Optimization of Dental Adhesive Interfaces Using Tissue Biomodulation with DESIGNER Biopolymers

Abstract

Objective:

To investigate the modulatory effects of four proanthocyanidin-DESIGNERS (PAC-DESIGNERs) on the long-term bond strength of the resin-adhesive interface, the degree of conversion of resin monomers, the chemical-mechanical properties of dentin matrix, and cell biocompatibility.

Methods:

Standardized formulations of PACs with a dominant degree of polymerization - DP (trimer: PM-AB and CV-AB; tetramer: PM-ABA and CV-ABB) were prepared from 2 sources of AB-Type PACs using a DESIGNER approach. Resin-dentin interface was assessed after 24h and 1 year using a microtensile bond strength (µTBS) test. The degree of conversion (DC) of resin monomers and chemical analysis of the dentin matrix were analyzed by ATR-FTIR spectroscopy. The viscoelastic properties of the dentin matrix were assessed by dynamic mechanical analysis (DMA). Cell viability was analyzed using a 3D cell culture model. Data were analyzed using two and one-way ANOVA and post-hoc tests (α = 0.05).

Results:

All PAC-DESIGNER biomodulation increased the µTBS when compared to control (p<0.05), regardless of source, DP, and aging. The DC of resin adhesive was not negatively impacted, and an increase in DC was observed with the incorporation of PM-AB and PM-ABA DESIGNERs (p<0.05). PAC-DESIGNER increased the dentin matrix complex modulus (153–79MPa) and storage modulus (151–78MPa) when compared to control (~9MPa, p<0.05). All DESIGNERs decreased the intensity of amide II/CH₂ ratio; a decrease in the amide III/CH₂ ratio was observed for CV-ABB (p<0.05). PAC-DESIGNERs exhibited good cell biocompatibility and healthy cell morphology.

Significance:

All PAC-DESIGNERs optimized the dentin-resin µTBS. The different molecular structures played a modulatory role in the chemical-mechanical properties of the dentin matrix, the degree of conversion of adhesive, and cell biocompatibility.

1. INTRODUCTION

The contemporary landscape of dental restorative materials has seen a burgeoning interest in innovative biotechnologies aiming at enhancing biostability, biomechanics, and biocompatibility critical for the service life of adhesive-based dental restorative therapies [1]. Biomodulation of dental tissue is a bioinspired approach that multi-functionally enhances the performance of dentin-resin interfaces [2,3]. A well-established dentin biomodulation concept encompasses the use of phytochemicals that mimic inherent mechanisms of functional resilience and self-shielding. Specifically, this approach exploits proanthocyanidins (PACs), a class of plant-derived oligomers, from certain plants, which chemically mediate and/or mimic multi-scale length covalent, non-covalent, and covalent-like interactions within the dentin matrix [2,4,5] resulting in the mechanical reinforcement and reduced biodegradability of dentin [4].

PAC-rich crude extracts of select plants have been demonstrated to diminish the prolonged breakdown of the adhesive interface [6]. Further investigation revealed that purified, single compounds from Cinnamomum verum and Pinus massoniana enhance the dentin mechanical properties by mediation of non-enzymatic collagen cross-linking and reduce collagen biodegradability by coverage of cleavage sites of endogenous proteases in type I collagen [6–8]. As a result of these effects in dentin, PACs increase the short- and long-term bond strength and decrease micro-permeability at the resin-dentin interface [9]. These beneficial properties can be attributed to the amphiphilic nature of PACs, involving hydrophilic phenolic groups connected with aromatic rings, which collectively enhance and stabilize resin adhesion via interactions between the hydrophilic dentin matrix and the hydrophobic resin [3].

The degree of polymerization (DP) of PACs is a key regulator of the dentin biomodulation mechanism. Trimeric and tetrameric PACs consistently showed higher potency to increase the dentin biomechanical profile when compared with other oligomers [5,10]. A DP of 3 and 4 gives the compounds the best structural ability to fit into the interfibrillar spaces within the collagen scaffold, as the molecular size and shape favor interactions with the dentin matrix and, therefore, their modulatory role [10]. The inter-flavan linkage (IFL) is another structural feature that has recently been shown to further modulate PAC-dentin interactions. Two types of IFLs exist: the B-type has one carbon-carbon (4→8 or 4→6) bond between monomeric units, while the A-type has an additional ether bond (C-O-C, 2→O→7, or 2→O→5) (1) [5]. Previous studies [5,11] have shown that A-type IFLs exert more dentin stability and lower degradation and lower the oxidation ratio to the tissue compared with analogous B-type PACs. Accordingly, the additional ether bond appears to be a key regulator of the long-term stability of the dentin matrix [11].

