Influence of flavonoids on long-term bonding stability on caries-affected dentin


  • Cross-linkers improve adhesive and mechanical properties in caries affected dentin.

  • Dentin pretreatment with flavonoids reduce hybrid layer degradation over time.

  • Flavonoids improve morphology of adhesive interface on caries affected dentin.



To evaluate the effect of experimental dentin pre-treatment solutions formulated with different flavonoids on microtensile bond strength (μTBS), nanohardness (NH) and ultra-morphological characteristics of artificial caries-affected dentin (CAD) bonded using a universal bonding system.


A microbiological method was used to create an artificial CAD in 91 human molars. Five experimental pre-treatment solutions were created using the following flavonoids: quercetin (QUE); hesperidin (HES); rutin (RUT); naringin (NAR), or proanthocyanidin (PRO). A placebo solution (PLA) with no flavonoids added was also evaluated. The flavonoids or placebo solutions were applied to the CAD prior to the application and photoactivation of a universal adhesive (Scotchbond Universal, 3M Oral Care). A control group (CON), in which only the bonding agent was applied without any flavonoid solution, was also evaluated. A 3-mm-thick block of resin composite (Opallis, FGM) was built up on the flat bonded CAD surfaces and was light-cured following the manufacturer’s instructions. Specimens were sectioned to obtain resin-dentin slices and sticks (cross-sectional area of 0.8 mm 2 ). The μTBS, NH, and confocal ultramorphology analysis of resin-dentin interface was evaluated at 24 h and after thermo-cycling aging (25,000 cycles). The results were analyzed using 2-way ANOVA followed by Bonferroni’s post hoc test (pre-set α = 0.05).


The specimens from groups QUE, NAR, and RUT presented greater μTBS values than those from CON group ( p< 0.05). Specimens from some of these experimental groups presented greater nanomechanical properties ( p< 0.05), and no morphological degradation at the resin-dentin interface after aging.


The use of exogenous cross-linkers as dentin pre-treatment before bonding procedures may represent a suitable strategy to improve the longevity of universal adhesive systems applied to caries-affected dentin.


According to the Global Burden of Disease (GBD) study, dental caries is the most prevalent pathological condition worldwide [ ]. Indeed, dental caries in primary and permanent teeth continue to be a major problem for the public health system [ ]. When cariogenic bacteria reach the dentinal substrate, endogenous metalloproteinases (MMPs) [ ] and cysteine cathepsins (CP) are activated [ ], resulting in denaturation of collagen fibrils. This process compromises the mechanical properties of dentin and accelerates its degradation [ ].

Considering the philosophy of minimally invasive dentistry, restorative procedures performed on specific substrates such as caries-affected dentin (CAD) [ ] have become a feasible option. However, there are still some concerns regarding the bonding on such challenging substrate; the quality and the durability of the bonding to CAD may be considered unreliable [ ] when compared to sound dentin [ , ]. Indeed, the degradation of the bonding interface created on CAD is much more evident than that observed on the bonding interface created on sound dentin [ , ].

Despite the immediate effect of chlorhexidine as an MMPs inhibitor, the limited, short-lasting effect of this substance has shifted the attention of researchers towards therapeutic substances that may offer a long-lasting effect. In this context, the use of exogenous cross-linking agents has been proposed as an effective method to inhibit the activity of MMPs, therefore preserving collagen within the dentin-bonded interface [ ]. Among these agents, hesperidin (HES), a glycoside vasoactive flavone, has been advocated to reduce the degradation at the bonding interface when incorporated into a two-step self-etching adhesive system [ ]; HES was more efficient in preserving the integrity of the resin-dentin interface than proanthocyanidin (PRO) [ , ]. In addition, despite the benefits of PRO as a cross-linker agent, its incorporation within the composition of adhesives may compromise their bonding performance to dentin [ , , ].

Moreover, glycoside flavones may also have antimicrobial effects on several Gram + and Gram bacterial strains, as well as on Staphylococcus aureus [ ]. The multiple effects of flavonoids, such as HES on sound dentin, have brought a new alternative for the development of therapeutic substances to improve bonding stability on CAD. Among phenolic compounds, there are other molecules with similar structure, such as quercetin (QUE), rutin (RUT), and naringin (NAR) (glycoside flavones). However, to date, no evidence about the potential effects of these substances on long-term bonding on CAD is available.

