**1. Introduction**

Dental resin composites are an excellent material for direct restorations of anterior teeth and in many cases posterior teeth due to their esthetics and ease of placement [1]. Nevertheless, composites are known to accumulate more oral bacterial plaque and biofilm than other direct restorative materials, which could expose the restored tooth to a higher risk for future recurrent caries [2]. Indeed, most failed restorations due to secondary caries are restored with composites [3,4]. The formation of plaque starts with the salivary-acquired pellicle formation. The glycoprotein found in the acquired pellicle promotes bacterial cell adherence. The microbes in the biofilm then produce acids which lowers the pH and lead to mineral loss over time resulting in dissolution of the tooth structure, the formation of caries, and failure of the restoration [5–7]. Unfortunately, currently available commercial composites lack antibacterial properties. Accordingly, efforts were made to overcome the presence of cariogenic bacteria, in an effort to prevent recurrent caries [8].

The incorporation of calcium fluoride nanoparticles (nCaF2) into composites has the potential to reduce demineralization [9]. Fluoride (F) ions work by stimulating the remineralization and suppressing the oral microorganisms [10,11]. The presence of F ions in the event of demineralization enhances the precipitation of calcium and phosphate ions and forms fluorapatite [Ca5(PO4)3F] to protect the tooth surface [12–14]. Fluoride was also shown to have the advantage of reducing bacterial acid production to reduce recurrent caries.

Designing a composite containing calcium fluoride nanoparticles would enhance the fluorapatite deposition in the affected tooth structure. When the tooth structure is subjected to acidic attack by the cariogenic pathogens, calcium and phosphate ions are lost from enamel. Using remineralization approaches to restore the lost minerals is required to enforce and strengthen the tooth structure. Therefore, the composite with calcium fluoride nanoparticles would enhance remineralization and form fluorapatite that is able to resist future acidic challenges. Several studies have demonstrated the ability of forming fluorapatite using nanotechnology [15,16]. In one study, they manufactured fluorapatite nanoparticles and examined its doping with silver ion nanoparticles and evaluated its physical and antimicrobial effects. The results showed 30% inhibition of bacterial growth after 4 h of incubation while maintaining the natural morphology of fluorapatite [15]. In another study, fluorapatite was incorporated into chitosan scaffolds. Fluorapatite maintained its structure, granted antimicrobial effects, and showed osteoconductive capability [16].

Furthermore, the incorporation of antibacterial agents into composites have also been investigated. Imazato et al. integrated 12-methacryloyloxydodecylpyridinium bromide (MDPB) into composites and showed successful antibacterial effects [17–20]. The incorporation of quaternary ammonium polyethylenimine (QPEI) into composites also produced a potent and wide-spectrum antimicrobial effect against salivary microorganisms [21]. Antimicrobial peptides (AMPs) were also demonstrated to have antimicrobial properties by bacterial membrane permeabilization and intracellular targeting [22]. Other studies developed antibacterial agents such as dimethylaminohexadecyl methacrylate (DMAHDM) [23,24] and showed a strong antibiofilm activity without compromising the mechanical properties [25].

Previous studies indicated that the salivary protein accumulation on composite surface could lower the efficiency of "contact-killing" mechanisms [26,27]. Accordingly, efforts were made to improve protein-repellent strategies including the addition of protein-repellent agents such as (2-methacryloyloxyethyl phosphorylcholine, or MPC) into resins [28–30]. This method provided resistance to protein adsorption and bacterial adhesion due to the hydrophilic characteristic of MPC [29,30]. However, to date, there has been no report on the development of a novel bioactive dental composite that contains nCaF2, DMAHDM, and MPC in combination.

The objectives of this study were to develop a new composite consisting nCaF2, DMAHDM, and MPC, and to investigate the mechanical, ion release and oral biofilm properties for the first time. The following hypotheses were tested: (1) Adding DMAHDM and MPC into the nCaF2 composite would have mechanical properties similar to a commercial control composite; (2) Adding DMAHDM and MPC into the nCaF2 composite would not compromise the F and Ca ion release; and (3) The new bioactive composite would have much less microorganisms, produce less biofilm acid, and have better remineralizing properties than the commercial control composite.