In the pursuit of standardizing PAC-based interventional biomaterials, we recently introduced the DESIGNER (Depletion and Enrichment of Select Ingredients Generating Normalized Extract Resources) concept as a means of producing materials selectively enriched in dentin-bioactive PAC trimers and tetramers, with well-characterized chemical profiles and dentin biomodification profiles [12,13]. The DESIGNER approach employs centrifugal partition chromatography (CPC), a form of countercurrent (liquid-liquid) separation, which is loss-free and can enrich the desired components while removing (knock-down/knock-out) undesired compounds. PAC-DESIGNER intervention materials are also a cost-effective strategy as they advance the use of biopolymers contained in natural waste materials into new commercial products, in a highly reproducible and scalable manner. Furthermore, DESIGNER materials are chemically much more defined and better suited for analytical standardization.

This study involved the preparation of four different combinations of A- and B-type PAC-DESIGNER materials (PAC-DESIGNERs in the following) to assess the role of DP in trimeric and tetrameric PACs from two priority plants sources, Cinnamomum verum bark and Pinus massoniana bark. By studying dentin-resin adhesion, degree of conversion of resin monomers, dynamic mechanical properties of dentin matrix, biochemical characteristics, and 3D cell culture models, this study shed comprehensive light on the multifaceted structure-activity relationships and potencies of PAC-DESIGNERs. The null hypothesis tested was that no significant differences would arise among the four PAC-DESIGNERs regarding their modulatory roles and effects on dentin matrix, dental pulp stem cell proliferation, and resin-adhesive interface as indicators of the viscoelastic properties of dentin extracellular matrix and biocompatibility.

2. MATERIAL AND METHODS

2.1. PAC-DESIGNERs preparation

The PAC-DESIGNERs were prepared from Cinnamomum verum (CV) and Pinus massoniana (PM) barks. The phytochemical procedures and spectroscopic analysis for their purification and structural characterization have been reported in detail previously [13,14] and are briefly summarized as follows. The production of standardized ligand-PACs was based on the DESIGNER approach [5,13]. The methyl acetate partition (83.6 g) of CV was fractionated by DIAION HP-20 chromatograph column eluted by 40% MeOH in H2O to yield CV oligomers enriched fraction A (16.7 g). Fraction A (5.00 g) was further purified by CPC (EtOAc/n-Butanol/MeOH/H₂O, 6/0.1/1/5, v/v), in ascending mode at a flow rate of 10.0 mL/min and 3,000 rpm, the stationary retention volume ratio (Sf) was 0.80. The CPC fractions were pooled into two subfractions according to their Thin Layer Chromatography (TLC) profiles, yielding the trimeric CV-AB (845 mg) and tetrameric CV-ABB (633 mg) DESIGNERs, respectively.

The EtOAc-soluble extract (48.8 g) of PM was fractionated on CPC-I using n-hexane/EtOAc/MeOAc/H₂O (2/4/1/4, v/v) in descending mode at a flow rate of 30.0 mL/min and 2,500 rpm, the stationary retention volume ratio (Sf) was 0.74. The CPC-I fractions were pooled as three fractions based on their TLC profiles, with the first fraction being the dimer/monomer knock-out fraction A (24.9 g). Fraction A (5.40 g) was further subjected to CPC-IIa using n-hexane/EtOAc/MeOAc/H₂O (0.5/4/1/4, v/v) in ascending mode (10 mL/min, 3,000 rpm, Sf 0.84). The CPC-IIa fractions were combined as five subfractions 1–5, with subfraction 2 enriched in trimers representing DESIGNER PM-AB (497 mg). The dimer/tetramer enriched subfraction 1 (1.20 g) was loaded to another CPC-III separation using n-hexane/EtOAc/MeOAc/H₂O (2/4/1/4, v/v) in descending mode (30 mL/min, 2,500 rpm, Sf 0.7), yielding the tetrameric DESIGNER PM-ABA (746 mg).

2.2. Formulation of the DESIGNER primers and experimental adhesive resin

The DESIGNERs were formulated as a neutral primer (20 mmol/L, pH~7). Each DESIGNER material was precisely weighed into amber tubes for light protection and dissolved in HEPES buffer to achieve final concentrations of 6.5% (w/v), with pH adjustment to 7.0 at room temperature.

An experimental dental adhesive with the following components in percent by weight (wt.%) was prepared: 2-hydroxyethyl methacrylate/ HEMA (10.3); camphorquinone/ CQ (0.50); diphenyliodonium hexafluorophosphate/ DPIHP (1.0); ethyl-4-(dimethylamino) benzoate/ EDMAB (0.50); triethylene glycol dimethacrylate/ TEGDMA (20.6); bisphenol glycidyl dimethacrylate/ Bis-GMA (37.7); and ethanol 100% (29.4). The hydrophobic photoinitiator (PI) components were first mixed with HEMA and ethanol to enhance dissolution. Next, the viscous BisGMA was added, and the mixture was vortexed for 5 min, then agitated on a tube rotator for 24 h. Finally, the resin was aliquoted into 1.0 mL amber vials for storage [2].

