Novel dental resin infiltrant containing smart monomer dodecylmethylaminoethyl methacrylate

Objectives White spot lesions (WSLs) are prevalent and often lead to aesthetic problems and progressive caries. The objectives of this study were to: (1) develop a novel resin infiltrant containing smart monomer dodecylmethylaminoethyl methacrylate (DMAEM) to inhibit WSLs, and (2) investigate the effects of DMAEM incorporation on cytotoxicity, mechanical properties, biofilm-inhibition and protection of enamel hardness for the first time. Methods DMAEM was synthesized using 1-bromododecane, 2-methylamino ethanol and methylmethacrylate. DMAEM with mass fractions of 0%, 1.25%, 2.5% and 5% were incorporated into a resin infiltant containing BisGMA and TEGDMA. Cytotoxicity, mechanical properties and antibacterial effects were tested. After resin infiltration, bovine enamel was demineralized with saliva biofilm acids, and enamel hardness was measured. Result DMAEM infiltration did not increase the cytotoxicity or compromise the physical properties when DMAEM mass fraction was below 5% (p > 0.05). Biofilm metabolic activity was reduced by 90%, and biofilm lactic acid production was reduced by 92%, via DMAEM (p < 0.05). Mutans streptococci biofilm CFU was reduced by 3 logs (p < 0.05). When demineralized in acid and then under biofilms, the infiltrant + 5% DMAEM group produced an enamel hardness (mean ± sd; n = 6) of 2.90 ± 0.06 GPa, much higher than 0.85 ± 0.12 GPa of the infiltrant + 0% DMAEM group (p < 0.05). Significance A novel resin infiltrant with excellent mechanical properties, biocompability, strong antibacterial activity and anti-demineralization effect was developed using DMAEM for the first time. The DMAEM resin infiltrant is promising for inhibiting WSLs, arresting early caries, and protecting enamel hardness.