#### **2. Materials and Methods**

#### *2.1. Fabrication of Composites*

The experimental resin consisted of bisphenol A glycidyl dimethacrylate (BisGMA, Esstech, Essington, PA, USA), and triethylene glycol dimethacrylate (TEGDMA, Esstech) at 50:50 mass ratio. Camphorquinone at 0.2% (Millipore Sigma, Burlington, MA, USA) and 0.8% ethyl 4-N, N-diethylaminobenzoate (Millipore Sigma) were incorporated for photoactivation. The resin is referred to as BT resin. MPC (Millipore Sigma) was added at a mass fraction of 3% and incorporated into the BT resin with magnetic stirring bar at 150 rpm to be dissolved completely into the resin.

The synthesis of DMAHDM was performed using a modified Menschutkin reaction [31]. Briefly, 10 mmol of 2-(dimethylamino) ethyl methacrylate (Millipore Sigma), 10 mmol of 1-bromohexadecane (TCI America, Portland, OR, USA), and 3 g of ethanol were combined in a reaction vessel and then stirred for 24 h at 70 ◦C. After the evaporation of the solvent and removal of impurities, the DMADHM was collected. DMAHDM was added into the BT resin at a mass fraction of 3% and was stirred using a magnetic stirring bar at 150 rpm until it was completely dissolved into the resin.

The nCaF2 was manufactured using a spray-dry method as described in previous studies, yielding a mean particle size of 32 nm [9,32–34]. The mass fraction of nCaF2 incorporated into BT resin was 15%, based on our preliminary study. A previous study tested di fferent concentrations of nCaF2 in composite and, after long-term water-aging, the composite with 20% nCaF2 had a flexural strength of 60 MPa [33]. In the present study, 15% nCaF2 was integrated into the resin to achieve good mechanical strength. Silanized barium boroaluminosilicate glass particles with a mean size of 1.4 μm (Dentsply Sirona, Milford, DE, USA) were incorporated into the BT resin for mechanical enhancement. As a commercial control composite, Heliomolar (Ivoclar Vivadent, Mississauga, ON, Canada) was also tested. Heliomolar contains 66.7% filler mass fraction of ytterbium-trifluoride and nanofillers of 40–200 nm of silica. The following groups were tested (Table 1 summarizes the materials used in the study):



**Table 1.** Materials used in the study.

#### *2.2. Characterization of nCaF2*

Transmission electron microscopy (TEM, Tecnai T12, FEI, Hillsboro, OR, USA) was used to assess the nanoparticles. Samples were prepared through placing nanoparticles on a perforated copper grid coated by a carbon film. To avoid particle agglomeration, the sample was ultrasonicated for 5 min in acetone prior to deposition. Particle size distribution was measured using a laser di ffraction particle size analyzer (SALD-2300, Shimadzu North America, Columbia, MD, USA).

#### *2.3. Mechanical Properties Testing*

Each composite paste was mixed in a disposable plastic container using a speed mixer (DAC 150.1 FVZ-K SpeedMixer ™, FlackTec Inc., Landrum, SC, USA) at a speed of 2800 rpm for 1 min, and then thoroughly mixed by hand on a plastic slab for 5 min. The paste was then placed in a rectangular mold of 2 × 2 × 25 mm3. Mylar strips were placed on both sides, followed by two glass slides. The specimen was light-cured using a curing unit at 1200 mW/cm<sup>2</sup> (Labolight DUO, GC America, Alsip, IL, USA) on each side for 1 min [35]. After demolding, the samples were stored in a 100% humidity chamber for 24 h at 37 ◦C. Flexural strength and elastic modulus were tested at a crosshead-speed of 1 mm/min with a 10 mm span with a three-point flexural test using a computer-controlled universal testing system (Insight 1, MTS, Eden Prairie, MN, USA) [36,37]. Flexural strength and elastic modulus were measured after 24 h of specimen immersion in distilled water at 37 ◦C. Flexural strength: S = 3Pmax/L(2bh2), where Pmax is the fracture load, L is span, b is sample width and h is thickness. Elastic modulus: E = (P/d) (L<sup>3</sup>/[4bh3]), where load P was divided by displacement d which is the slope in the linear elastic region. Six specimens were tested for each group (*n* = 6).