2.3. Production of resin-dentin biointerfaces and study of the strength of adhesion

Dentin specimens were sourced from forty extracted (sound) human molars, obtained under an approved protocol (IRB # 2023–0717). The occlusal surfaces of the molars were flattened using silicon carbide abrasive papers (180 and 320 grit) to remove occlusal enamel and expose mid-coronal dentin surfaces. The flat dentin surfaces were further polished (600 grit for 20 s) to standardize a smear layer. The specimens were randomly divided into five groups (n = 8): unprimed (control) and primed with CV-AB, CV-ABB, PM-AB, or PM-ABA.

The restorative protocol included surface etching with 35% (w/v) glycolic acid (pH= 1.3) for 15 s, rinsing for 15 s, blot drying of the surface, and experimental primers were actively applied on the dentin surface for 1 min. [3,15] Primed surfaces were thoroughly rinsed with ultrapure water for 30 s. Two consecutive layers of an experimental adhesive resin were applied on the surface followed by gentle air drying for 20s for solvent evaporation, and light curing for 40 s (3M Elipar Deep Cure-S Curing Light, 1200 mW/cm², Solventum). A commercial resin composite (Filtek Supreme Ultra, Lot: N959543, 3 M ESPE, St. Paul, MO, USA) was employed to incrementally build a restoration block in 3 increments, each increment light cured for 40 s. Specimens were immersed in a simulated body fluid - SBF (5 mmol/L HEPES, 2.5 mmol/L CaCl₂, 0.05 mmol/L ZnCl₂, and 0.3 mmol/L NaN₃ (pH 7.4) [16] and incubated at 37 °C. After 24 h incubation period, the specimens were serially sectioned into resin-dentin beams with a cross-sectional of 0.8 ± 0.05 mm², as previously described [2]. The bond strength was assessed after 24h and 1-year storage in SBF. SBF was replaced every 2 weeks.

Five resin-dentin beams were selected from each tooth and affixed by their edges to a Ciucchi jig using cyanoacrylate glue (Super Glue gel Loctite; Henkel Corporation, Rocky Hill, Connecticut, USA). Tensile testing was performed on a Shimadzu AGS-X universal testing machine (200 N; Columbia, MD 21046 USA), with a crosshead speed set at 1 mm/min. Micro-tensile bond strength (µTBS) was determined by dividing the peak load force by the cross-sectional area of the interface (Figure 2A). Data were statistically analyzed using two-way ANOVA, followed by Games-Howell post-hoc tests (α = 0.05, SPSS v.25, SPSS).

Figure 2.

(A) Schematic overview of the resin-dentin specimen preparation process for the micro-tensile bond strength test. (B) Results of the dentin-resin micro-tensile bond strength (µTBS) of dentin treated with four PAC-DESIGNER primers after 24h and 1 year aging. An interaction was observed between experimental groups and time points (p=0.029). Different letters indicate statistical differences between groups (α=0.05). (C) Distribution of the various failure modes for the treatment and control groups: adhesive at dentin (interface dentin–adhesive), adhesive at resin (interface resin–adhesive), mixed (dentin–adhesive–resin), cohesive in dentin, cohesive in resin.

The failure modes of all tested specimens were analyzed using a stereomicroscope and classified using a previously published rank [17]. The debonded areas were classified as adhesive at dentin (interface dentin–adhesive), adhesive at resin (interface resin–adhesive), mixed (dentin–adhesive–resin), cohesive in dentin, and cohesive in resin. Fractured specimens were analyzed in a Dino Lite Microscope (AM4515ZT - EDGE- Dunwell Tech., Inc.).

2.5. Effect of PAC-DESIGNERs on the Degree of Conversion of Experimental Adhesive

To determine the effect of AB-type PAC-DESIGNERs on the degree of conversion of experimental resin adhesive, we incorporated DESIGNERs on the experimental adhesive at three concentrations: 0.4, 1.0 and 2.5 wt. %. The formulations were vortexed for 3 minutes and placed on a tube rotator for 2 hours to ensure complete homogeneity.

The degree of conversion (DC) was determined by Fourier Transform Infrared Spectroscopy (FTIR, Thermo Fisher Scientific, USA) with the attenuated total reflectance (ATR). Briefly, 3 μL of the adhesives (n = 5) were placed on the diamond crystal, and the spectra were immediately obtained with 32 accumulations and at a resolution of 4 cm⁻¹. The increments were photoactivated for 40s at a distance of 1.5 cm. The peaks of 1635 cm⁻¹ and 1740 cm⁻¹ correspond to carbon-carbon double bonds in the aliphatic chain and the carbonyl groups, respectively. The DC was calculated using Equation 1 [18] The data were analyzed using two-way ANOVA followed by Tukey’s post-test (α=0.05, SPSS v.25, SPSS).

$$DC\left(\%\right)=\left(1-\frac{\frac{A_{sample}^{1635}}{A_{sample}^{1740}}}{\frac{A_{monomer}^{1635}}{A_{monomer}^{1740}}}\right)\times 100$$ (1)

2.6. Dynamic mechanical analysis of dentin matrix

The viscoelastic properties of the dentin matrix were evaluated using the bulk dynamic mechanical analysis (DMA) method [5,7]. Mid-coronal dentin specimens of 1.5 × 7 × 0.5 mm (width × length × thickness) were prepared from extracted human sound molar and demineralized in 10% phosphoric acid (Ricca Chemical Company, Arlington, TX, USA) for 5 h. A dimple was created with a diamond bur (835.31.014 F G, Brasseler USA Dental, Savannah, GA, USA) in one of the edges for constant specimen positioning during the experiment.