Introduction
Dental caries is one of the most prevalent oral diseases globally (Peres et al., 2019). White spot lesions (WSLs) are early-stage caries, which have been reported to affect between 50% and 96% of patients receiving fixed orthodontic treatments (Sardana et al., 2019;Sonesson et al., 2020). Treatments for patients suffering from WSLs include fluoride, phosphopeptide compounds, and resin infiltration (Wegehaupt and Attin, 2010;Leila et al., 2017). Resin infiltant can penetrate into the pores of enamel and seal the passage of acid. It is minimally invasive and has been an emerging therapeutic modality in the treatment of WSLs. However, resin infiltant can only seal off about 30-60% of the enamel pore volume (Kielbassa et al., 2009;Yim et al., 2014). Furthermore, after the attack of acid produced by cariogenic microorganisms, the hardness of enamel decreased significantly (Dai et al., 2022).
In the oral cavity, dental caries starts with the breakdown of the dynamic balance of oral microecology (Pitts et al., 2017). During the development of caries, oral microbial diversity is decreased, and the acidogenic bacteria can multiply and produce acids, causing the pH to decrease and leading to hard tissue demineralization (Selwitz et al., 2007). Antibacterial agents such as silver nanoparticles (AgNP) (Kielbassa et al., 2020;Li et al., 2020), quaternary ammonium methacrylate (Yu et al., 2020), and ionic liquid-loaded microcapsules (Cuppini et al., 2021) were incorporating into dental materials to inhibit the growth of plaque. However, oral commensal microbiome colonizes on dental tissue plays an essential role in oral microecology (Rosier et al., 2018). The oral probiotics were also inhibited by the agents mentioned above, that would break the balance of oral microecology (Pitts et al., 2017;Liang et al., 2020). Therefore, an intelligent resin infiltrant that could show an antibacterial effect only during dysbiosis, rather than killing all the bacteria, is highly preferred to inhibit WSLs.
Recently, microecosystem-regulating effects of intelligent pH-sensitive resin materials have received more attention (Fenton et al., 2018). Dodecylmethylaminoethyl methacrylate (DMAEM) is a novel tertiary amine (TA) smart material that responds to pH change. Its central nitrogen atom is connected to 3 alkyl or aromatic groups. When the local pH is low, such as during biofilm acid attacks, the nitrogen atoms of TA is protonated to form quaternary ammonium salts (QAMs), which are antibacterial. The QAMs-modified dental materials demonstrated excellent antibacterial effect (Han et al., 2017;Wang et al., 2019;Zhou et al., 2019). In previous studies, the MIC and MBC of DMAEM against Streptococcus mutans UA159 (S. mutans), Streptococcus gordonii DL1 (S. gordonii), and Streptococcus sanguinis SK1 (S. sanguinis) under different pH were tested by serial microdilution assays . When the pH was 5, the MIC of DMAEM against S. mutans, S. sanguinis and S. gordonii were 0.18 mg/mL, 0.37 mg/mL, and 5.95 mg/mL, respectively. The MBC against S. mutans, S. sanguinis and S. gordonii were 1.4 mg/mL, 2.9 mg/mL, and 11.9 mg/mL, respectively. These values indicate a strong antibacterial activity at pH 5.
However, when the pH was 7.4, the MIC and MBC of DMAEM became much higher, at more than 13.5 mg/mL, which meant much lower antibacterial activity . Therefore, DMAEM was strongly antibacterial only when pH was low and when tooth-protection was most needed. These results indicate that DMAEM was smart and had a pH-sensitive capability . In addition, the smart DMAEMmodified resin adhesives could successfully combat secondary caries in vivo and in vitro . Furthermore, smart DMAEM sealants showed the potential to reduce microleakage, thus preventing dental caries (Li et al., 2021).
To date, there has been no report on pH-sensitive modification of resin infiltrant to inhibit WSLs. Considering the unique pH-responsive feature of DMAEM, which was suitable for the unique acidic environment of WSLs, we designed an intelligent pH-sensitive resin infiltrant containing DMAEM for the first time. The objectives of the present study were to: (1) develop a novel intelligent resin infiltrant containing smart monomer dodecylmethylaminoethyl methacrylate (DMAEM) to inhibit WSLs, and (2) evaluate the cytotoxicity, mechanical properties, biofilm-inhibition and protection of enamel hardness of the novel resin infiltrant. The following hypotheses were tested: (1) Novel resin infiltrant containing DMAEM could be successfully synthesized; (2) DMAEM resin infiltrant could inhibit biofilm growth and acid production; and (3) DMAEM resin infiltration could protect enamel and retain its hardness after biofilm acid attacks.
2 Methods and materials 2.1 Synthesis of DMAEM and preparation of specimen According to the work described previously, DMAEM was synthesized and verified Li et al., 2021). In brief, 100 mmol of 1-bromododecane was added to 500 mmol of 2-methylamino ethanol in 80 mL isopropanol at room temperature. After stirring for 8-10 h under reflux at 85°C, the mixture was cooled to room temperature and slowly poured into 150 mL of diethyl ether. Then the mixture was washed with deionized water and brine, dried over anhydrous Na 2 SO 4 and vacuumized. Then 31.8 mL methylmethacrylate, catalyst CAA (107 mg, 0.4 mol%) and polymerization inhibitor methoxyphenol (100 mg, 2 mol%) were added at room temperature. After stirring at 100-110°C for 12 h, CAA (150 g, 0.6 mol%) was supplementary, and then stirred for another 12 h, cooled to room temperature and evaporated.