#### *2.4. Ca and F Ion Release*

The ion releases for all groups containing nCaF2 were tested. A solution of sodium chloride (NaCl) (133 mmol/L) was bu ffered with 50 mmol/L HEPES to pH 7 [36,38]. Three specimens of 2 × 2 × 12 mm<sup>3</sup> were placed into 50 mL of solution, accommodating a specimen volume/solution ratio of 3.0 mm<sup>3</sup>/mL, similar to those in previous studies [36,38,39]. The specimen's F and Ca ions release were measured at 1, 2, 4, 7, 14, 21, 28, 35, 42, 49, 56, 63, and 70 days. At every time point, aliquots of 2 mL were collected and substituted by a fresh 2 mL solution of NaCl. The aliquots were investigated for Ca ions by a colorimetric assay using a microplate reader (SpectraMax M5, Molecular Probes, San Jose, CA, USA) as previously described, using known standard and calibration curves [36,38,39]. The F ion release was tested with a F ion selective electrode (Orion, Cambridge, MA, USA). Fluoride standard solutions were measured to form a standard curve. The standard curve was used to establish the F concentration. The F ion concentration measurement was performed by combining 0.5 mL of sample and 0.5 mL of undiluted TISAB solution (Fisher Scientific, Pittsburgh, PA, USA).

#### *2.5. Sample Preparation for Biofilm Tests*

The cover of a 96 well plate was used to fabricate composite discs for microbiological experiments yielding samples 0.5 mm in thickness and 8 mm in diameter [31]. Composite paste was placed at each indent in the in 96-well plate cover then covered with Mylar strips and glass slides to form a smooth surface. It was then light cured as described previously and then stored for 24 h at 37 ◦C. The following day discs were magnetically stirred for 1 h at 100 rpm in distilled water to remove uncured monomers [18,40,41]. The specimens were sterilized using ethylene oxide (Anprolene AN 74i, Andersen Products, Haw River, NC, USA) for 24 and allowed to de-gas for 7 days, following the instructions of the manufacturer.

#### *2.6. Saliva Collection and Dental Plaque Microcosm Biofilm Model*

Saliva collection was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Institutional Review Board at the University of Maryland Baltimore (IRB #: HP-00050407). The advantage of the dental plaque microcosm biofilm model is the use of an inoculum of human saliva to mimic the heterogeneity and complexity of the bacteria that are present in human dental plaque [18]. An equal amount of saliva was simultaneously gathered from ten healthy contributors with normal dentition, free of active caries, and no antibiotic use within the prior 3 months. Contributors were instructed not to brush their teeth 24 h preceding collection and not to eat or drink 2 h preceding the collection. Subsequently, the collected saliva from all participants was mixed and diluted to 70% in sterile glycerol. Then the saliva–glycerol solution was stored at −80 ◦C until use [42].

