Ca/P-PILP was synthesized at high Ca2+ concentration without precipitation.
Electron Probe Micro Analyzer was used to investigate element contents.
Artificial caries dentin lesion could be biomimetic remineralized by Ca/P-PILP.
Remineralization could improve the bond strength of artificial caries dentin lesion.
To assess the ultrastructural change of demineralized dentin collagen during calcium phosphate polymer-induced liquid precursor (Ca/P-PILP) mediated remineralization process and to evaluate the biomimetic remineralization potential of high concentration Ca/P-PILP at demineralized artificial caries dentin lesion, additionally to investigate the bond interfacial integrity as well as the bonding strength of the biomimetic remineralized artificial caries dentin lesion.
Demineralized dentin collagen of 5 μm thick was biomimetically remineralized with low, medium concentration Ca/P-PILP for 10 days and high concentration Ca/P-PILP for 10, 15, 20 days. Artificial caries dentin lesion at a thickness of 150 ± 50 μm was biomimetically remineralized with high concentration Ca/P-PILP for 20 days. The biomimetic remineralization of demineralized dentin collagen was observed by scanning electron microscopy (SEM). The biomimetic remineralization intensity and depth of artificial caries dentin lesion was assessed by Electron Probe Micro Analyzer (EPMA). The bonding interfacial integrity between remineralized artificial caries dentin and composite resin was observed by Swept-source optical coherence tomography (SS-OCT) and the bonding strength of remineralized artificial caries dentin was evaluated by micro-tensile bond strength analysis (μTBS).
Solely PAA-PASP solution and solely saturated Ca/P solution can’t achieve dentin collagen remineralization. Increased concentration of Ca/P-PILP and prolonged remineralization time can enhance the biomimetic remineralization intensity of demineralized dentin collagen. After treating with high concentration Ca/P-PILP, a 150 ± 50 μm thick layer of demineralized artificial caries dentin lesion was not fully remineralized, and the biomimetic remineralization intensity reached up to 88.0%. Furthermore, a better bonding interfacial integrity with less microgap and increased bond strength at both baseline level and aging level were observed when artificial caries dentin lesion was biomimetically remineralized with high concentration Ca/P-PILP.
Biomimetic remineralization of demineralized caries dentin lesion promotes its clinical properties for resin composited adhesive restoration.
Dental caries is a dynamic process of continual changes of pathological demineralization and physiological protective remineralization of dental hard tissue which is mainly composed of hydroxyapatite (HA), collagen and non-collagenous matrix proteins [ ]. According to the conservative and preventive concept raised by minimal invasive dentistry (MID), the outer layer of carious-infected dentin (CID) should be removed, however the inner layer of carious-affected dentin (CAD) should be retained to avoid potential pulp complication [ ]. Nevertheless, the remaining CAD demonstrates a lower bond strength to resin composite compared to non-carious dentin [ , ]. This is attributed to an increased water content [ ], lower physical properties [ ] and a thicker hybrid layer [ , ] of demineralized dentin lesion. In addition, demineralized dentin maintains excessive water from collagen compartments that cannot be entirely replaced by adhesives, which may lead to nanoleakage within hybrid layer [ ] and further increase the risk of secondary caries [ ]. Besides, without the protection of minerals, dentin collagen is more susceptible to endogenous proteases which may cause collagen degradation [ ]. Thus, remineralizing demineralized dentin lesion is of paramount importance.
Biomimetic remineralization of demineralized dentin with Calcium phosphate polymer-induced liquid precursors (Ca/P-PILP) is a promising strategy which is based on a non-classical crystallization pathway theory, by backfilling demineralized dentin with amorphous calcium phosphate (ACP) and non-collagenous protein (NCP) analogues including biocompatible polymers polyacrylic acid (PAA), polyaspartic acid (PASP), and so on [ ]. This remineralization does not rely on the epitaxial growth of residual apatite in the demineralized dentin [ ]. In the process of biomimetic remineralization, Ca/P-PILP was stabilized by NCP analogues [ ]. At the early stage of Ca/P-PILP, ACP nanoparticles were initially formed [ ], and NCP analogues prevented these ACP nanoparticles from aggregation and auto-transformation into apatite prior to entering into intrafibrillar compartment of dentin collagen; this followed by the nucleation and maturation of ACP nanoparticles along the intrafibrillar and extrafibrillar spaces [ , ].
Previous studies have investigated individually the function of PAA or PASP [ , , ]. In this study, we use double NCP analogues containing both PAA and PASP to sequester and stabilize Ca/P-PILP. The maximal amount of Ca 2+ that can be chelated by PAA-PASP (26.1 and 13.0 mg/mL, respectively) is 43.5 × 10 −3 M. This Ca/P-PILP could successfully realize osteoporotic bone biomimetic remineralization according to Yao’s research [ ], and in this study, its biomimetic remineralization of artificial caries dentin lesion will be investigated. Moreover, the effect of this double NCP analogues process-directing agent on internal adaptation of bonding interface and the bonding strength of the remineralized artificial caries lesion will be evaluated.