Cell viability test
A 96-well plate (Costar; Corning Inc., NY, USA) was used as a mold for the samples (Yu et al., 2020). 5 mL of resin infiltrant were added to each well, and then light-cured for 40 s (Triad 2000; Dentsply, York, PA, USA). To remove the uncured monomers, all the samples were immersed in deionized water for 24 h. The samples were sterilized by ethylene oxide in an ethylene oxide sterilizer (Anprolene AN 74i; Andersen, Haw River, NC, USA). Samples of the same group (n = 6) were immersed in 200 mL of cell medium and soaked at 37°C for 24 h to obtain the extracts, which were collected and diluted to 10 mL, and the extract was diluted to 36 times, 64 times, 128 times with fresh medium . The NOKSI immortalized normal oral keratinocyte cell line (provided by Dr. Tao Ma, University of Maryland) were cultured and maintained in keratinocyte serum-free medium with growth factor supplement (Gibco; Thermo, Carlsbad, USA) and 1% antibiotic/antimycotic (AA) (Millipore Sigma; Merck KGaA, Darmstadt, Germany) at a density of 40,000 cells per mL. Cells were cultured at 37°C and 5% CO 2 (Wisniewski et al., 2021). The negative control group was inoculated into the medium without extract. All groups were replaced with fresh medium after 24 h. After 48 h of incubation, 10 mL of CCK-8 solution (Cell Counting Kit-8; Dojindo, Kumamoto, Japan) was added and incubated at a constant temperature incubator for one hour. Optical density (OD) was measured at a wavelength of 450 nm using a microplate reader (SpectraMax M5; Molecular Devices, Sunnyvale, CA, USA). Six replicates were tested for each group.

Mechanical properties test
Rectangular molds (2 × 2 × 25 mm) were used for mechanical testing . After composite bars were immersed in distilled water at 37°C for 1 d, a computercontrolled Universal Testing Machine (5500R; MTS, Cary, NC, USA) was used to test the mechanical properties. In brief, the specimens were fractured in three-point flexure with a 10-mm span at a crosshead-speed of 1 mm/min. Flexural strength (S) was calculated as: S = 3PmaxL/(2bh2), where Pmax is the fracture load, L is span, b is specimen width, and h means specimen thickness. And, elastic modulus (E) was calculated as: E = (P/d) (L3/[4bh3]), where load P divided by displacement d is the slope of the load-displacement curve in the linear elastic region. Six specimens were tested for each material (n = 6).

Bacterial inoculation
Saliva derived biofilm was used in the study. Ten healthy individuals without active caries with natural dentition served as donors. Donors did not take any antibiotics in the three months previous to donation and fasted for two hours. The saliva was diluted two-fold with sterile 50% glycerol. Then the saliva was stored at -80°C .
Each well of 96-well plate with resin infiltrant was added with 300mL Mcbain medium (Yu et al., 2020). The salivaglycerol stock was seeded (1:50 final dilution) into plates and incubated under aerobic environment at 5% CO 2 with 37°C. The medium was refreshed every 12 h. After 24 h, phosphatebuffered saline (PBS) was used to rinse the biofilms to remove loose bacteria.

Anti-bacterial test 2.5.1 Biofilm accumulation
A crystal violet assay was performed to analyze biomass accumulation (Huang et al., 2019). Briefly, each group included six duplicate samples. After the biofilm was rinsed with PBS, 200 mL 95% methyl alcohol was added to each well and incubated for 15 min to fix the biofilm. After the biofilm was rinsed with PBS, submerging in 300 mL 0.1% crystal violet solution for 30 min. Then, washing with PBS. Finally, 300 mL 95% ethanol solution was added to each well and the plate was shaken at 80 rpm for 45 min at room temperature. Subsequently, 100 µL ethanol of the solution from each well was transferred to a 96-well plate, and a microplate reader was used to measure the absorbance of the solution at a wavelength of 595 nm.

MTT test
The biomass accumulation reflects the whole biofilm which contains live cells and dead cells. It is important to evaluate the metabolic activity that was produced by live cells only. In the present study, the MTT (3-(4,5-Dimethyl-thiazol-2-yl)-2,5diphenyltetrazolium bromide) assay was used to measure metabolic activity . Each group involved six duplicate samples. After proliferation for 24 h, the biofilms were rinsed with PBS and 200 mL MTT dye was added to each well (0.5 mg/mL MTT in PBS). Then, the biofilm plates were cultured for 1 h at 37°C. After removing the MTT solution, 300 mL dimethyl sulfoxide (DMSO) was added and shaken at 80 rpm for 20 min in the dark to dissolve the formazan crystals. Finally, 100 µL of DMSO was placed into a 96-well plate, and the absorbance was read at a wavelength of 540 nm using a microplate reader.