For all biofilm experiments, McBain artificial saliva growth medium was used. McBain medium contained 2.5 g/<sup>L</sup> Type II mucin (porcine, gastric, Millipore Sigma), 2.0 g/<sup>L</sup> bacteriological peptone (Becton Dickinson, Sparks, MD, USA), 2.0 g/<sup>L</sup> tryptone (Becton Dickinson), 0.35 g/<sup>L</sup> NaCl, 1.0 g/<sup>L</sup> yeas<sup>t</sup> extract (Fisher Scientific), 0.2 g/<sup>L</sup> potassium chloride (Millipore Sigma), 0.1 g/<sup>L</sup> cysteine hydrochloride (Millipore Sigma), 0.2 g/<sup>L</sup> calcium chloride (Millipore Sigma). The pH of the medium was adjusted to 7 and autoclaved. After cooling the medium, 0.0002 g/<sup>L</sup> vitamin K1, 0.001 g/<sup>L</sup> hemin were added. During biofilm experiments, 2% sucrose solution and the saliva–glycerol solution were used as an inoculum at a ratio of 1:50. The sucrose and inoculum were added to the medium and 1.5 mL of the medium was placed in each well of a 24-well plate containing a composite specimen from each groups. Specimen were incubated in 5% CO2 at 37 ◦C for 8 h to permit biofilm growth on the samples. The same procedure was repeated after 8 h without the addition of saliva and incubation occurred again for 16 h. After 16 h the samples were moved to a new 24-well plate which contained fresh medium and sucrose, and was further incubated for 24 h. Composites were exposed to bacterial culture for a total of 48 h, which resulted in reasonably mature dental plaque microcosm biofilms on composites [25,43].

#### *2.7. Biofilm Colony Forming Units (CFU) Counts*

Nine discs were prepared for each group. Following the 48 h incubation, the disc samples containing biofilm were transported into a vial filled with 1 mL of cysteine peptone water (CPW). This was vortexed for 5 s then sonicated for 5 min and vortexed again to harvest the biofilm [29]. Serial dilutions of the suspensions of bacteria were prepared and transported to agar plates to grow. The CFU were counted on three di fferent agar plates. To determine total streptococci count, mitis salivarius agar (MSA, Becton Dickinson, Sparks, MD, USA) were used. To determine the growth of mutans streptococci, 0.2 units per mL bacitracin (Millipore Sigma) was added to the mitis salivarius agar (MSB). To evaluate the growth of the total microorganisms, tryptic soy blood agar (TSBA) agar plates were used by adding defibrinated sheep blood to tryptic soy agar (TSA, Becton Dickinson). The agar plates were kept at 37 ◦C in a 5% CO2 incubator for 48 h. CFU calculation was based on the colony number and multiplied by the dilution factor [29].

#### *2.8. Biofilms Metabolic Activity Evaluation (MTT)*

The MTT (3-[4,5-dimethylthiazol-2-yl]-2,5- diphenyltetrazolium bromide) assay was performed to investigate the biofilm metabolic activity. Following 48 h of incubation, the discs (*n* = 9) were transferred into a clean 24-well plate, then 1 mL of tetrazolium dye was placed to every disc. The discs were then incubated in an incubator of 5% CO2 at 37 ◦C. Discs were then transported to another 24-well plate, and 1 mL of dimethyl sulfoxide (DMSO) was added to every disc and incubated for 20 min in a dark room [29,44]. After incubation, 200 μL of the DMSO solution was collected and the absorbance at 540 nm was measured [29,44] using a microplate reader (SpectraMax® M5, Molecular Devices, San Jose, CA, USA).

#### *2.9. Biofilms Lactic Acid Production*

Following 48 h incubation, discs (*n* = 9) were moved to a different 24-well plate comprising 1.5 mL buffered-peptone water (BPW) with 0.2% sucrose then incubated for 3 h in a 5% CO2 incubator at 37 ◦C to release acids. After 3 h, the BPW solution lactic acid concentrations were measured by recording the absorbance at 340 nm [29,44] using a microplate reader (SpectraMax® M5, Molecular Devices). Standard curves were produced by means of lactic acid standards.

#### *2.10. Scanning Electron Microscopy (SEM) of Biofilms*

For biofilm visualization and confirmation of bacterial attachments on composite discs, biofilms formed at 24 h, 48 h, and 96 h were sputter-coated with platinum. Scanning electron microscopy (SEM, Quanta 200, FEI Company, Hillsboro, OR, USA) was used to examine the bacterial accumulation (Figure 7).

#### **3. Statistical Analysis**

All data were evaluated with one-way analysis of variance (ANOVA), and post hoc multiple comparison using Tukey's honestly significant difference test was performed. All statistical analysis was completed using the GraphPad Prism 8 software package (GraphPad Software, San Diego, CA, USA) at 0.05 level of significance.