Dentin specimens (n = 5) were divided into experimental groups of 4 PAC-DESIGNERs and control (untreated). PAC-DESIGNER primers were prepared at 1% w/v in 20 mmol/L HEPES buffer (pH 7) [19]. A control group was immersed in HEPES buffer only. Specimens were immersed in 80 uL of solution (pH 7) for 1 h at room temperature and rinsed with ultra-pure water [19]. A strain sweep method employing a 3-point bending submersion clamp was used with pre-set parameters of 1 Hz frequency, 1 to 100 μm varying amplitude, and a preload force of 0.01 N (Q800 DMA, TA Instruments, New Castle, DE, USA) [7]. Viscoelastic properties, represented by storage (E’), loss (E”), and complex (E*) moduli - indicating elastic, viscous, and overall bulk viscoelastic properties were assessed. Additionally, damping capacity, calculated as the ratio of viscous to elastic components (tan δ = E”/E’), was determined. Statistical analyses employed one-way ANOVA and Games-Howell post-hoc tests (α = 0.05, SPSS v.25, SPSS).

2.7. Biochemical analysis of collagen in dentin matrix

Fourier-transform infrared spectroscopy (FTIR) was employed for the biochemical characterization of functional groups within the dentin matrix. The analysis was conducted using the FTIR spectrometer (Thermo Fisher Scientific, USA) with an attenuated total reflectance (ATR) unit. Spectra were collected across the range of 4,500 to 600 cm⁻¹, with 128 scans accumulated per sample (n = 3). Before each analysis, background spectra were collected without a sample on the ATR crystal to compute absorbance spectra. The specimens were pressed onto the ATR plate surface for spectrum acquisition. Following acquisition, the spectra underwent a 14-point interpolated baseline subtraction and were normalized based on the mean values of all datasets.

To assess the secondary structure of collagen, the amide bands were correlated with hydrogen bonding patterns of collagen polypeptide chain backbones, such as α-helices and β-sheets. The study focused on the primary bands of type I collagen: amide I (1,630 cm⁻¹), amide II (1,550 cm⁻¹), amide III (1,240 cm⁻¹), and CH₂ wagging vibrations (1,450 cm⁻¹). Intensities and band areas (integrals) were used as measures of investigating structural variations in collagen cross-linking and within the collagen triple helix. Ratios between the amide bands II and III vs. the CH₂ scissoring band were calculated. FTIR indices (amide II/CH₂ and amide III/CH₂) of untreated and DESIGNER treated dentin matrices were statistically analyzed to assess the effects of dentin biomodulation [20]. Spectra were processed using OMNIC Spectra and Origin Pro 8 software, and the resulting data were statistically analyzed using One-way ANOVA and Games-Howell post-hoc tests.

2.8. 3D cell culture model of the dentin-pulp complex

Human dental pulp stem cells (DPSCs) were cultured in Dulbecco’s modified Eagle’s medium (D-MEM; Corning; USA) supplemented with 50% of Fetal Bovine Serum (FBS), 1% L-glutamine and 1% antibiotic/antimycotic solution. Cells were seeded on the flask and incubated with 5% of CO₂ at 37^oC. Type I bovine collagen (6 mg/mL, Nutragen, Advanced Biomedical) was enriched with D-MEM powder without phenol red solution (13 mg/mL), Fetal Bovine Serum (8.5% v/v), L-Glutamine (1% v/v) and sodium bicarbonate solution (5 mg/mL), and pH adjust (pH = 7.4) with sodium hydroxide solution. The scaffold was synthesized on ice maintaining the low viscosity. The scaffolds were distributed in the well plate and stored in 5% CO₂ at 37^oC for 2 h. The DPSCs concentration was 3.5 × 10⁵ cell/scaffold [21].

Mid-coronal dentin specimens of 1.7 × 0.5 × 3 mm (thickness × width × length) were prepared from extracted human sound molars. The specimens were etched using 35% glycolic acid for 30 seconds and rinsed for 30 seconds. For dentin biomodulation, the specimens were treated with 20 μL of the 6.5% of PAC-DESIGNERs primers for 1 min. Each dentin specimen was placed on transwell permeable support (polystyrene/6.5 mm insert/8.0 μm PET membrane) over the collagen scaffold cell culture model. A dentin specimen placed over the collagen scaffold was used as a positive control and a blank (regular media without cells) was used as a negative control. The cell viability was assessed using PrestoBlue reagent (Invitrogen, A13262), with the fluorescence being measured on a microplate reader (BioTek Synergy HTX plate reader, excitation at 540 nm) after 1, 3, and 5 days. The cell viability data variability was determined using Levene’s test. The data was statistically analyzed using two-way ANOVA and Tukey’s post-hoc test (α = 0.05; SPSS v.25, SPSS) [22].