Lactic acid test
First, biofilms were rinsed by cysteine peptone water (CPW) to remove loose bacteria . And 300 mL of buffered peptone water (BPW) supplemented with 0.2% sucrose was added, and incubated at 5% CO 2 with 37°C for 3 h to produce acid. After 3 h, the BPW solution was used for lactate analysis. An enzymatic (lactate dehydrogenase) method was used to evaluate the lactate concentrations. Microplate reader was used to measure the absorbance at 340 nm for the collected BPW solutions. Standard curves were prepared using a lactic acid standard (Supelco Analytical; Bellefonte, PA, USA).

Live/dead staining
BacLight Live/Dead Bacterial Viability Kit (Molecular Probes; Eugene, OR, USA) was used in the study . Live bacteria cells were stained with SYTO 9 to produce green fluorescence, and cells with compromised membranes were stained red by propidium iodide. The disks were examined using inverted epifluorescence microscope (TE2000-S; Nikon, Melville, NY, USA), and percent of live bacteria (%) was calculated by ImageJ software (ImageJ_v1.8.0; National Institutes of Health, USA). Percent of live bacteria (%) = live bacteria/(live bacteria + dead bacteria). Six replicates were tested for each group.

CFU
After 24 h incubation, biofilms were harvested for colonyforming unit (CFU) analysis (Cheng et al., 2016). Three types of agar plates were prepared. First, tryptic soy blood agar culture plates were used to determine the total number of microorganisms. Second, mitis salivarius agar (MSA) culture plates, containing 15% sucrose, were used to determine the total number of streptococci. Third, MSA agar culture plates plus 0.2 units of bacitracin per mL were used to determine the number of mutant streptococci .

Enamel hardness test
For anti-demineralization test, bovine teeth were used as previously described (Yu et al., 2020). In brief, bovine permanent incisors free of lesions and cracks were selected. Crowns were cut into sections measuring 4 × 4 × 2 mm by using a diamondcoated band saw with continuous water cooling (Isomet; Buehler, Lake Bluff, IL, USA). These surfaces were then ground flat with water-cooled carborundum discs made of waterproof silicon carbide paper (Extec; Enfield, CT, USA) with various grits (1000, 1200, 2400, 3000, and 4000) (Cheng et al., 2008;Iizuka et al., 2014;He et al., 2015). To remove the residual abrasives, all of the polished samples were individually sonicated in distilled water for five minutes.
Initial enamel caries was produced in enamel blocks, as described earlier (Rocha Gomes Torres et al., 2011). The demineralization solution contained 75mM acetic acid (pH 4), 8.7 mM Ca(Cl) 2 , 8.7 mM KH 2 PO 4 , and 0.05 ppm NaF (acetic acid, Ca(Cl) 2 , KH 2 PO 4 , NaF; Biofroxx, Hessen, Germany) (Gao et al., 2020). The blocks were immersed in the demineralization solution (8 mL per block) at 37°C for 16 h with stirring. After dried, 37% phosphoric acid was coated on the demineralization site of the samples for two minutes. Then, rinsed with water for 30 s, and the samples were dried with an air syringe without oil or water. In the demineralization area, 1 mL 99% ethanol was used to dehydrate the area thoroughly. After 30 s, the areas were dried with a water-free and oil-free air syringe. Resin-infiltrant (1 mL) was applied to the demineralized zone for three minutes. After removing the excess material, the area was light-cured for 40 s. Afterward, infiltrant was applied again. After one minute, the excess material was removed again and light-cured for 60 s (Figure 1). All samples were sterilized in an ethylene oxide sterilizer.
Enamel hardness was measured before the demineralized treatment, after demineralization in acid solution, after resin infiltration and after demineralization under biofilms. A hardness tester (Shimadzu HMV-2000, Kyoto, Japan) was employed using a Vickers diamond indenter with a load of 25 g for 5 s dwell time (Liang et al., 2019;Gao et al., 2020). Six indentations were made in each enamel specimens, and each group had six specimens.