The aim of this in vitro study was to evaluate the influence of experimental dentin pre-treatment solutions containing QUE, HES, RUT, NAR, or PRO on microtensile bond-strength (μTBS), nanohardness (NH), and the morphology of the resin-dentin interface created by a universal adhesive system applied to CAD in etch-and-rinse mode. The first null hypothesis of this study was that there would be no differences in μTBS and NH values when flavonoids are applied to CAD during the bonding procedure. The second null hypothesis was that the μTBS and NH values in experimental groups with flavonoids after ageing would be no different from those before ageing. The third null hypothesis was that there would be no change in the morphology of resin-dentin interface when flavonoids were used.

Materials and methods

Formulation of experimental solutions

Experimental solutions with QUE, HES, RUT, NAR (Sigma Aldrich, St. Louis, MI, USA) and proanthocyanidin (95%) (PRO) (grape seed extract from Vitis vinifera . Active Pharmaceutica, Palhoça, SC, Brazil) were made. The physical and chemical properties of the molecules are displayed in Table 1 . The primer solutions were determined considering purity, solubility index, hydrophobic nature, and the critical micelle concentration (CMC) of each flavonoid. An exclusive equation was used to the maximum availability of flavonoids in a liquid state without affecting their properties ( Table 2 ) [ ]. A placebo solution (PLA), containing only the vehicles ( Table 2 ) used in the solution but with no flavonoid added, was included as a further experimental group. A control group (CON) where only the adhesive was applied to CAD following the manufacturer’s instructions was also included.

Table 1
Physical and Chemical properties of the molecules used in this study.
HESPERIDIN 610.56 g/mol 2 6 1.06 mM 0.02 mg/mL
NARINGIN 580.53 g/mol 2 6 1.12 mM 1 mg/mL at 40 °C
PROANTHOCIANYDIN 595.55 g/mol 7 a 2 a 1.09 mM a 0.130 mg/mL a
QUERCETIN 302.24 g/mol 5 2.15 mM 0.06 mg/mL
RUTIN 610.52 g/mol 4 6 1.06 mM 0.125 mg/mL

a Expected properties of the Proanthocianydin mer, which may vary according to the number of mers present in the final molecule (oligomer or polymer), reducing solubility and increasing the molecular mass.

Table 2
Composition of a hydro-alcoholic solution of flavonoid (6.5% mass).
Component Compound Quantity %
Active Compound Flavonoid 6.5% mass
Vehicle (Pure Ethanol) Pure Ethanol 30% (3 mL)
Surfactant (Polysorbate 20) SPAN 20 1% (0.1 g)
Aqueous medium Distilled Water QS 10 mL

Specimen preparation

Ninety-one caries-free extracted human third molars obtained from patients (range: 18 – 35 years old) were used. After approval by the local Ethics Committee (protocol # 41.2017), teeth were collected from several patients, and an informed consent for surgery was obtained through a written document. All the extracted teeth were disinfected in 0.5% chloramine, stored in distilled water and used within three months after extraction. A flat mid-dentin surface was exposed on each tooth using a 180-grit SiC paper under continuous irrigation.

Simulated microbiological caries

Such a method to create a simulated microbiological-based caries lesion on dentin was validated in a previous study [ ]. All specimen surfaces were covered with a layer of epoxy resin (Araldite, Brascola Ltda, São Bernardo do Campo, SP, Brazil), followed by a layer of nail varnish, while only the occlusal surface was left exposed. All specimens were sterilized in steam autoclave (Phoenix Ind. Brasileira, Araraquara, SP, Brazil) for 15 min at 121 °C [ ], and each tooth was individually immersed in an 8-mL Falcon tube containing an artificial caries solution. Such a solution was composed of 9.25 g of brain heart infusion culture supplemented with 1.25 g of yeast extract, 5.0 g of sucrose, in 250 mL of distilled water and 100 μL of primary culture of S. mutans (INCQS 00446), with the pH around 4.0. The specimens were incubated in an anaerobic jar (5% CO 2 ) at 37 °C. The specimens were transferred to another 8-mL Falcon tube containing a new artificial caries solution every 48 h. After 14 days, all specimens were sterilized again as previously described and washed with deionized water [ ].