For the live and dead assay, collagen scaffolds were sectioned and transferred to 24 well plates, and a solution containing 4 μmol/L ethidium homodimer-1 (EthD-1) and 2 μmol/L calcein AM (Live/Dead viability kit-Invitrogen, Cat#L3224) was used. Plates were incubated for 30 min at 37^oC/5% CO₂. After this step, the scaffolds were rinsed twice with sterile deionized water for 10 min. The analysis was performed using a fluorescence microscope (EVOS^® FL Auto, life technologies).

3. RESULTS

3.1. Dentin-resin micro-tensile bond strength

Figure 2B illustrates the 24 h and 1 year results of the micro-tensile bond strength assay (µTBS, in MPa). An interaction was observed between experimental groups and time points (p=0.029). All PAC-DESIGNER primers increased the µTBS when compared to the control group immediately and 1 year of aging (p<0.05). There was no significant difference among the PAC DESIGNERs primers at each timepoint (p>0.05). PAC-DESIGNERs primers promoted stability bond strengths after 1-year (p≥0.05), except CV-ABB which exhibited a significant decrease after aging (p>0.05). Percent distribution of the failure modes [adhesive at dentin (interface dentin–adhesive), adhesive at resin (interface resin–adhesive), mixed (dentin–adhesive–resin), cohesive in dentin, cohesive in resin] are reported in Figure 2C. With dentin-adhesive (d-a) interfaces withstanding > 52 MPa, less than 10% of the DESIGNER-treated specimens debonded entirely at the critical d-a interface, indicating robust hybrid layers at 24h evaluation. Conversely, in the control group, failures generally occurred at the d-a interface (60%), and after 1 year aging specimens exhibited failures at the d-a interface (29.2%) and mixed failures (37.5%). The CV-ABB group yielded a higher proportion of failures at the interface resin-adhesive (85%) at 24h and cohesive at resin failure (31.4%) after 1-year aging. CV-AB and PM-ABA showed a similar distribution of failure at interface resin-adhesive (> 47.5%) and dentin-adhesive (< 7.5%). In addition, CV-AB had the highest values of mixed failure (35%). After 1 year aging, CV-AB showed the highest failure percentage of cohesive at dentin (31.4%) and PM-ABA resin-adhesive (40.0%). Among all treated groups, PM-AB presented the lowest percentage of failure at the interface resin-adhesive (47.5%), and after 1 year aging, mixed failures were the most prevalent fracture pattern (31.4%).

3.3. Degree of conversion of experimental adhesive

The results of the degree of conversion of the experimental adhesive with DESIGNERs as additives are displayed in Figure 3. Incorporation of PAC-DESIGNER material into the solvated experimental adhesive increased the degree of conversion (%) at all concentrations for PM-AB and PM-ABA groups, compared to the control (p<0.05). In contrast, the conversion rate did not differ statistically between the control and DESIGNERs groups CV-ABB and CV-AB (p>0.05).

Figure 3.

Degree of Conversion (%) for neat experimental adhesive resin (control) vs. experimental adhesive resin mixed with each of the four PAC-DESIGNERs, at different concentrations (0.4, 1, and 2.5 wt.%) at 15 min after photoirradiation at room temperature. Distinct Greek letters show statistical differences between the experimental groups and concentrations (α = 0.05).

3.4. Dynamic mechanical analysis of the dentin

The bulk effect of the PAC-DESIGNERs on the viscoelastic properties of the dentin matrix is illustrated in Figure 4. Biomodulated dentin significantly influenced the viscoelastic properties, including storage (E’), loss (E”), and complex (E*) moduli (p<0.05). Biomodulation by PAC-DESIGNERs showed a 12-to-20-fold increase in their mechanical properties (E’ and E*) compared with control (p<0.05). The PM-AB DESIGNER elicited the highest increases in E’ and E* of dentin, followed by PM-ABA, CV-AB, and CV-ABB (p<0.05). The CV-AB, PM-AB, and PM-ABA DESIGNERs elicited higher values of E” when compared the CV-ABB and control group (p<0.05). The DESIGNERs affected the damping capacity (tan δ) of dentin differently: dentin modulated with CV-AB presented the highest tan δ, followed by CV-ABB (p<0.001), and no statistical differences were found between control and PM-AB or PM-ABA (p>0.05).

Figure 4.

Results of the dynamic mechanical analysis of dentin matrices treated with PAC-DESIGNERs. The graphs illustrate the mean values along with standard deviation bars for (A) storage modulus (E’), (B) loss modulus (E”), (C) tan 𝞭, and (D) complex modulus (E*). Different symbols indicate statistically significant differences among groups (p< 0.05).