The objective of this study was therefore, to assess the ultrastructural change of demineralized dentin collagen during Ca/P-PILP mediated remineralization process and to evaluate the biomimetic remineralization potential of Ca/P-PILP on artificial caries dentin lesion, additionally to investigate the bonding interfacial integrity as well as the bond strength of the biomimetic remineralized artificial caries dentin lesion with resin composite. The null hypotheses were (1) that solely PAA-PASP solution and solely saturated Ca/P solution can’t achieve dentin collagen remineralization, (2) that the concentration of Ca/P-PILP and the remineralization time have no influence on the degree of dentin collagen biomimetic remineralization, (3) that artificial caries dentin lesion can’t be completely remineralized with Ca/P-PILP, (4) that Ca/P-PILP don’t improve the bonding interfacial integrity nor (5) the bond strength of artificial caries dentin lesion with resin composite.
Materials and method
The study was approved by local Ethics Committee of the School and Hospital of Stomatology Wuhan University, China. 55 extracted non-carious human third molars were collected with the patients’ informed consent and stored in 0.5 wt% thymol (Sigma-Aldrich, USA) solution at 4℃.
For demineralized dentin collagen model, a 1 mm thick dentin disk was prepared by 2 parallel cuts perpendicular to the longitudinal axis of each tooth using the low speed Isomet saw (Buehler, Lake Bluff, IL, USA) under running water. The upper surface was polished with 1200-grit silicon carbide paper under running water for 2 min, then etched with 37% phosphoric acid (DenFil™ Etchant-37, VERICOM CO LTD, Korea) at room temperature (25℃) for 20 s and rinsed thoroughly with distilled water for 2 min to create the a 5 μm thick layer of demineralized dentin collagen matrix.
For artificial caries dentin lesion model, the occlusal third of the crown was removed using the low speed Isomet saw. Then each tooth was polished with 1200-grit silicon carbide paper under running water for 2 min. Furthermore, 2 layers of acid-resistant varnish were applied to cover all the outer surfaces of each dentin specimen except the occlusal surface. Partially demineralized dentin at a depth of 150 ± 50 μm starting from the uncovered occlusal surface was created by a pH-cycling protocol [ ] to imitate demineralized caries dentin lesion [ , ] . Each tooth was immersed in 20 mL demineralization solution (50 mM Acetic acid, 2.2 mM CaCl 2 , 2.2 mM NaH 2 PO 4 , pH = 4.8) for 8 h, followed by 20 mL remineralization solution (1.5 mM CaCl 2 , 0.9 mM KH 2 PO 4 , pH = 7.0) for 16 h. The de-/remineralization cycle was repeated for 2 weeks and the solution was refreshed for each cycle (24 h per cycle). pH cycling consists of both demineralization and remineralization processes was selected to prepare artificially demineralized dentin, this mimic the dynamic de-/remineralization process during cares development [ , ]. And that After pH-cycling, teeth were rinsed with deionized water for 30 min.
Synthesis of biomimetic remineralization solution
To synthesize Ca/P-PILP at a high concentration without precipitation, a double NCP analogues process-directing agent of PAA-PASP were used. 2.0 mL of 0.1 M Na 2 HPO 4 (Sigma-Aldrich, USA) solution was mixed with 0.4 mL solution consisting of 0.3 g/mL PAA (average Mw = 450 kDa; Sigma-Aldrich, USA) and 0.15 g/mL PASP (Mw = 6−8 kDa; Aike Agent, China), the mixing process was done using magnetically stir at room temperature for overnight. Then the mixed solution was successively dropped into another mixed solution containing 2.0 mL of 0.1 M CaCl 2 (Sigma-Aldrich, USA) and 0.2 mL of 0.3 g/mL PASP with vigorous stirring. Finally, the pH of the finally mixed solution was adjusted to 7.4 using 3 M NaOH (Sigma-Aldrich, USA). The ACP nanoparticles produced by aforementioned procedure are about 1 nm in size [ ].
For demineralized dentin collagen model, 24 demineralized dentin disks were divided into 4 groups:
Non-treated (NT) group (n = 3): The demineralized dentin disks were observed by scanning electron microscopy (SEM; Zeiss SIGMA, England) without any treatment.
Ca/P-PILP group (n = 15): The freshly prepared Ca/P-PILP was diluted with distilled water to 10%, 50% and 100% to represent the low, medium and high concentration Ca/P-PILP respectively. 9 demineralized dentin disks were distributed among the 3 solutions evenly and randomly and each disk was remineralized in 5 mL Ca/P-PILP for 10 days at 37℃. Another 6 demineralized dentin disks were remineralized with high concentration Ca/P-PILP for 15 days and 20 days, respectively. The solution was refreshed every 5 days.
