfaces for reliable bonding stability [ , ].
Wettability plays a vital role in facilitating dentin bonding performance [ ]. Water wet-bonding method was once widely used in dentin bonding [ ]. However, controlling the humidity is technology-sensitive because excessive wetting or drying may cause a negative effect on bond strength [ ]. Ethanol wet-bonding was introduced since it enables ethanol to replace water in dentin matrix and support demineralized collagen fibers, thereby prompting the penetration of hydrophobic adhesive monomers and avoiding collagen collapse to achieve desirable bonding efficiency [ ]. Whereas, ethanol has high vapor pressure, the replacement of adequate water is time-consuming, which seems impractical in clinical applications [ ].
Dimethyl sulfoxide (DMSO), an aprotic polar solvent with a highly polar S O group and two hydrophobic CH 3 groups, can dissolve polar and nonpolar compounds [ ]. Compared with ethanol, DMSO possesses lower volatility and stronger permeability. These characteristics compensate for the high technique susceptibility of ethanol wet-bonding. DMSO can dissociate the highly cross-linked dentin collagen into a discrete fibril network, thereby improving the wettability of demineralized dentin and the penetration of monomer into etched dentin matrix and re-expanding collapsed collagen to a moderate level [ , ]. Previous studies have proven that the application of a 50% (v/v) DMSO aqueous solution can significantly improve the dentin bonding durability and hybrid layer stability [ , ]. These findings suggest DMSO as an ideal solvent to achieve long-term dentin bond strength.
Considerable attention has been focused on MMP inhibitors in recent decades because of their clinical feasibility and ability to prevent the degradation of collagen fibrils with incomplete resin infiltration [ ]. Owing to concerns on drug resistance and potential cytotoxicity of chemical synthetics, such as chlorhexidine [ , ], naturally sourced materials are highly desired and widely investigated. Epigallocatechin-3-gallate (EGCG), a sort of green tea extract, has become popular on account of its anti-inflammatory, antioxidant, antibacterial, and anticancer efficacy [ ]. EGCG possesses favorable biocompatibility and suppresses MMP-2 and MMP-9 activities to prevent dentin collagen degradation [ ]. The formation of cariogenic bacteria biofilm can be effectively inhibited by EGCG [ , ], which is beneficial to the stability of the dentin–adhesive bonding interfaces. However, the biological functions of EGCG may be limited because of its low solubility in water [ ]. Although EGCG is more soluble in ethanol than in water, the technical sensitivity of ethanol wet-bonding and the rapid volatilization of ethanol when exposed to air remain a conundrum. Fortunately, DMSO is of particularly low volatility and EGCG can be well dissolved in DMSO aqueous solution at room temperature.
To maximize the efficacy of DMSO and EGCG in stabilizing the dentin–adhesive bonding interfaces, the application of DMSO wet-bonding combined with EGCG may be a promising strategy for improving dentin bonding durability and preventing secondary caries. To the best of our knowledge, no relevant information is available.
Therefore, the present study aimed to investigate the synergistic effects of DMSO wet-bonding and EGCG on improving the dentin–adhesive interface stability. The null hypotheses were that the adjunctive application of DMSO wet-bonding and EGCG on dentin (i) would not affect the immediate or long-term dentin bonding performance and (ii) would not reduce bacterial biofilm growth along the dentin–adhesive interface.
Materials and methods
Specimen preparation and bonding protocols
Sixty-five sound third molars were obtained under an informed consent from the donors reviewed and approved by the Ethics Committee for Human Studies of the School and Hospital of Stomatology, Wuhan University, China [no. 2011(067)]. These teeth were preserved in 0.5% (w/v) thymol solution at 4 °C before use. A 50% (v/v) DMSO aqueous solution was immediately prepared before use by mixing DMSO (Sigma-Aldrich, St. Louis, MO, USA) in sterilized deionized water. EGCG powder (Sigma-Aldrich, St. Louis, MO, USA) was freeze-dried and dissolved into the 50% (v/v) DMSO aqueous solution to obtain four experimental solutions at concentrations of 0, 0.01, 0.1, and 1 wt%, respectively.