Statistical analysis
Statistical analyses were performed with SPSS, version 22.0 (SPSS; Chicago, IL, USA). One-way analysis of variance (ANOVA) was performed to detect the significant effects of the variables. Tukey's multiple comparison test was used to compare the means of each group at a p-value of 0.05.

Results
Cell viability results of resin infiltrant are plotted in Figure 2. Results showed that all the groups except for infiltrant + 7.5% DMAEM group (p < 0.05) had acceptable cell viability (p > 0.05).
Mechanical properties of resin infiltrant are shown in Figure 3. All DMAEM resin infiltant had flexural strength and elastic modulus comparable to the infiltrant + 0% DMAEM group (control), which are around 130 MPa and 4 GPa respectively (p > 0.05). The DMAEM modification did not affect the mechanical properties of resin infiltrant. Figure 4 showed the quantification of biofilm viability. MTT assay showed that the metabolism of biofilm decreased with the increase of DMAEM concentration (p < 0.05), while the infiltrant + 1.25% DMAEM group didn't show significant difference (p > 0.05). Crystal violet assay and lactic acid production assay plotted similar results, with the increase of DMAEM concentration, the biofilm accumulation and the lactic acid production decreased (p < 0.05).
Live/dead staining images ( Figure 6) showed that the infiltrant + DMAEM groups had much more red staining, indicating a strong anti-bacterial effect, while the infiltrant + 0% DMAEM group (control) had more green staining and little red staining, which represented little anti-biofilm activity. Figure 6E showed live/dead ratio, the infiltrant + DMAEM groups had lower percent of live bacteria (p < 0.05), and the percent of live bacteria decreased with increasing DMAEM concentration (p < 0.05).
Enamel hardness test is plotted in Figure 7. Results showed that the hardness of enamel decreased after being demineralized in acid solution (p < 0.05), while the resin infiltration recovered the hardness (p > 0.05) ( Figure 7A). After 48 h biofilm demineralization, the hardness of all the groups except for the infiltrant + 5% DMAEM group decreased (p < 0.05). After two stages of demineralization: first in acid solution, second in biofilm culture, the enamel showed the lowest hardness in all the group (p < 0.05). While the hardness value increased as the increasing of the DMAEM concentration (p < 0.05) ( Figure 7B).

