Effects of Streptococcus mutans and its fluoride resistant strains on the adhesion of CAD/CAM ceramics to teeth and resin

2.1 Strain identification

Bacteria frozen at - 80 degrees were recovered. To ensure the reliability of the bacteria, 16s rDNA was amplified and sequenced, and finally phylogenetic tree analysis was performed.

Bacteria were dissolved in 50 μ l TaKaRa Lysis Buffer for Microorganism to Direct PCR (TaKaRa, Japan), denatured and centrifuged to take the supernatant as a template. The target fragment was amplified by PCR using 2 × TransTaq ® High Fidelity (HiFi) PCR SuperMix I (TransGen Biotech, China). Using 27F and 1492R (Table 1) as primers, DNA sequencer (model: 3730XL, ABI, USA) was used for sequencing. The sequencing results were entered into Genebank, and the BLAST functional components were used to compare the measured gene sequences with the 16S rDNA sequences in the Genebank database for homology. The 16S rDNA sequences of the 22 strains with higher scores in the BLAST results were selected, and the phylogenetic tree analysis was performed according to Neighbor-Joining using MEGA5.2.1 software.

Inoculate S. mutans on BHI medium containing NaF 50 μ g/ml, anaerobically culture at 37 ∘ C in anaerobic condition (95%N 2 , 5%CO 2 ) for 36 h. Then serially passage them in mediums whose concentration of NaF increased with a trend of 50 μ g/ml in each generation, until 1000 μ g/ml, screen out colonies with complete morphology, and then inoculate them on mediums without NaF for 20 consecutive generations. Finally medium containing 1000 μ g/ml NaF was used to screen out fluoride-resistant strains.

2.3 Observation of cracks in the bonding interface of ceramic- adhesive-tooth complex

2.4 Observation of cracks under a stereo microscope

After 14 days, the specimens were taken out and placed in sterile distilled water for 10 min of ultrasonic vibration. Under a stereo microscope, the bonding interfaces were observed with a magnification of 20 times. Two interfaces were observed: a. ceramic-adhesive interface, b. adhesive-tooth complex (enamel/dentin/flowable resin) interface. Save the photos, and use Image-Pro Plus 4.0 software to calculate the percentage of crack length to the total length of bonding surface on each interface.

2.5 Observation of ceramic-adhesive interface by scanning electron microscope (SEM) magnified 1000 times

The specimens were dehydrated in ethanol with gradient concentration, sprayed with gold. The bonding interface was observed by SEM under 20 kV accelerating voltage. The experimental flow chart is shown in Fig. 1D.

2.6 Interface observation and micro-tensile test of the direct bonding between ceramic and resin with adhesive

3.1 Identification of S. mutans UA159 and induction of its fluoride resistant strains

The colony form of S. mutansua159 is round, light yellow and opaque, and the surface is slightly rough, the boundary is irregular (Fig. 3A).

OPEN IN VIEWER

Figure 3B shows the sequencing result and the phylogenetic tree of 16sRNA.

It can be seen that the genetic distance between the sample and S. mutans is the closest (Fig. 3C). Through further sequence alignment with Genebank Blast, it is found that the homology with S. mutans UA159 is 100% (Fig. 3D), which further proves that the recovered strain in this experiment was S. mutans UA159.

After continuous passage, strains with stable morphology were screened out by BHI medium with NaF. Colonies of fluoride resistant strains are round, slightly larger than the wild strain (Fig. 3E).

OPEN IN VIEWER

Figure 4A shows that under the stereomicroscope, the three ceramics are bonded to tooth complexes through adhesive. We mainly observe two interfaces: the upper, ceramic-adhesive interface and the lower, adhesive-tooth complex interface. Figure 4B further shows the cracks in the lower interface.

Figure 4C shows the cracks in ceramic-adhesive interface. It can be seen that compared with the control group, UA and FR cause significantly more cracks between ceramic and adhesive, and the influence of wild strain UA is greater than that of fluoride resistant strain. Compared the three kinds of ceramics with each other, cracks between resin-matrix ceramic and adhesive are higher than those between glass ceramic IPS Empress CAD and adhesive. Lava Ultimate-adhesive interface has the most cracks.

Figure 4D shows the cracks in the adhesive-tooth complex interface. Compared with the control group, the cracks of the specimens immersed in bacterial solution increased significantly, but there was no significant difference between FR group and UA group.