Bonding procedures

The experimental design is presented in Fig. 1 . Prior to the bonding procedures, the surrounding enamel of each specimen was removed with a diamond bur (#4137, KG Sorensen, Barueri, SP, Brazil), until the dentin surface was totally exposed. Subsequently, the occlusal dentin surface of each specimen was polished using a 600-grit silicon-carbide paper for 30 s to obtain a standardized smear layer [ ]. Afterwards, the teeth were randomly allocated to the following experimental groups: QUE, HES, RUT, NAR, PRO, PLA, and CON.

Fig. 1
Experimental design used in this study.

The occlusal dentin surface was etched with 37% phosphoric acid for 15 s, water-rinsed for 30 s and air-blown dried for 5 s. Each respective experimental solution was actively applied for 1 min to re-wet the dentin surface. These were slightly air-dried for 2 s and the moisture was homogenized with absorbent paper leaving a wet surface. The universal adhesive system Scotchbond Universal (3M Oral Care, Saint Paul, MN, USA) was applied following the manufacturer’s instructions ( Table 3 ) and light-cured for 10 s at standard mode using a polywave LED curing system (Valo, Ultradent Products, South Jordan, UT, USA). A 3-mm thick resin composite block (Opallis, FGM Prod. Odont. Ltda, Joinville, SC, Brazil) was built up on the bonded surfaces in three increments of 1-mm thick; each one was individually light-cured for 40 s (Valo, Ultradent Products). A single operator carried out all the bonding and restorative procedures in an environment with controlled temperature and humidity.

Table 3
Adhesive and experimental solutions used in this study.
Substance Composition Application
3M Scotch Bond Universal Ingredients: MDP, Dimetacrilate resins, HEMA, Vitrebond TM copolymer. Filling particles, ethanol, water, initiators, Silane. Total etch technique − Apply 1 drop of Single Bond Universal over the etched dentin regions for 20 s, dry for 5 s and cure for 10 s
Hesperidin Solution Hydro alcoholic solution of Hesperidin 6.5% Actively apply for 1 min over etched dentin to rewet.
Quercetin Solution Hydro alcoholic solution of Quercetin 6.5% Actively apply for 1 min over etched dentin to rewet.
Rutin Solution Hydro alcoholic solution of Rutin 6,5% Actively apply for 1 min over etched dentin to rewet.
Naringin Solution Hydro alcoholic solution of Naringin 6.5% Actively apply for 1 min over etched dentin to rewet.
Proanthocyanidin Solution Hydro alcoholic solution of GSE 6,5% (95% of PRO) Actively apply for 1 min over etched dentin to rewet.
Placebo Hydro alcoholic solution with tensoactive SPAN 20. Actively apply for 1 min over etched dentin to rewet.

The specimens were stored in distilled water at 37 °C for 24 h. After storage, only 21 specimens were longitudinally sectioned in “x” direction across the bonded interface (IsoMet 1000; Buehler, Lake Bluff, USA), under water cooling at 300 rpm to obtain 1.2-mm thick resin-dentin slices for NH analysis. Forty-nine specimens (n = 7) were longitudinally sectioned in both “x’’ and “y’’ directions across the bonded interface to obtain resin-dentin sticks with a cross-sectional area of 0.8 mm 2 ; the exact dimensions were measured using a digital caliper and recorded to determine the μTBS values (Absolute Digimatic, Mitutoyo, Tokyo, Japan). Half of the sticks were evaluated after 24 h, while the other half were submitted to thermocycling (25,000 cycles; dwell time of 30 s from 5 °C to 55 °C; Odeme, Joaçaba, SC, Brazil) prior to μTBS testing [ ].