Discussion
In the present study, a novel pH-sensitive DMAEM infiltrant with antibacterial and anti-demineralizing properties was developed. The hypotheses were proven that the new resin infiltrant had the acceptable mechanical properties without jeopardizing the biocompability, which had normal oral keratinocyte viability similar to that of the infiltrant + 0% DMAEM group (control). Additionally, the antibacterial effect enhanced the anti-demineralizing properties under the attack of biofilm for 7 days.
New materials used for dental treatment are required to be non-cytotoxic and biocompatible. In the present study, cytotoxicity test was carried out according to ISO 10993-5-2016 standard . According to the ISO standard, in vitro cytotoxicity tests, the MTT value is not less than 80% prove to be slightly cytotoxic . NOKSI were used in the present study because they are clinically relevant and in close proximity to dental restorations. Our findings suggest that the new resin infiltrant did not induce any significant toxicity when the mass fraction of DMAEM was below 5%, which was similar to the previous study . Therefore, the concentration of subsequent experiments was selected. These findings indicate that DMAEM resin infiltrant is suitable for clinical applications. Future in vivo experiments are needed to investigate the biocompatibility of the infiltrant containing DMAEM.
Mechanical properties of resin infiltrant may influence the hardness of the enamel, as well as the longevity of treatment (Paris et al., 2013). To evaluate whether DMAEM would jeopardize the mechanical properties, we made the resin infiltrant into a cuboid with the size of 2 × 2 × 25 mm. The result demonstrated that the modification did not affect the mechanical properties of resin infiltrant (p > 0.05). In detail, flexural strength of the resin infiltrant was around 120-130 MPa, while the elastic modulus was about 4 GPa, which were similar to other results (Prodan et al., 2022). The DMAEM could covalently grafted to resin infiltrant by reaction of acrylate groups with methacrylate groups, yielding the non-release antibacterial dental material. The covalent binding explained the similar mechanical properties of infiltant + DMAEM groups to the control group. The mechanism was similar to that of the DMADDM-modified and DMAHDM-modified dental resins, which had been reported before (Han et al., 2017;Cheng et al., 2017). WSLs are early-stage of dental caries, which are common side effects of orthodontic treatments (Paula et al., 2017). WSLs are caused by cariogenic biofilm, leading to demineralization of the enamel (Zou et al., 2018). In detail, during the dental caries, the oral microbial diversity is decreased, cariogenic microorganisms proliferate, whereas pH value decreases in a certain period of time (Zou et al., 2018). The balance of demineralization and re-mineralization is shifted to mineral loss, finally the caries is formed (Zou et al., 2018). Therefore, biofilm inhibition is the first step to solve WSLs (Abdullah and John, 2016;Horst, 2018).
The resin infiltration concept was first developed in the 1970's (Cheng et al., 2017), low viscosity resin materials had been used to restore decalcified enamel at that time (Paula et al., 2017;Zou et al., 2018), such as sealant and adhesive. At the end of the 2000's, with the development of dental materials, resin infiltrant was applied for caries arrestment, which meet the principle of microinvasive therapies nowadays (Borges et al., 2017). With resin infiltration, demineralization of enamel is hampered, for the reason a diffusion barrier of acids produced by biofilm is created (Paris et al., 2020). A three-year follow-up clinical study revealed that resin infiltration reduced 65-90% of the risk of caries compared with varying self-applied noninvasive interventions alone (Paris et al., 2020). However, it was difficult for infiltration treatment when lesions extending into deeper enamel or even dentine: complete penetration is less reliable (Yim et al., 2014;Min et al., 2015). Furthermore, it has been reported that the application of resin infiltrant can only seal off around 30-60% of lesion depth (Yim et al., 2014;Min et al., 2015). In addition, enamel surfaces are constantly exposed to the oral microflora, many in vitro studies reported that there even was a hardness loss for the resin-infiltrated area after acid attack (Tawakoli et al., 2016). Common methacrylates in restorative materials, such as TEGDMA and BisGMA, do not possess strong antimicrobial activities (Flor-Ribeiro et al., 2019). To give the material antibacterial properties, researchers added antibacterial agents into resin infiltrant. Several studies modified the materials with AgNP (Kielbassa et al., 2020), quaternary ammonium methacrylate (Yu et al., 2020), ionic liquid (Cuppini et al., 2021), etc., which has potential results. But they had the same limitation, that is killing all the bacteria instead of showing an antibacterial effect only during microdysbiosis. They not only killed cariogenic bacteria, but also inhibited probiotics, the balance of oral eubiosis was affected once again Li et al., 2021). Thus, a novel intelligent antibacterial resin infiltrant was needed.