Among adhesive-enamel interface, adhesive-dentin interface and adhesive-flowable resin interface in the bacterial solution (UA and FR groups), cracks in adhesive-flowable resin interface are the lowest, while cracks in adhesive-dentin interface occur most, cracks in adhesive-enamel interface are in the middle, indicating that both UA and FR can cause cracks. Dentin is the most fragile part when bonding with adhesive in S. mutans environment.

3.3 Observation of ceramic-adhesive interface under SEM magnified 1000 times

It can be seen from Fig. 4E that glass ceramic impress IPS is closely bonded with adhesive, and there are fewer cracks after soaking in bacterial solution compared with the other two resin-matrix ceramics. In addition, the bonding surface of glass-ceramic is wavy under the electron microscope, forming a lock like structure with the adhesive, but the one of resin-matrix ceramic is relatively flat.

Irregular leucite particles can be seen in the ceramic matrix of glass ceramic, and there is no obvious crack even if there is a bacterial condition. The structure of Lava Ultimate shows that dispersed fillers are distributed in a relatively uniform resin matrix. After soaking in bacterial solution, small cracks can be seen between the resin matrix, as shown by the arrow. In the meanwhile, the visual field of Vita Enamic is relatively messy. Resin matrix is distributed between the brighter glass ceramic skeleton, and there are fewer cracks after being soaked in bacterial solution compared with Lava Ultimate. A small amount of fillers can be seen in the cracks between the two resin-matrix ceramic and adhesive, especially Lava Ultimate. There are many cracks in the adhesive part of each group.

OPEN IN VIEWER

Figure 5A shows that the layers of 1a, 2a and 3a in the control group are closely bonded without large cracks. However, specimens immersed in bacterial solution UA and FR have large and irregular cracks in the interface between ceramic and adhesive, and even the ceramic and adhesive are completely separated. Moreover, cracks of glass-ceramic group (IPS group) are significantly fewer than those of resin-matrix ceramic (Lava Ultimate and Vita Enamic groups) in ceramic-adhesive interface. The interface between adhesive and flowable resin is still tightly bonded without cracks.

3.5 Micro tensile test

Regardless of the type of ceramics, the micro tensile strength of UA and FR groups is lower than that of the control group, but there is no significant difference between UA and FR groups, indicating that S. mutans UA159 and its fluorine resistant strain weaken the bonding strength, but the degree of weakening is similar. In addition, in the control group, the bonding strength of glass ceramic IPS Empress CAD is significantly greater than that of the two resin-matrix ceramics, but there is no significant difference in the tensile strength between Enamic and Lava Ultimate. The strength of IPS Empress in UA and FR solution is slightly higher than that of the other two ceramics in UA and FR (Fig. 5B).

4.1 Effect of S. mutans on ceramic-adhesive interface

Resin-matrix ceramic is a kind of composite material rising in recent years. It combines the advantages of ceramic’s high strength and resin’s high ductility. Lava Ultimate and Vita Enamic, two products widely used in the market, are different from ordinary resins, because they are cured at high temperature and high pressure instead of traditional light and heat curing, which can effectively reduce the release of monomers. A study shows no monomer release was detected after the two products were soaked in acetone for 7 days [8]. However, the oral environment is very complicated and there are a large number of bacteria, which may affect the adhesion of the prosthesis. Bourbia et al. added resin into a bacterial solution of S. mutans UA159 for 14 days and found that resin monomer bis-HPPP could be detected, and showed obvious signs of surface erosion [5, 6]. The experimental results of Wang et al. also showed that S. mutans could weaken the bonding strength between resin and dentin [9]. Further studies show that esterase SMU_118 is an intracellular protein in S. mutans, which is expressed under cariogenic acidic conditions (ph-5.5) [6]. It can hydrolyze methacrylate [10], and plays a major role in biodegradation. It can act on the fragile ester bond of the resin and degrade the resin matrix into BisGMA, TEGDMA and UDMA, and their small monomers which will lead to cracks in the resin and affect the quality of bonding. These monomers can be detected by liquid chromatography, mass spectrometry and UV absorption [11]. In contrast, the polymer degradation activity of SMU_C gene knockout strain of S. mutansUA159 decreases significantly [12]. S. mutans UA159 was selected as the experimental strain because it has the strongest esterase activity with p-nitrophenol butyrate (p-NPB) as the substrate among all strains of S. mutans. This activity has been proved to be related to the degradation of composite resin [13].