PAA-PASP control group (n = 3): In this group the treatment solution only contains PAA and PASP, and the concentrations of PAA and PASP were the same as the Ca/P-PILP. Also, each disk was remineralized in 5 mL PAA-PASP solution for 20 days at 37℃ and the solution was refreshed every 5 days.
Ca/P control group (n = 3): In this group the treatment solution involves saturated calcium and phosphate provided by CaCl 2 and Na 2 HPO 4 solutions without NCP analogues, and the concentrations of Ca and P were the same as the high concentration Ca/P-PILP solution. Also, each disk was remineralized in 5 mL Ca/P solution for 20 days at 37℃ and the solution was refreshed every 5 days.
After the treatment, all specimens in 4 groups were washed with deionized water for 20 min, then stored at 4 ℃. Consequently, the potential remineralization and ultrastructural changes were observed by SEM.
For artificial caries dentin lesion model, after 2-week demineralization, 40 demineralized specimens were divided into 4 groups:
Non-treated (NT) group (n = 10): The demineralized teeth were assessed by field-emission-gun Electron Probe Micro Analyzer (Feg-EPMA; JXA-8530 F, JEOL, Japan) (n = 3), Swept-source Optical Coherence Tomography (SS-OCT; Shengqiang, China) (n = 3) and Micro-tensile bond strength (μTBS; Bisco, USA) (n = 4) without other treatment.
Ca/P-PILP group (n = 10): Each demineralized tooth was remineralized in 5 mL high concentration Ca/P-PILP without dilution for 20 days at 37℃ and the solution was refreshed every 5 days.
PAA-PASP control group (n = 10): Each demineralized tooth was remineralized in 5 mL PAA-PASP solution (same as demineralized dentin collagen model) for 20 days at 37℃ and the solution was refreshed every 5 days.
Ca/P control group (n = 10): Each demineralized tooth was remineralized in 5 mL Ca/P solution (same as demineralized dentin collagen model) for 20 days at 37℃ and the solution was refreshed every 5 days.
After treatment for teeth in artificial caries dentin lesion model, all specimens ware rinsed with deionized water and stored at 4 ℃. For each group of artificial caries dentin lesion model, 3 teeth were assessed by EPMA, 3 teeth were examined by SS-OCT, and 4 teeth were used for μTBS.
All dentin disks used in demineralized dentin collagen model were fractured into 2 halves. One half of the disk was prepared to observe the surface ultrastructure of dentin collagen and the other half was prepared to observe the cross profile of dentin tubules. The specimens were fixed with 4% glutaraldehyde (Sigma-Aldrich, USA) for 12 h at 4℃. Subsequently, specimens were rinsed with deionized water and then successively dehydrated with increasing concentration of ethanol (25, 50, 75, 95, 100%). After dehydration and drying with Critical Point Dryer (Leica EM CPD300, Germany), the specimens were sputter-coated with gold (JFC 1600, JEOL, Japan) and observed under SEM at the condition of 10 kV voltage.
Artificial caries dentin lesion model was examined by EPMA at a spatial resolution of ±0.05 μm. Teeth were cross-sectioned parallelly to the longitudinal axis of each tooth using the low speed Isomet saw under water cooling. The cross profile of each specimen was polished by 400–1200 grit silicon carbide paper with running water, then fixed, dehydrated and dried as mentioned in SEM samples preparation. Subsequently, all specimens were sputter-coated with carbon layer (JEE-420, JEOL, Japan). For each specimen, the depth and intensities of chemical elements Ca and P from the occlusal surface to deeper dentin along a 300 μm long line-scan were detected. The percentage of mean Ca intensity in remineralized dentin as compared with that in deeper unaffected sound dentin was quantificationally calculated as the relative remineralization intensity (I RM ). Besides, one representative specimen per group was regionally mapped by Ca, P and carbon (C) for the remineralized area and the corresponding back scattered electron (BSE) images were recorded.