We modified the resin infiltrant with DMAEM at the first time, which showed an antibacterial effect only during microdysbiosis. The total microorganisms were reduced by 1-4 logs. And the CFUs of Mutans Streptococci were reduced by 1-3 logs. Moreover, lactic acid production of the infiltrant + 5% DMAEM group was greatly reduced, which reduced by 92% compared to the infiltrant + 0% DMAEM group (control). Metabolic ability and biofilm accumulation were all reduced significantly with the increase of DMAEM concentration (p < 0.05). Therefore, the new intelligent resin infiltrant containing DMAEM are promising to inhibit biofilm growth. The pH-sensitive ability of DMAEM confers intelligent antibacterial properties to the novel resin infiltrant. DMAEM is a kind of intelligent materials, that could respond to the decrease of pH. In our previous work, DMAEM showed strong antibacterial Cell viability of normal oral keratinocyte was used to test the biocompability of DMAEM resin infiltrant. The viability was acceptable when that was not less than 80% of the control group according to ISO 10993-5-2016 standard. effects when pH was 5. While when pH was 7.4, the MIC and MBC of DMAEM were more than 13.5, which showed lower antibacterial activity . Both in vivo and in vitro experiments indicated that in resin adhesives, DMAEM provided long-term antibacterial effect via its reversible pH-responsive activity . And DMAEM sealant demonstrated possibility to prevent microleakage in sealant application (Li et al., 2021). Also, it has been proved that DMAEM could maintain the diversity of oral microbiome, and was friendly to commensal microbiota due to its pH-sensitive activity . The reason why DMAEM showed antibacterial effect when pH was 5 is for the nitrogen atoms of DMAEM. In lower pH, the nitrogen atoms could be protonated Antibacterial effects of composites on saliva-derived biofilm. (A) The biofilm metabolism evaluation, (B) biofilm accumulation, (C) production of lactic acid of the biofilm sites (mean ± SD; n = 6). The different letters indicate the significant difference between the bars (a, b, c). CFUs of the biofilm. (A) CFU of total microorganisms, (B) CFU of total streptococci, (C) CFU of mutans streptococci (mean ± SD; n = 6). The different letters indicate the significant difference between the bars (a, b, c).
to form QAMs, which showed strong antibacterial effect, while as pH increases, they are deprotonated and returned to DMAEM structure Li et al., 2021). QAMs were proved strong antibacterial effect: the electrostatic interaction between the negatively-charged bacterial cell membrane and the positivelycharged (N+) sites of a QAM resin causes the bacterium bursts; in addition, QAMs with long alkyl chains can physically pierce the bacterial cell wall, puncturing the cell membrane and releasing its cellular contents (Cheng et al., 2017). Therefore, DMAEM resin infiltrant could overcome the limitations of the present materials. Although DMAEM yielded covalently grafting to the resin infiltrant, which should have a long-term effect theoretically, future experiments are still needed to investigate the longevity of antibacterial effect. Representative images of live/dead stained biofilms grown for 24 h on composites. (A) Infiltrant + 0% DMAEM group, (B) infiltrant + 1.25% DMAEM group, (C) infiltrant + 2.5% DMAEM group, (D) infiltrant + 5% DMAEM group, (E) percent of live bacteria(%). Live bacteria were stained green and dead bacteria were stained red. Infiltrant + 0% DMAEM group had primarily live bacteria, while infiltrant + 5% DMAEM group produced mostly red staining.
Enamel surfaces are constantly exposed to the oral microflora (Arslan et al., 2015). Therefore, we used a salivaderived biofilm to evaluate antibacterial effect and antidemineralization effect of resin infiltrant. To simulate the conditions of caries, sucrose was added to the culture medium. Thus, cariogenic bacteria such as S. mutans and Lactobacilli metabolize carbohydrates to acids, causing demineralization of the tooth structure and the tooth-restoration margins beneath Enamel hardness. (A) Enamel samples were demineralized in acid solution, after then, infiltrant was applied as Figure 1. (B) Enamel samples were first demineralized like (A) and after infiltration, all the samples were second demineralized under biofilm (mean ± SD; n = 6). The different letters indicate the significant difference between the bars (a, b, c). The resin infiltration showed similar hardness value of sound enamel (p > 0.05), and after biofilm attack, only the hardness of the infiltrant + 5% DMAEM group did not decrease (p > 0.05).
the biofilm . Furthermore, to imitating an extreme clinical situation, and allowing demineralization to occur in a short time, the biofilms were cultured for 7 days continuously. The biofilm model successfully resulted in significant demineralization within 7 days, comparable to that with the use of a chemically-prepared acidic gel system for 21 days (Zhang et al., 2019). The biofilm model used in this study had main advantages over the traditional chemically-induced demineralization. It is more clinically relevant as it mimics a cariogenic situation, and it takes less time to construct a caries demineralization model.