Whether S. mutans has an effect on resin-ceramic composites has not been fully studied yet. Therefore, in this experiment, two typical composite ceramics on the market, Lava Ultimate and Vita Enamic, are used as the research objects, while the glass ceramics Empress CAD is equivalent to a comparative study. Although these two composite ceramics are composites of ceramics and resins, their structures are completely different. The former has a glass-ceramic filler embedded in a uniform resin matrix, while the latter has a mesh-like glass-ceramic frame, and the liquid resin penetrates into the mesh-like frame at high temperature. Because the latter has a relatively stable framework, and previous experiments have shown that the latter has slightly stronger mechanical properties than the former [14], it was assumed that the two have different resistance to cariogenic bacteria in the early stage of this experiment. The experimental results show that there are more cracks observed between the Lava Ultimate resin matrix under SEM, while there are very few cracks in the Vita Enamic, and the glass ceramic basically do not have cracks. It shows that S. mutans can affect resin-matrix ceramic, especailly Lava Ultimate, probably because the structure of Lava Ultimate is not as stable as Vita Enamic, a polymer infiltrated ceramic network (PICN) material. PICN has a framework in which resin and glass-ceramic are cross-linked. However, the micro-tensile test in this experiment showed that there was no significant difference in the micro-tensile strength between the two, probably because the generation of cracks was not enough to affect the effect of force on the ceramic bonding.

The adhesive used in this experiment is the double curing adhesive resin VariolinkN, which means the curing process requires both long enough light and certain chemical curing time. Therefore, the bonded specimens were stored in normal saline at 37 ∘ C for curing for 24 hours. The loss of adhesion of prosthesis in oral cavity is largely due to the poor control of the bonding process and insufficient curing time.

Variolink N is mainly composed of bisphenol A, glycidyl methacrylate (Bis-GMA) and polyurethane dimethacrylate, both of which belong to methacrylate. Therefore, as mentioned above, it can be biodegraded in the environment of S. mutans. In this experiment, cracks between adhesive were observed, indicating that S. mutans can destroy the polymer structure in adhesive, which is consistent with the conclusions drawn by previous scholars. This suggests that the bonding edge line of prosthesis should be minimized, because the long edge line will expose more bonding interfaces to the oral bacterial environment. We get the hint that when selecting the prosthesis, we must consider the bonding surface factor and try to reduce the percentage of the adhesive interface directly exposed to the oral cavity. If we have to adopt the design of long edge line, we should pay attention to the bonding process, in which the prosthesis should be continuously pressurized for enough time, patients can chew food only after the adhesive has solidified, and the excess adhesive must be removed. What’s more, the ability of total acid etching system to resist biodegradation is stronger than that of the acid etching system [15].

Experiments show that the percentage of cracks between ceramic and adhesive is related to ceramic materials, and cracks between glass ceramic IPS Empress CAD and adhesive are the least. Under SEM, it is also observed that compared with resin-matrix ceramic, the acid etching bonding surface of glass ceramic and adhesive form a lock like structure, the adhesive in the flowable state penetrates into the pits produced by hydrofluoric acid etching. Since there is such a lock like structure, so that the adhesive is not easy to separate from the ceramic body, which may explain why glass ceramic is less affected by S. mutans. In addition, the results of micro tensile test also show that the bonding strength between glass ceramics and flowable resin is stronger than that of resin-matrix ceramic, which also provides a certain basis for this conclusion. On the contrary, hydrofluoric acid cannot produce enough pits on the surface of resin-matrix ceramics, and the adhesive cannot form a lock like structure, which will be easier to separate from the ceramics.

4.2 S. mutans and adhesive-tooth complex interface

Pulp chamber retention is normally needed after the onlay restoration of posterior teeth finished with perfect root canal treatment. We usually use flowable resin to form a flat bottom in pulp chamber, which may help with the bonding of ceramic. In addition, if the tooth tissue is subgingival in Class II cavity, the flowable resin should be used to restore the gingival wall to supragingival position, which is called Cervical margin relocation, also known as gingival wall lifting. In these two cases, enamel, dentin and flowable resin are bonded to the ceramic through the adhesive, and then there are interfaces of adhesive-enamel, adhesive-dentin and adhesive-flowable resin which may contact with S. mutans in the oral cavity. The purpose of this experiment is to observe whether these three interfaces will be affected by S. mutans. Compared with the method of bonding enamel, dentin and flowable resin with ceramic respectively, this experiment is simpler and more convenient. In one specimen, three interfaces bonded with one ceramic can be observed at the same time, which reduces cumbersome steps of making specimens separately. What’s more, the interfaces can be controlled in the same bacterial environment, other influencing factors can be ignored. In addition, the design also simulates the interface directly exposed in the oral cavity during onlay bonding, which is closer to the physiological environment.