3.5. Biochemical analysis of dentin matrix

Figure 5A shows representative FTIR spectra of untreated vs. treated dentin matrices. The spectra include band assignments for key features such as amide I (~1,630 cm⁻¹), amide II (~1,550 cm⁻¹), CH₂ scissoring (~1,450 cm⁻¹), and amide III (~1,240 cm⁻¹). Across all dentin matrix spectra, absorption bands within the range of 1,200 to 937 cm⁻¹ were observed, corresponding to C–O bond stretching. Distinct peaks at 1,031 and 1,080 cm⁻¹ were attributed to C–O vibrations and C–O–C absorptions of carbohydrate moieties, [23] as well as the symmetric extension of the C–O–C group in the pyran ring of condensed tannins [24]. Additionally, bands within 1,160 to 1,145 cm⁻¹ range were associated with the asymmetric stretching of C–O–C, [25] as well as the symmetric bonding of aliphatic C–H, O–H, or C–O stretch of different groups [26]. The identification of qualitative spectral differences of type I collagen after dentin treated with PAC-DESIGNERs relied on several parameters: the emergence of new bands, the broadening and shifting of previously assigned bands, and an elevated absorbance compared to the unmodified control. DESIGNER treatment led to a noticeable decrease of intensity in the band at∼1,340 cm⁻¹ and the shoulder at ∼1,321 cm⁻¹, especially in the PM-ABA group. In addition, an increase in the intensity at ∼1,283 cm⁻¹ was observed in all treated dentin groups. It was observed band shifts at ∼1160 cm⁻¹ and ∼1031 cm⁻¹ which are possibly expressed at ∼1145 and ∼1168 cm⁻¹, respectively. New band was detected at ∼1,177 cm⁻¹ for PM-treated specimens, while decreases of intensity in those peaks for CV-treated, especially for CV-ABB. A modification of the band at 1,160 cm⁻¹ was observed from CV-ABB treatment. Dentin biomodulation significantly decreased the intensity of amide II/CH₂ ratio, while a decrease in the amide III/CH₂ ratio was only observed for CV-ABB.

Figure 5.

(A) Characterization of biochemical properties and structural modification in dentin matrix induced by PAC-DESIGNERs using Fourier-transform infrared (FTIR) spectroscopy. Groups included control (untreated), trimer CV-AB, trimer PM-AB, tetramer CV-ABB, and tetramer PM-ABA DESIGNERS. Notably, the FTIR spectra between 1,400 and 800 cm⁻¹ (right panel) reveal new bands indicative of modifications of dentin matrices treated with all DESIGNERs of this study. (B-D) Illustration of the mean values along with standard deviation bars of ratios derived from intensities and areas under the assigned peaks, specifically the amide I vs. CH₂ ratio (1,630/1,450 cm^–1), amide II vs. CH₂ ratio (1,550/1,450 cm^–1) and the amide III vs. CH₂ ratio (1,240/1,450 cm^–1). Different Greek letters indicate statistically significant differences between groups. Statistical significance was set at p< 0.05.

3.6. Cell culture

Figure 6B shows the results of cell viability (%) in a 3D cell culture model. Overall, the PAC-DESIGNERs exhibited good cell biocompatibility in each experimental group and each timepoint. The PM-ABA group showed cell viability decrease after 3 and 5 days (p<0.001). Hence, a cell proliferation process was observed for the CV-ABB and CV-AB groups after 3 days (p<0.001). No significant difference in cell viability occurred among PAC-DESIGNERs after 3 and 5 days (p = 0.121). The live and dead assay showed a healthy cell morphology and similar amounts of cell death among experimental groups (Figure 6C).

Figure 6.

(A) Schematic of 3D cell culture model. (B) Results of cell viability (expressed as %) using a 3D cell culture model to study the effects of leached PAC-DESIGNERs. Different Greek letters show statistically significant differences between the time points (1, 3, and 5 days). The statistical difference threshold was set at p<0.01. (C) Live and Dead assay images of the effect of leached PAC-DESIGNERs over the DPCS after 1, 3, and 5 days, no numerical or morphological differences were observed among groups.

4. DISCUSSION

Biomodulation of dentin using PACs leverages exogenous collagen cross-links to enhance the mechanical stability of the tissue and mitigate the biodegradation rates of collagen through multifaceted interactions with dentin matrix components [11,27,28]. Four PAC-DESIGNERs from two priority study plants, Pinus massoniana and Cinnamomum verum, were found to enhance the mechanical properties of dentin, improve the outcomes of resin-dentin adhesion, and were fully biocompatible. Therefore, all null hypotheses were rejected in the present study.