Demineralization, which could reduce the enamel hardness, is an important process as well as outcome in the occurrence of WSLs. The microorganisms along the demineralized area could re-penetrate deeper, leading to further demineralization, and finally the cavities formed (Paris et al., 2013;Neres et al., 2017). Therefore, it is important to evaluate whether the resin infiltrant can inhibit demineralization. It has been proved that resin infiltrant could s i g n i fi c a n t l y i n c r e a s e b o t h m i c r o -h a r d n e s s a n d demineralization resistance of enamel, which prevent or reduce the progression of caries (Brignardello-Petersen, 2020). Reported surface micro-hardness (352 HVN ≈ 3.45 GPa) for sound human enamel was consistent with sound bovine enamel hardness in our study; moreover, the hardness of natural carious enamel (0.29-3.29 GPa) was similar to hardness of artificial caries found in this study, either (Maupoméet al., 1998;Huang et al., 2010;Paris et al., 2013). After resin infiltration, the hardness of demineralized enamel recovered as sound enamel. It is doubtful whether the hardness could recover after resin infiltration. Some studies showed that the hardness of infiltrated enamel was less than sound enamel, while others showed increased hardness value (Maupoméet al., 1998;Huang et al., 2010;Paris et al., 2013;Dai et al., 2022). For instance, in Dai's in vitro study, the enamel treated with ICON showed lower hardness (1.7 GPa) compared to sound enamel and experimental resin infiltrant containing TEGDMA and BisGMA (Dai et al., 2022). The reason may be that the ICON infiltrant possessed a polymer network mainly consisting of TEGDMA, which composition is likely to lead to lower hardness properties for the ICON infiltrant (Chen et al., 2019;Dai et al., 2022). Furthermore, BisGMA and TEGDMA based infiltrants may reduce polymerization shrinkage due to higher molecular weight (Paris et al., 2013). And except for the different materials, the operation could be another reason. For example, removing the excess infiltrant, polishing the surface, and the way to etch the surface could be different, that may influence the outcome (Zakizade et al., 2020). Then, after oneweek biofilm attack, the hardness decreased again except for the infiltrant + 5% DMAEM group. In addition, the hardness value increases with the increase of DMAEM concentration. Our results indicated that under the cariogenic microbial environment, the resin infiltant may not be enough to resist demineralization. The possible reason for the continued decrease in the other groups is the continuous biofilm attack under an extremely severe cariogenic condition, for three reasons. First, unlike the oral environment, brushing teeth or chewing will partially eliminate the biofilm. Second, there wasn't any buffer in the medium, and more sugar was added to mimic an environment prone to caries, and that was much more aggressive than what happened intraorally, where the saliva served as a buffer. Third, no source of minerals for remineralization was added in the medium, but there were minerals in the saliva intraorally. Considering resin infiltrant only seals the demineralized area, and the DMAEM resin infiltrant was a contact antibacterial material, future studies could combine release type of antimicrobial agent to further prevent the caries progression. Furthermore, although the experimental resin infiltrant was used in several studies, it did not outperform the commercial infiltrant (Paris et al., 2013), so more work needs to be done before it can be applied clinically.
Under these conditions, with the modification by DMAEM, the novel resin infiltrant showed comparable biocompability, mechanical properties, and strong antibacterial effect in an acidic environment. Moreover, these results demonstrated the antidemineralization properties in the carious oral environment, which could prevent the progression of WSLs.

Conclusion
Development of novel dental materials that show an antibacterial effect only during microdysbiosis is an ideal way to inhibiting WSLs while maintaining a healthy oral eubiosis. In the present study, a novel pH-sensitive resin infiltrant containing DMAEM was synthesized for the first time. The new resin infiltrant presented good biocompability when the mass fraction of DMAEM was below 5% (p > 0.05). Biofilm metabolic activities, biofilm biomass, lactic acid production were substantially reduced, and biofilm CFU was reduced by up to 3 log. After acid attack by acid solution and then under by biofilms, the infiltrant + 5% DMAEM group produced an enamel hardness of 2.90 GPa, much higher than 0.85 GPa of the control infiltrant + 0% DMAEM group. Therefore, the novel intelligent resin infiltrant is highly promising for enamel infiltration to protect tooth structures and inhibit dental caries.