Thirty of the sixty-five teeth were sectioned below the enamel-dentinal junction by using a diamond saw (Isomet, Buehler Ltd., Lake Bluff, IL, USA) with low speed to expose flat mid-coronal dentin surfaces under water-cooling. These dentin surfaces were burnished by using a 600-grit silicon carbide paper for 1 min under water irrigation to yield a standardized smear layer. After etched with 35% phosphoric-acid gel (3 M ESPE, St. Paul, MN, USA) for 15 s, each dentin surface was thoroughly sprayed with deionized water and blot-dried. These teeth were randomly assigned to five groups ( n = 6 teeth each group) based on different treatment protocols:
Group 1: deionized water (water wet-bonding, WWB group).
Group 2: 50% (v/v) DMSO aqueous solution (DMSO wet-bonding, DWB group).
Group 3: 0.01 wt% EGCG-incorporated 50% (v/v) DMSO aqueous solution (0.01%EGCG/DWB group).
Group 4: 0.1 wt% EGCG-incorporated 50% (v/v) DMSO aqueous solution (0.1%EGCG/DWB group).
Group 5: 1 wt% EGCG-incorporated 50% (v/v) DMSO aqueous solution (1%EGCG/DWB group).
The dentin surface of each group was pretreated with the corresponding solution via a microbrush for 60 s, respectively, and then blot-dried with filter papers to generate a visible moist surface which liquids were no longer able to remove. The specimens were bonded with a Singlebond Universal (3 M ESPE, St. Paul, MN, USA) adhesive in accordance with the manufacturer’s instructions by the same proficient dentist. After 15 s polymerization by a Bluephase Style light-curing unit (Ivoclar-Vivadent, Amherst, NY, USA), build-ups of resin composite (Charisma, Haraeus Kulzer, Hanau, Germany) were constructed on top of the bonded surfaces in four 1-mm increments, and each increment was light-cured for 20 s.
Microtensile bond strength (μTBS) test
After 24 h storage in deionized water at 37 °C, the bonded teeth were sectioned perpendicular to the bonding interfaces into slabs of 0.9 mm thickness. Six middle slabs randomly selected from each group were stored for interfacial nanoleakage assessment ( n = 4 slabs each group) and in situ zymography of the hybrid layer ( n = 2 slabs each group). The remaining slabs were further sliced into beams of 0.9 mm × 0.9 mm. Unqualified beams either situated peripherally or accompanied enamel residual were excluded, and ten qualified beams were selected from each tooth. Five of them ( n = 30 each group) were immediately tested for μTBS, and the other five ( n = 30 each group) were examined after 1-month collagenase ageing. The ageing solution of collagenase was obtained by dissolving collagenase of Clostridium histolyticum (Sigma-Aldrich, St. Louis, MO, USA) into artificial saliva (20 mM HEPES buffer, 30 mM KCl, 4.0 mM KH 2 PO 4 , 0.2 mM MgCl 2 ·6H 2 O, 0.7 mM CaCl 2 , 0.3 mM NaN 3 , pH 7.4 [ ]) to achieve a concentration at 0.1 mg/mL. The ageing specimens were stored in this collagenase solution at 37 °C, protected from light.
Beams used for μTBS were individually fixed to a machine for microtensile testing (Bisco Inc., Schaumburg, IL, USA) via a Zapit adhesive (Dental Ventures of America, Corona, CA, USA), and were loaded in tension at a cross-head speed of 1 mm per min until fracture. After recording the maximum fracture load, a digital caliper was used to measure the cross-sectional area of each beam and the actual bond strength values were calculated in Megapascal (MPa).
Fracture pattern analysis
After μTBS test, the fractured surface of each beam was collected, desiccated, sputter-coated with Au-Pd alloy (JFC-1600, JEOL, Tokyo, Japan), and then examined by a field-emission scanning electron microscopy (FESEM, Carl Zeiss Sigma, Jena, Germany). The fracture patterns were distributed into four groups: ( A ) adhesive failure; ( CC ) cohesive failure in composite; ( CD ) cohesive failure in dentin; and ( M ) mixed failure [ ].
Interfacial nanoleakage evaluation
Four of the stored middle slabs selected from each group were randomly classified to be treated immediately or after 1-month collagenase ageing ( n = 2 slabs each subgroup). After coated with two layers of nail varnish which left 1 mm away from the bonded interface, the slabs were soaked in a 50% (w/v) ammoniacal silver nitrate solution for 24 h in darkness, followed by thoroughly rinsing in deionized water. The slabs were further dipped in a photo-developing solution for another 8 h exposure to fluorescent light, then wet-ground using 600, 800, 1200, and 2000-grit silicon carbide papers and 0.25 μm diamond paste, ultrasonically cleaned, desiccated, and sputter-coated with carbon (JFC-1600, JEOL, Tokyo, Japan).