The AB-type DESIGNERs were formulated to be enriched in different major trimeric and tetrameric PACs with distinct configurations related to inter-flavan linkages (IFLs) AB, ABA and ABB (Figure 1). The use of 6.5% PAC-DESIGNERs neutral primers resulted in a significant increase of the dentin-resin micro-tensile bond strength when compared to the unprimed control group (Figure 2). These bioadhesive promoting properties of PAC-DESIGNERs can be attributed to their physicochemical characteristics as the hydrophilic phenol groups in PACs are associated with hydrophobic aromatic rings, this hydrophilic/-phobic balance apparently enhances bonding interactions between the hydrophobic resin and the dentin matrix [3]. The observed effects can also be attributed to other factors such as an increase in the mechanical properties of the dentin matrix at the hybrid layer and underlying dentin [29]. These findings could also explain the observed fracture pattern at immediate and 1 year aging (Figure 2), which showed the majority of failures were at the adhesive at resin pattern for all the DESIGNER-treated groups, meaning the bond failure happened at the composite resin and not at the PAC-reinforced hybrid layers. In addition, the outcomes corroborate previous reports showing an increase in the resin-dentin bond strength after biomodulation with a PAC oligomer enriched grape seed extract [2,3,9]. The degree of oligomerization and the IFLs did not have an apparent effect on the immediate resin-dentin bond strength, meaning that all DESIGNER primers increased the bond strength compared to the control. However, after 1 year aging, CV-ABB was the only DESIGNER primer to exhibit a significant decrease in the dentin-resin bond strength (Figure 2). Interestingly, CV-ABB elicited the lowest increase in the stiffness of dentin matrix when compared to other DESIGNERs (Figure 4). CV-ABB is enriched with tetramers and B-type linkages. B-type linkages are more flexible than A-type IFL due to their single-bond nature as opposed to double-bond linkages (Figure 1). While the less pronounced dentin biomodification induced by CV-ABB did not affect the immediate TBS, it appears to have induced less stability of the dentin-resin interface, and this outcome could be attributed to the higher proportion of B-type linkages.

Figure 1.

The PAC Block Arrays (PACBARs), full chemical structures, and heat maps showing the relative concentrations (%) of the different degrees of polymerization (DP) and the most predominant molecule in the four PAC-DESIGNERS from PM and CV PM-AB (A) and PM-ABA (B); CV-AB (C) and CV-ABB (D). Polymer refers to DP>=5, dimers (DP=2) were absent in all DESIGNERs.

The degree of conversion (DC) provides insights into the resin monomer’s conversion of carbon-carbon double to single bonds [30], crucial for maintaining the polymerization behavior of the resin and ensuring adequate physical properties [31]. Here PACs were incorporated into resin to determine if contact with DESIGNERs would influence the polymerization process. We found that PAC-DESIGNERs, regardless of concentration, did not affect the degree of polymerization of the solvated experimental adhesive. Interestingly, the DESIGNERS even led to a DC increase for the PM-based DESIGNERS compared to control (Figure 3). Few studies have suggested a positive copolymerization effect between flavonoids and the bonding agent, resulting in ester-type chemical bonds that improve the mechanical properties of the adhesive layer [32–35]. Accordingly, such factors may have increased the monomer conversion of the experimental adhesive, independent of the concentration of the PM source, which mainly explains the higher DC values observed in the present study. However, this increased effect of the DC was only observed in PM, therefore additional factors may have contributed to the observed results.

To broaden the investigation of PAC-DESIGNERs in dentin biomodulation, the viscoelastic characteristics of the modified dentin matrix was examined using dynamic mechanical analysis (DMA) assay [7]. This method provides measurement outcomes of storage (E’), loss (E”), complex (E*) moduli, and damping capacity (tan 𝜹) [5,7]. PAC-DESIGNERs from PM induced the highest increases in the E’ and E* of the dentin matrix when compared to CV DESIGNERs (Figure 4). The findings point to differences in the IFL as an explanation of the distinct behavior between the tetrameric compounds from PM and CV. The PM ABA tetramer has two double linkages (C→O→C; A-type), giving these compounds more stability than the CV tetramer ABB, which bears two B-type linkages that involve two single-bond connections (C→C; B-type) [10]. The degree of polymerization presented an effect on these properties as well, showing that DP3 DESIGNERs elicited higher potency than DP4 DESIGNERs. The difference between these two DPs can be explained by the suitability of the major compounds in fitting into interfibrillar spaces promoting a higher efficiency in PAC-matrix interactions, which favors the smaller trimeric compounds [5,10]. In addition, the different molecular polarities based on the number of hydroxyl groups and aromatic rings influences the hydrogen bonding [5].

The PM-AB, PM-ABA, and CV-AB DESIGNERs elicited the highest increase in loss modulus (E”) of the dentin matrix when compared CV-ABB (Figure 4). The lower viscosity observed in the CV-ABB group when compared with the other biomodulated groups can be explained by the presence of two B-type IFLs, which provide more conformational flexibility than the A-type IFL [36]. Nevertheless, the tetramer CV-ABB compound has shown a potency to increase the viscoelastic properties of the dentin matrix when compared with other compounds configuration. This can be attributed to the presence of a 4→6 linkage, which is characteristic of the B-type IFL [36,37]. The difference in the chemical structures related to the IFLs can explain the effect of the CV-ABB on the dentin matrix viscosity in the present study [36,37]. The viscoelastic properties affect the tissue’s damping capacity, and an increase in this component means that the tissue has more ability to dissipate energy during oscillatory stress [5]. Both PAC-DESIGNERs from CV increased the damping capacity compared with the other groups (Figure 4). This fact shows the impact of the source and their respective PAC constituents on the collagen capacity to absorb or dissipate energy [5,10,38].