Microtensile bond strength testing

Each stick was attached to a modified device for μTBS test with cyanoacrylate resin (IC-Gel, bSi Inc., Atascadero, CA, USA) and subjected to a tensile force in a universal testing machine (Kratos, São Paulo, SP, Brazil) at 0.5 mm/min. The failure mode was evaluated under an optical microscope (SZH-131, Olympus, Tokyo, Japan) at 40x and classified as cohesive in dentin (failure exclusive within cohesive dentin – CD); cohesive in resin (failure exclusive within cohesive resin – CR); adhesive (failure at resin-dentin interface – A), or mixed (failure at resin-dentin interface that included cohesive failure of the adjacent substrates, M). The number of premature failures (PF) was recorded and was not included in the average μTBS results.

Nanohardness within adhesive layer, hybrid layer and dentin

The resin-dentin slices obtained previously from 21 restored teeth (n = 3) were wet-polished using 1000 to 4000-grit SiC papers for 30 s each and cleaned in an ultrasonic water bath for 5 min. The specimens were attached to a metal stub and placed in a nanoindenter device (UNAT nanoindenter, Asmec, Dresden, Germany), which had a Berkovich indenter (20 nm radius) to evaluate the NH. The adhesive interface was visualized with a microscope, and a net of 24 indentations was created (6 in the “x” axis and 4 in the “y” axis) with a load of 5000 nN and a function time of 10 s starting from the adhesive layer (AL) and moving down towards the resin-dentin interface and dentin. This procedure was done to evaluate the hybrid layer (HL) as well as the dentin at a 50 μm depth. The distance between each indentation was consistently maintained by adjusting the distance range by 100 μm (±10 μm) per range on the “x” axis. The values obtained after the indentation were analyzed in a software to calculate NH (InspectorX, ASMEC GmbH, Dresden, Germany).

Confocal ultramorphology evaluation

Prior to the bonding procedures, the adhesive system was doped with Rhodamine B (83689-1 G, Sigma-Aldrich, Saint Louis, MI, USA) at approximately 0.2 wt.% [ ]. The specimens were restored as previously described for μTBS test. Half of the slices (n = 3) were immersed in 0.1 wt.% sodium fluorescein (46960-25G-F, Sigma-Aldrich, Saint Louis, MI, USA) for 4 h [ , ], while the other half (n = 3) were aged by thermocycling, as previously described, and then immersed in Fluorescein.

Specimens were polished with 1000 to 2500-grit SiC for 30 s and ultrasonically cleaned (2 min), air-dried, and the resin-dentine interfaces were analyzed using Confocal Laser Scanning Microscopy (CLSM) (DMi8 Cell Advanced Leica, Mannheim, Germany) equipped with a 63×/1.4 NA oil immersion lens. The emission fluorescence was recorded at 512–538 nm (Fluorescein) and 585–650 nm (Rhodamine B). Ten images from each slab were randomly captured at 5 and 10 μm and were analyzed with the CLSM image software (LAS X, Leica, Heidelberg, Germany).

Statistical analysis

The analysis was performed using the tooth as the statistical unit. The μTBS and NH results were averaged to obtain the mean bond strength for each tooth. The values were submitted to a 2-way repeated ANOVA, followed by Bonferroni’s post hoc test (pre-set α = 0.05). Post-hoc power analysis was performed using the SPSS19 (IBM Company, Armonk, NY, USA). The morphological characteristics of the HL were qualitatively evaluated in CLSM.


Microtensile bond strength and fracture analysis

The study had adequate power for both factors (treatment; time), (> 90%; α = 0.05). The μTBS values are displayed in Table 4 . After 24 h, the groups QUE, RUT, NAR showed the highest μTBS values, which were significantly higher than those observed in CON ( p= 0.005, p= 0.001, p= 0.008, respectively). No significant difference was observed between the other experimental groups; CON showed the lowest values at both periods. After thermocycling, RUT and NAR groups exhibited the highest μTBS values ( p< 0.001), while the specimens in CON and QUE groups presented the lowest μTBS values after thermocycling. Only specimens from RUT, CON, and QUE groups exhibited a significant drop in μTBS after thermocycling ( p= 0.013, p= 0.012, p< 0.01, respectively; Table 4 ).

Aug 18, 2020 | Posted by in Dental Materials | Comments Off on Influence of flavonoids on long-term bonding stability on caries-affected dentin
Premium Wordpress Themes by UFO Themes