**5. Discussion**

The present study developed a new composite consisting of nCaF2, DMAHDM and MPC, and investigated the mechanical, ion release and oral biofilm properties. The protein-repellant nCaF2+DMAHDM+MPC had slightly lower mechanical properties when compared to all other groups. However, nCaF2+DMAHDM had good mechanical properties matching those of a commercial composite, while possessing high levels of F and Ca ion release and a strong antibacterial effect. The new bioactive composite nCaF2+DMAHDM decreased biofilm CFU by four orders of magnitude over that of commercial control composite, and substantially reduced acid production to inhibit caries.

Critical to the development of any new composite restorative material are its mechanical properties. The flexural strength of EC was around (170 ± 38) MPa. With the addition of 15% nCaF2, the strength was reduced slightly to (157 ± 5) MPa. Those values are consistent with the results in a previous study [32]. The composite in that study had 65% total fillers and 20% CaF2, similar to the 70% total fillers and 15% nCaF2 composition in the present study. The flexural strength for the control group containing no nCaF2 in the previous study was (145 ± 9) MPa, while the group which had 20% nCaF2 had a flexural strength of 121 MPa. The nCaF2+DMAHDM and nCaF2+MPC had similar strengths of (95 ± 6), and (96 ± 5) MPa, respectively. Meanwhile, for nCaF2+DMAHDM+MPC, the flexural strength was around (64 ±3) MPa. Those are similar to the results of a previous study using the same filler mass fraction where the flexural strength was around 90 MPa when 3% MPC and 3% DMAHDM were added to the BT resin [43]. The decreased strength in nCaF2+DMAHDM+MPC could be due to two reasons. First, the BT base resin is composed of BisGMA and TEGDMA, which contain two reactive groups each. However, both DMAHDM and MPC are monomethacrylates and contain one reactive group. The addition of these monomethacrylates may substantially change the reaction kinetics and resulting crosslinked network, which can result in inferior mechanical properties. Another possible explanation for the decrease in mechanical properties is due to the hydrophilic nature of MPC. The specimens for flexural strength were stored for 24 h in water at 37 ◦C. The hydrophilicity of MPC makes it more likely that the composite would absorb water during the 24 h immersion prior to testing. This would likely lead to the reduction in flexural strength compared with the non-MPC groups.

Fluoride releasing materials have been known to render teeth more resistant to decay [45]. However, many available materials with fluoride-releasing properties, such as glass ionomer and resin modified glass ionomers, have inferior mechanical properties that do not serve the high load requirements for restorations in stress-bearing areas. Therefore, the incorporation of F ions into composite restorations would have the dual benefit of fluoride release and improved mechanical properties [46–48]. The current study synthesized nCaF2 through a spray-drying method, which produced nanoparticles of CaF2 with a median particle size of about 32 nm. The use of small amounts of nCaF2 fillers prevented the mechanical properties to be compromised. As seen in previous studies, the use of fluoride fillers with larger particle sizes require higher mass fractions of fillers to achieve significant fluoride ion release. This would lead to a decrease in the composite mechanical properties [49].

The incorporation of nanoparticles with a high surface area resulted in the release of high levels of F and Ca ions using a low nCaF2 filler level [33]. F ions foster remineralization by forming fluoroapatite [Ca5(PO4)3F] [50]. Furthermore, it was shown that the lower the pH, the higher the release of ions. In this study, the release of F ions was significantly increased with the integration of 15% nCaF2 at a pH of 7, measuring (0.04 ± 0.01) mmol/L. This was comparable to the 0.03 mmol/L F release level in a study that had a pH of 6, ye<sup>t</sup> with an increased percentage of CaF2 (23%) [51]. Groups containing MPC had the highest F ion release over time. The levels of release in nCaF2+MPC and nCaF2+DMAHDM+MPC were (0.40 ± 0.02) mmol/L and (0.25 ± 0.03) mmol/L, respectively. This could be due to the hydrophilic nature of MPC, resulting in a higher level of water uptake. As more water was absorbed by the composite, the Ca and F ions were solubilized, leading to an increase in the initial ion release. However, nCaF2+DMAHDM F ion release was (0.20 ± 0.03) mmol/L, which was superior to the control. While the amount of nCaF2 was the same in the two groups, the lower level of initial release in nCaF2+DMAHDM, when compared to MPC containing groups, could result in a more sustained level of F ion release over a longer period. The releasing trend of F and Ca ions were the same for all groups. The groups containing nCaF2 had a high F release initially, and subsequently were followed by a steady state release. The higher release of Ca and F ions in all nCaF2 composites revealed that CaF2 nanoparticles were fast released from the nCaF2 composites because of the hydrolytic breakdown of the interface between the resin matrix and the CaF2 nanoparticles. A likely explanation for such a trend is that the ions were released from the near-surface Ca and F reservoir of the nCaF2 composite, and the reservoir near the surface diminished with the increase in immersion time. However, in clinical situations, the superficial layer of nCaF2 would experience repeated chewing forces and tooth-brushing, hence the CaF2-depleted surface would be removed by wear. As a result, a fresh surface would be exposed with the ability to further continue the calcium and fluoride release. Further studies are needed to investigate the outcome of wear on Ca and F ion release, as well as whether these materials can be made rechargeable to extend the lifetime of ion release.