Total etching is the most common way for enamel bonding. Its principle is that phosphoric acid can cause micro pits on hydroxyapatite, thus adhesive can form a lock-like structure with enamel. In the meantime, the treatment of dentin can be either total etching or self-etching. The principle of the former is to demineralize the dentin surface and expose the collagen network structure, and then use pretreatment agent to facilitate the infiltration of resin into the network; the principle of the latter is to form a smear layer [16]. In this experiment, the total etching adhesive is used. It was shown that its bonding effect is stronger than that of self-etching adhesive [17], but the total etching steps are complex, and each step needs to be strictly moisture-proof. After phosphoric acid etching, the locking effect of enamel produces strong mechanical retention, but the demineralized dentin collagen fiber is easy to collapse, which is not conducive to bonding. Therefore, the bonding between adhesive and enamel is stronger than that between adhesive and dentin, which can explain that cracks in adhesive-enamel interface is fewer than dentin.

The flowable resin used in this experiment is TetricN-Flow, whose component is similar to the adhesive, Variolink N double curing resin adhesive, so the two are bonded most closely, which can explain that the cracks between the two are the least after soaking in bacterial solution.

S. mutans can not only degrade the adhesive through esterase, resulting in cracks between the adhesive and tooth tissue, but also produce hydrolase and bacterial collagenase. The content of organic in dentin can reach 30%, of which collagen accounts for the largest proportion. It is very likely to be affected by bacterial collagenase when exposed to a high concentration of bacterial environment.

4.3 Micro tensile test

The effect of a bonding system can be evaluated by bonding strength, which is often measured by tensile test and shear test. However, many studies have found that the fracture type of shear test is closer to cohesive failure (resin internal failure), rather than interface failure, which is more common in clinic [18]. The fracture type of tensile test is the interface failure, which is more conducive to evaluate the bonding strength of intraoral prosthesis. However, the results of tensile and shear tests are easily affected by specimen’s geometry and uneven stress distribution during force application, so micro tensile test is introduced. This test requires the bonding area to be close to 1 mm 2 , so the stress distribution will be more average, which can produce higher bonding strength and less cohesive failure, so it is more accurate and sensitive [19].

However, the micro tensile test has high requirements for the test piece. Therefore, the test is carried out only after the strip test piece without cracks is selected under the stereomicroscope.

The experimental results show that the bonding strength of glass ceramic Empress IPS is the largest, which is also consistent with the fewest cracks in ceramic-adhesive interface observed. Moreover, the bonding strength of glass-ceramic is still stronger than that of the other two resin-matrix ceramic even if it is under the environment of S. mutans. This suggests that traditional glass ceramics cannot be completely replaced by new emerging materials, and we should pay special attention to the removal of plaque when applying resin-matrix ceramics.

4.4 S. mutans and its fluorine resistant strain

Fluoride can achieve bacteriostatic effect by affecting the acid production, acid resistance and adhesion of bacteria. However, if fluoride preparations are used at specific sites for a long time, it may induce the growth of fluoride resistant strains, especially the fluoride resistant strains of S. mutans, which have stronger acid production and acid tolerance than their wild strains.

In this experiment, both kinds of bacteria can weaken the bonding strength of ceramic blocks and cause cracks, and there is little difference between the two. However, at the ceramic-adhesive interface, the wild strain cause more cracks than its fluorine resistant strain, which may be due to the change of esterase synthesis of fluorine resistant strains, the hypothesis needs to be verified by further experiments. This result also suggests that the use of fluoride such as fluoride toothpaste will not interfere too much with the bonding of resin based restorations at the bacterial level.

The results of this experiment can suggest that S. mutans, the most cariogenic bacterium in the oral cavity, can affect the bonding between ceramic and tooth tissue. In clinical practice, it is necessary to strictly control the bonding process to ensure moisture isolation and tight edges. In daily maintenance, it is necessary to ensure the cleanness of prosthesis and remove plaque in time.

Moreover, this experiment needs to further study whether the biodegradation of S. mutans occurs at the bonding interface. Liquid chromatography-mass spectrometry (LC-MS), which is commonly used in metabonomics, can be used to detect whether there are free monomers in composite ceramics.

In addition, it is necessary to further detect whether the gene encoding of the esterase SMU_C in the fluoride resistant strain of S. mutans mutates, whether the expression of SMU_C is changed under the condition of fluorine or no fluorine, so as to more accurately explain whether the biodegradability of the fluoride resistant strain of S. mutans changes.