The interfacial nanoleakage was evaluated with FESEM under a back-scattered electron mode. Ten fields-of-view along the bonding interface of each slab were randomly captured (20 images for each subgroup). NIH Image J software (Bethesda, MD, USA) was used to compute the nanoleakage percentage of silver nitrate deposition within the dentin–adhesive layer. The percentage was scored by two examiners on a range of 0–4 according to a protocol previously described as below [ ]: 0, no nanoleakage; 1, <25% nanoleakage; 2, 25%≤50% nanoleakage; 3, 50%≤75% nanoleakage; and 4, >75% nanoleakage. Consistency between the results of observers was determined by Kappa test (K = 0.86).
In situ zymography of the hybrid layer
Two of the stored middle slabs randomly selected from each group were utilized for in situ zymography test. A mixture of fluorescein-conjugated gelatin (E-12055, Molecular Probes, Eugene, OR, USA) was prepared immediately before use according to the manufacturer’s protocols. The slabs were wet-ground to approximately 50 μm thickness and placed on a microscope slide. The gelatin mixture was then dropped individually on each slab, followed by covering a coverslip. All the slabs were incubated in a humidified chamber protected from light at 37 ℃ for 24 h. Each slab was visualized with a confocal laser scanning microscope (CLSM) (Fluoview FV1200, Olympus, Tokyo, Japan) in fluorescence mode by using 40 × objective lens of 0.95 NA under the excitation/emission wavelengths of 488/530 nm. Three images obtained from the same z layer were randomly captured for each slab. All images ( n = 6 images each group) were analyzed and quantified using a NIH Image J 1.8.0 software (Bethesda, MD, USA) to inspect the hydrolysis of the fluorescein-conjugated gelatin substrates; this process can be suggestive of the activity of the endogenous gelatinolytic enzyme based on the value of relative green fluorescence [ ].
Contact angle measurement
Twenty of the sixty-five third molars were sectioned below the enamel-dentinal junction by using the diamond saw with low speed to produce dentin disks (0.5 mm thickness). All the disks were wet-ground with 600, 800, 1200, and 2000-grit silicon carbide papers and 0.25 μm diamond paste, ultrasonically cleaned, and etched with 35% phosphoric-acid gel for 15 s. After rinsed with deionized water and blot-dried, these disks were randomly assigned to five groups ( n = 8 disks each group) in a same procedure as descried in Section 2.1 .: group 1 (WWB); group 2 (DWB); group 3 (0.01%EGCG/DWB); group 4 (0.1%EGCG/DWB); group 5 (1%EGCG/DWB). Contact angle measurement was conducted by using a Contact Angle System OCA (Dataphysics Instruments; Filderstadt, Germany). A droplet of 5 μL of Singlebond universal adhesive was deposited on each dentin disk and the optical data of the droplet was captured and measured with a digital-imaging camera. All parameters were kept constant, especially the distance between the dentin surface and the tip.
Specimens preparation and bacterial culture
Streptococcus mutans ( S. mutans , ATCC 25175), provided by the School of Stomatology, Wuhan University, was cultivated in a Brain Heart Infusion (Becton-Dickinson & Co., Sparks, MD, USA) broth at 37 °C for 24 h in an anaerobic environment. The concentration of the bacterial suspension was confirmed at 10 8 colony forming units (CFU)/mL before use.
According to the same procedure as described in Section 2.1 ., the remaining fifteen of the sixty-five third molars were sectioned, wet-ground, etched, rinsed, blot-dried, and randomly distributed into five groups ( n = 3 teeth each group) : group 1 (WWB); group 2 (DWB); group 3 (0.01%EGCG/DWB); group 4 (0.1%EGCG/DWB); group 5 (1%EGCG/DWB). The dentin surface of each tooth was then bonded with Singlebond Universal adhesive, light-cured, and constructed with 4 mm resin composite build-ups. After 24 h storage in deionized water at 37 °C, the bonded teeth were longitudinally sectioned into a series of slabs of 0.9 mm-thick across the adhesive interface. Eighteen qualified middle slabs were collected from each group, among which, nine of them randomly selected from each group were immediately tested for the antibacterial activity, and the rest nine slabs were tested after 1-month ageing in sterile deionized water at 37 °C.