Recurrent caries is considered a major drawback of composite restorations [1,52,53]. Studies proposed that the greater vulnerability to recurrent caries may be associated with the absence of composites' antibacterial capability when compared to other commonly used restorative materials such as glass ionomers and amalgams [2,54]. Therefore, the improvement of composites by incorporating antibacterial, protein repellent, and remineralizing properties is essential to lengthen the service life of composite restorations. DMAHDM has demonstrated strong antibacterial e ffects on an extensive antimicrobial spectrum through its contact-killing properties [26,55]. Since DMAHDM contains a reactive methacrylate group, it is copolymerized within the resin matrix, rendering it immobilized and preventing its release or loss over time [26,56]. Hence, its antibacterial capability is long-lasting. However, the antibacterial ability of DMAHDM is limited to biofilms that contact the composite surface [26], as biofilms that are not in contact with DMAHDM cannot be killed via DMAHDM's mechanism of action. Even with this limitation, the composite containing nCaF2+DMAHDM substantially lowered the biofilm development and production of lactic acid, reducing the CFU of biofilm by 4 logs.

It can be seen in Figure 6B that the experimental control group and the group containing nCaF2 and MPC exhibited a greater lactic acid production when compared with the other groups. DMAHDM has been shown in previous work to have a strong antibacterial e ffect, as illustrated by the CFU data in Figure 5. This antibacterial activity is reflected indirectly by the lactic acid production. When an *S mutans* biofilm accumulates, the secretion of lactic acid occurs. In the absence of a viable biofilm, the concentration of lactic acid is expected to be very low, as is the case in the DMAHDM-containing groups. Interestingly, the nCaF2 and nCaF2+MPC groups did not have a significant reduction in CFU (Figure 5), but the nCaF2 group showed a significant reduction in lactic acid production in contrast to the nCaF2+MPC group. Previously, it has been shown that the inclusion of CaF2 in a composite formulation has a moderating e ffect on the lactic acid production [57]. It has been speculated that the F ion release helped reduce the acid production of the bacteria via the inhibition of metabolic pathways such as the fermentation pathway for lactic acid production, biofilm plaque, and hydrodynamic e ffects on mass transfer, fluoride delivery, and caries [58]. However, when MPC was added to the composite containing nCaF2, there was no decrease in lactic acid production. This response may be related to the interactions between the calcium and fluoride ions and the phosphorylcholine fragment of MPC. The protein-repellent nature of MPC from the hydrogen bond network of water molecules surrounding the phosphorylcholine fragment limited the ability of the protein to adsorb on the surface. It was found, however, that the presence of halide ions (Cl<sup>−</sup>, Br<sup>−</sup>, I−) influenced the hydrogen bond network and the di ffusion through it [59]. It is possible that the presence of F− ions would have a similar result, which may a ffect F ion di ffusion and disrupt the creation of a protein-repellant coating, leading to biofilm formation.

SEM images in Figure 7 confirm the presence of substantial biofilm on all groups except those containing DMAHDM. At 24 h and 48 h, Heliomolar (CC), experimental control (EC), nCaF2, and nCaF2+MPC groups exhibited the formation of microcolonies and early biofilm multilayers. The biofilms on these groups at 96 h were fully mature, substantial, and dense. However, in DMAHDM-containing groups there was minimal attachment of biofilm observed. At 96 h the DMAHDM-containing groups showed a reduction in biofilm attachment compared to earlier time points. It is possible that the antimicrobial e ffect of DMAHDM as a contact-killing agen<sup>t</sup> had eliminated the bacteria left on the samples over time. These results correlate very well with the CFU results.

It is possible that the salivary proteins adsorption on composite resin surfaces might reduce the effectiveness of this "contact-killing" mechanism. Therefore, a protein repellent agen<sup>t</sup> was added. MPC is a methacrylate that contains phospholipid polar groups. MPC is highly hydrophilic and has been shown to decrease the adsorption of protein and reduce bacterial attachments [60]. Previous work indicated that integrating MPC into the resin decreased the adsorption of proteins by approximately one order of magnitude [43,61]. Additionally, it was shown that surfaces containing MPC were resistant to brushing mechanical stresses [62]. In this study, MPC was incorporated into the resin, and copolymerized into the composite resin. This is analogous to the process of incorporating DMAHDM, with a similar result of a stable, covalently bonded, and non-releasing functionality.

In the current study, the antibacterial capability of the experimental composite was improved with nCaF2 and was further substantially improved with the use of both nCaF2 and DMAHDM. However, the nCaF2+MPC group showed no reduction in the CFU counts. The composite that had nCaF2+DMAHDM+MPC displayed the most potent antibacterial effect with CFU 6 log biofilm reduction. Therefore, these results confirmed that the addition of DMAHDM is vital to improving the antibacterial effect. It is thought that DMAHDM serves to reduce the accumulation of biofilm and, as a result, improves the efficacy of MPC in resisting protein adsorption. However, in the absence of DMAHDM, the bacterial biofilm will accumulate and render MPC less effective.

The combination of nCaF2+DMAHDM promoted remineralization and had a significant antibacterial effect. Another potential benefit of using nCaF2 is that the release of F ions could act as a transmembrane proton carrier and an inhibitor to the glycolytic enzyme, thus preventing oral microorganisms by stimulating acidification of the cytoplasm and providing antibacterial effect at a long-distance to inhibit caries [10,63]. These properties would be highly valuable to inhibit recurrent caries around the margins of the restoration, since this is the location that most dental plaque tends to accumulate. In addition to its superior antibacterial and remineralization properties, the nCaF2+DMAHDM composite exhibits excellent mechanical properties which make it suitable in a variety of restoration applications. In comparison, the composite consisting of nCaF2+DMAHDM+MPC exhibited lower mechanical properties. However, the strength and modulus achieved by this group was still high enough to be used for restorations in low load areas such as class V restorations [37]. This is a critical area of utilization, since older patients tend to have exposed root surfaces that are more prone to caries. Further studies are needed to investigate the nCaF2+DMAHDM combination in dental cements, bonding agents, fissure sealants, and composites to remineralize tooth lesions and suppress biofilm and plaque buildup, especially in patients with a high caries risk.