Dentin biomodification by PAC-DESIGNERs biomaterials also resulted in conformational changes to the secondary collagen structure (Figure 5) and increased intermolecular and interfibrillar cross-links attributed to increased tissue stiffening. Decreases to the amide II/CH₂ ratios were observed in all biomodulated dentin, with the highest decrease observed for PM-AB on the amide III/CH₂ ratio. The absorption band of amide II corresponds to the quantity of -NH₂ present in collagen [39–41]. A reduction in the ratio of amide II/CH₂ implies a conversion of free -NH₂ to N-H groups due to collagen crosslinking [3,42]. The amide III band is indicative of the stability of the collagen triple helix [42], and alterations in amide III are associated with changes in protein conformation. The amide III/CH₂ ratios between PM-trimer AB were the only significant difference, suggesting a relationship between the IFL type and the efficiency of PAC phenyl groups in inducing protein conformational changes. It was noticed that the AB-type trimers decrease the amide III/CH₂ ratios, while the tetramers induced an increase on these same ratios. Given the critical role of hydrogen bonds in collagen stabilization and enhancement of tissue biomechanics, the energetic interaction of these bonds can affect tissue hydration, collagen residue substitutions, and molecular stability [43,44]. Moreover, differences were observed in characteristic collagen FTIR bands for all DESIGNER treatments. A close similarity in the IR patterns between 1,400 and 800 cm⁻¹ was noticed between the two CV DESIGNERs, as well as with the spectra of the two PM DESIGNERs. This may suggest that the modifications resulting from the PM and CV DESIGNERs might be impacted by the main monomer found in each of the respective materials, i.e., catechin and epicatechin, respectively [45,46].

The biocompatibility of dentin-leached PAC compounds was studied using a 3D cell culture model of DPSCs cultured on a type-I collagen scaffold to mimic the dentin-pulp complex microenvironment [21,22,47]. Leached compounds from PM-ABA decreased cell viability after 3 and 5 days (Figure 6). As noted in prior work, reactive phenol groups from enriched oligomeric PAC from grape seed extract and their chemical structure may lead to increased cell differentiation via increased anti-oxidative activity [47] (2). As a response to this increase in the cell differentiation, it is possible to observe a decrease in the DPSC viability [47,48]. Importantly, the present findings confirm that the PAC-DESIGNERs are fully biocompatible with DPSCs, and that the PACs can possibly even promote an upregulation of key genes involved in differentiation, dentinogenesis, and biomineralization [47]. A cell proliferation process was observed for both CV DESIGNERs after 3 days. Epicatechin is the basic monomeric unit of CV oligomers, therefore leached CV compounds may have played a role in the higher cell proliferation results. Epicatechin, in a specific concentration range (0.01 to 0.1 M) was found to increase DPSCs proliferations after 3 days when compared to higher concentration [45]. This cell behavior has been attributed to the presence of epicatechin in enriching the collagen structure, allowing cells to have greater fluid exchange, greater nutrition and, consequently, the proliferation process [45]. Further studies are needed to analyze the potential of PAC-DESIGNERs in differentiating DPSCs and evaluating protein expression could further support these findings. Overall, the investigated DESIGNERs seem to exhibit no local toxicity to cells and, therefore, are safe for oral topical usage.

5. CONCLUSION

Biomodulation by PM and CV PAC-DESIGNERs can enrich the biomechanical and bonding properties at the resin-dentin interface and increase the dentin-resin bond strength. Most remarkable were the stable resin-dentin bond strength elicited by DESIGNER primers, except for DESIGNERs higher proportion of B-type linkages. The trimers and tetramers from Pinus massoniana improved the degree of conversion of the experimental adhesive. Trimers from both plants showed superior performance on the viscoelastic properties of the dentin matrix when compared with their tetrameric counterparts from the same plant. All PAC-DESIGNER treatments showed good cell biocompatibility, especially the DESIGNERs from Cinnamomum verum. The multifaceted benefits of PAC-DESIGNERs demonstrated in the present study bring us closer to the availability of advanced biomaterials that can improve the durability of adhesive restorations clinically.

Highlights.

• The DESIGNER concept is an innovative approach to standardize the formulation of active Proanthocyanidins (PAC).

• PAC-DESIGNERs primers promote high dentin-resin bond strengths.

• Trimeric and Tetrameric PAC-DESIGNERs play a modulatory role on the dentin matrix.

• PAC-DESIGNERS derived from Pinus massoniana improve the model adhesive’s conversion degree.

• All PAC-DESIGNERs presented high biocompatibility with stem cells in the dentin-pulp complex.