The inoculation medium of S. mutans was obtained by diluting previously prepared medium with BHI broth supplementing 1% (w/v) sucrose. Each slab was positioned in one well of a 24-well plate and injected with inoculation medium of 1 mL. After 24 h anaerobic incubation at 37 °C for biofilm growth, the biofilm-coated slabs were gently rinsed thrice using sterile phosphate buffer solution (PBS) to wash away non-adherent bacteria.
Live/dead staining of biofilms
Six biofilm-coated slabs (three for immediate test, three for ageing test) randomly selected from each group were stained with a live/dead bacterial viability kit (Molecular Probes, Invitrogen, Eugene, OR, USA). This kit includes two dyes which are SYTO-9 and propidium iodide (PI), enabling live and dead bacteria to be stained to emit green and red fluorescence, respectively [ ]. CLSM was used to examine live and dead S. mutans biofilm along the dentin–adhesive interface in fluorescence mode by using 40 × objective lens of 0.95 NA under the excitation/emission wavelengths of 488/500 nm for SYTO-9 and 594/635 nm for PI. Three representative stacks (Z-stack) of each biofilm image were captured at a Z-step of 2 μm, starting from the bottom (contacting with the treated surface) to the top of the biofilm. A Bitplane Imaris 7.4.2 software (Zurich, Switzerland) was employed to analyze confocal images obtained from the first 10 layers of each Z-stack and to investigate the inhibitory efficacy on biofilm formation for each group.
Six biofilm-coated slabs (three for immediate test, three for ageing test) randomly selected from each group were utilized to survey S. mutans adhesion and biofilm formation along the dentin–adhesive interface. Each slab was fixed with 2.5% glutaraldehyde for 4 h then gradient dehydrated by ethanol (30%, 50%, 70%, 80%, and 90% for 20 min, respectively, and 100% for 20 min twice). After desiccated and sputter-coated with gold (JFC-1600, JEOL, Tokyo, Japan), all slabs were observed by FESEM under a secondary electron (SE2) mode at 5 kV in a high vacuum mode. Three fields-of-view were randomly captured for each slab.
The remaining six biofilm-coated slabs (three for immediate test, three for ageing test) obtained from each group were used to determine the cellular viability of S. mutans biofilm. Each slab was transferred to one well of a new 24-well plate containing 1 mL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; Sigma-Aldrich, St. Louis, MO, USA) solution (0.5 mg/mL), and anaerobically incubated at 37 °C for 4 h. After that, the MTT solution of each well was taken out and replaced by 1 mL of DMSO to dissolve the blue/purple formazan, followed by gently shaking for 20 min. The supernatant of each well was detected at 570 nm via a PowerWave XS2 spectrophotometer (BioTek Instruments Inc., Winooski, VT, USA). Four readings of each slab were recorded for each group ( n = 12). MTT assay was performed in triplicate.
An IBM SPSS Statistics 20.0 software (Armonk, NY, USA) was applied for statistical analysis. For μTBS test, statistical analysis was performed using tooth as the statistical unit; the mean μTBS obtained from the 5 beams of per tooth was used to represent the bond strength of the specific tooth. After the normal distribution of μTBS, gelatinolytic activity, contact angle, and MTT data was confirmed, a two-factor (variables: pretreatment method and collagenase ageing) analysis of variance (ANOVA) with post-hoc Tukey’s test was used to analyze the values of μTBS test. A Kruskal-Wallis test with Dunnett’s post-hoc test was utilized to analyze the statistical differences among the interfacial nanoleakage groups in their scores, while inter-examiner reliability was determined by Cohen’s kappa test. A one-factor ANOVA with post-hoc Tukey’s test was conducted to analyze the data of gelatinolytic activity, contact angles and MTT assay. The significance level for all tests was defined at 0.05.
Microtensile bond strength
The mean μTBS values for all groups are represented in Fig. 1 . Two-factor ANOVA indicated that the variables of pretreatment method ( F = 10.607, p = 0.000) and collagenase ageing ( F = 46.425, p = 0.000) significantly influenced the bond strength. The interaction of pretreatment method × collagenase ageing was significant ( F = 4.160, p = 0.006), suggesting that the variation of μTBS values among the five groups were dependent on the aforementioned two factors.