Swept-source optical coherence tomography (SS-OCT)
3 teeth in each group of artificial caries dentin lesion model were examined. In this experiment, one-step self-etch adhesive (Adper™ Single Bond Universal, 3 M, USA) and nanocomposite resin (Filtek™ Z350 XT, 3 M, USA) was used. Following the manufacturer’s application guide, 1 mm composite resin was build-up on the dentin surface of each specimen. The interfacial adaptation of biomimetic remineralized artificial caries dentin lesion was observed by SS-OCT. The light source in the system sweeps a center wavelength at 1310 nm (bandwidth 50 nm) with a scan rate of 60 kHz. The axial resolution of the OCT system is 11 μm in air, which is equivalent to 7 μm in oral hard tissue and resin composite, assuming a refractive index of about n = 1.5. The output is an image of 362 * 729 pixels. The scanning probe connected to the SS-OCT was placed over the top of composite resin, and the light beam was projected and cross-sectionally scanned across the specimen. The first SS-OCT image of each sample was taken parallel to the buccolingual plane of the teeth. After the image was taken, the specimen was rotated 30° clockwise, then the next image was taken. 6 images were taken for each sample. When there is a interfacial defect between dentin and composite, the presence of air in the microgap can cause different reflection of light, which can be visualized as a bright spot at the bonding interface on the SS-OCT image. The percentage of microgap as compared with the total length of the interface was measured in pixels using ImageJ. The mean value per group as well as standard deviation was determined and the groups were compared statistically by one-way ANOVA at a significant level of α = 0.05 using SPSS 22.0 software package (SPSS, Chicago, IL, USA).
Micro-tensile bond strength analysis
4 teeth in each group of artificial caries dentin lesion model and another 4 non-caries health teeth (ND group) were examined. In this experiment, one-step self-etch adhesive (Adper™ Single Bond Universal, 3 M, USA) and nanocomposite resin (Filtek™ Z350 XT, 3 M, USA) was used following the manufacturer’s application guide. 4 mm composite was build-up incrementally on the dentin surface of each specimen. After storage in water at 37℃ for 24 h, each specimen was sectioned perpendicular to the adhesive surface, the section area of each beam was approximately 1 mm 2 . For each group, 20 beams randomly selected and tested for μTBS as the baseline; and another 20 beams were selected and thermocycled (60 s of immersion, alternatively, in a 5 and 55 °C water bath) for 100,000 cycles. The beams were sticked with cyanoacrylate to the testing apparatus, and measured at a tensile force of 1 mm/min. The data of μTBS was analyzed by one-way ANOVA at a significant level of α = 0.05 using SPSS 22.0 software package.
Biomimetic remineralization of demineralized dentin collagen
SEM images illustrated changes during biomimetic remineralization process of demineralized dentin collagen. After treating with 37% phosphoric acid for 20 s, demineralized collagen fibrils were desultory collapsed and shrunken due to the lack of inorganic minerals ( Fig. 1 A-B). The demineralized collagen fibrils were desultory collapsed and shrunken due to the lack of inorganic minerals. For specimens in the PAA-PASP and Ca/P control groups, after treatment with solely NCP analogues ( Fig. 1 C-D) and solely saturated Ca/P solution ( Fig. 1 E-F) for 20 days, no obvious remineralization was found compared to Ca/P-PILP group. For specimens in Ca/P-PILP group, an increase in Ca/P-PILP concentration from 10% to 100% enhanced the degree of biomimetic remineralization ( Fig. 1 G-L, Fig. 2 A-C). At a low concentration of 10%, the demineralized dentin collagen fibrils were only slightly remineralized after 10 days ( Fig. 1 G-H, Fig. 2 A). Collagens were less like to collapse as a small amount of ACP nanoparticles have infiltrated into collagen gap zone and displaced the water. The banding structure indicated that the lowest concentration of Ca/P-PILP solution could not induce sufficient intrafibrillar remineralization since the banding pattern of collagen would be disrupted by intrafibrillar mineralization [ ]. No obvious extrafibrillar was observed and the interfibrillar spaces were relatively wide. After remineralization for 10 days with Ca/P-PILP in medium concentration of 50%, the diameter of the collagen fibrils increased ( Fig. 1 I-J, Fig. 2 B). With the support of minerals, the remineralized collagen became continuous and well-organized. The partially remineralized collagens have acquired enough support to prevent them from dehydration shrinkage. Remineralizing the collagen fibrils with Ca/P-PILP in high concentration for 10 days achieves both intrafibrillar and extrafibrillar remineralization ( Fig. 1 K-L, Fig. 2 C). The banding structure was not as distinct as remineralized in low concentration of Ca/P-PILP, and the interfibrillar spaces were decreased as a result of extrafibrillar remineralization on the outer surface of collagen fibrils. With regard to high concentration of Ca/P-PILP, the degree of remineralization was strengthened at a prolonged remineralization time (15 d and 20 d) ( Fig. 1 M-P, Fig. 2 D-E). All samples showed remineralization for entire layer of demineralized dentin collagen fibrils ( Fig. 2 ). Therefore, the first hypothesis that solely PAA-PASP solution and solely saturated Ca/P solution can’t achieve dentin collagen remineralization was accepted. And the second hypothesis that the concentration of Ca/P-PILP and the remineralization time have no influence on the degree of biomimetic remineralization of demineralized dentin collagen was rejected.