The purpose of this study was to compare the metabolism of Streptococcus mutans biofilms after 1-7 days of growth on different orthodontic adhesives.
Specimens of 6 commercial orthodontic adhesives were fabricated in custom-made molds and polymerized using a light-emitting diode light-curing unit. Bioluminescent S mutans (UA159:JM10) biofilms were grown on ultraviolet-sterilized specimens for 1, 3, 5, and 7 days (n = 18 biofilms/d/product) in anaerobic conditions at 37°C. The metabolism of biofilms (relative luminescence unit [RLU]) was measured 0, 2, 4, and 6 minutes after exposure to D-luciferin solution using a microplate reader. A linear mixed-effects model was used to analyze the logarithm of RLU (log RLU). The model included fixed effects of products, days, and minutes. Tukey-Kramer post-hoc tests were then performed on the significant predictors of log RLU (α = 0.05).
Days ( P <0.0001) and minutes ( P <0.0001) were independent predictors of log RLU, but the products were not ( P = 0.5869). After adjusting for minutes, the log RLU was analyzed with a post-hoc test, and all differences between days were significant with the exceptions of day 3 from day 5 ( P = 0.0731) and day 5 from day 7 ( P = 0.8802). After adjusting for day, log RLU was analyzed with a post-hoc test and all differences in minutes were significant.
No significant differences in the metabolism of S mutans biofilms were observed among the 6 orthodontic adhesives. Biofilms that were grown for 3 days demonstrated the highest levels of biofilm metabolism as evidenced by higher mean log RLU values relative to 1, 5, and 7-day growth durations.
A high-throughput bioluminescence assay was used to assess intact biofilms’ metabolism on orthodontic adhesives.
The metabolism of Streptococcus mutans biofilms on 6 orthodontic adhesives was not different.
Biofilms grown for 3 days showed a higher metabolism than 1, 5, and 7-day growths.
Potential side effects of fixed orthodontic appliances include plaque accumulation, enamel demineralization, and white spot lesions (WSLs), which can develop within 1 month after their placement. , These types of iatrogenic enamel lesions are typically located in regions surrounding bracket bases toward the gingival aspect of affected teeth. , Depending on their severity, the presence of WSLs after treatment, at best, compromises esthetic outcomes and, at worst, necessitates costly and invasive restorative procedures. The prevalence of WSLs in orthodontic patients has been reported to range from 2% to 96%, which is a significant issue in orthodontics. , This problem is exacerbated when WSLs are closer to the gingiva and/or when an excess adhesive flash is left outside the bracket base area. These factors have been demonstrated to adversely impact the ability of patients to perform adequate oral hygiene techniques , and to significantly increase retentive sites available for biofilm growth.
Several materials are used as orthodontic adhesives, including composite resins, glass ionomer cement (GIC), resin-modified glass ionomer cement (RMGIC), and poly-acid modified composites. Orthodontic adhesives on the basis of traditional composite resins formulations are typically used because of their significantly higher bond strengths at 30 min and 24 hours compared with GIC and RMGIC. Composite resins also offer superior mechanical properties and handling characteristics relative to their GIC counterparts and, therefore, have risen in popularity among clinicians. Previous studies have indicated that polymer-based materials upregulate the adhesion and growth of oral biofilms (plaque). , Sukontapatipark et al, while investigating the correlation between bacterial colonization and fixed orthodontic appliances, demonstrated the presence of bacteria-filled gaps (≅10 μm) at the tooth-adhesive interface, thereby indicating the susceptibility of that interface to bacterial accumulation. These findings suggest that orthodontic adhesives are one of the most important risk factors for the development of enamel demineralization during orthodontic treatment.
Two systematic reviews have recently concluded that the oral microbiome is altered by the presence of fixed orthodontic appliances and reported elevated counts of Streptococcus mutans and Lactobacillus species. , These bacterial strains play important roles in enamel demineralization and caries development, and may support enamel decalcification associated with WSLs. , Although the presence of S mutans in the oral cavity is normal, the ecology of biofilms may shift from a state of health to a disease-associated state because of increased levels of S mutans . , The simple incorporation of well-known antibacterial agents (eg, fluoride and chlorhexidine) into commercially available polymer compositions have been proposed as an approach to overcome the problem of enamel demineralization during orthodontic treatment. Although experimental antibacterial materials were initially capable of reducing bacterial attachment and growth of oral microorganisms, long-term antibacterial properties have not been demonstrated. In addition, leaching of antibacterial agents into the oral cavity has been demonstrated to result in experimental materials displaying rough surfaces and poor mechanical properties, thereby decreasing enthusiasm for their use.
Several studies have investigated the adhesion of radiolabeled-streptococci, S mutans and Streptococcus sobrinus to different types of brackets (shape and composition) and orthodontic adhesives (RC, GIC, RMGIC) over different periods ranging from 2 to 24 hours. Their findings revealed that the materials investigated were capable of modulating the early adhesion and growth of cariogenic bacteria. A recent study investigated the aspects of multispecies biofilm formation (1 and 4 days) on orthodontic adhesives with various surface characteristics using real-time polymerase chain reaction (RT-PCR). The results reported have indicated that no differences were observed in the adhesion of total bacteria. However, the adhesion of S mutans , S sobrinus , Porphyromonas gingivalis, and Aggregatibacter actinomycetemcomitans were higher to RMGIC when compared with the other orthodontic adhesives tested (RC and compomer).
Although RT-PCR and colony counts have been successfully used to determine bacterial viability, these assays require high levels of biofilm manipulation to separate cells from extracellular polymeric substances either by sonication, vortexing, or the use of matrix-dissolving enzymes. Such intense manipulation may impact cells’ metabolism and result in imprecise cell counts, which makes these methods unsuitable for longitudinal investigations. These methods may also cause damage to cells, loss of viability, and incomplete detachment of the biofilm from specimens, all of which may produce unrepresentative results. Esteban Florez et al have introduced and validated a minimally-invasive bioluminescence assay using a bacterial construct developed by Merritt et al (UA159:JM10). The assay reported was capable of assessing, in a real-time manner, the metabolism of intact S mutans biofilms grown on resin composites. More recently, Esteban Florez et al have optimized that assay as a high-throughput format capable of screening the antibacterial efficacies of commercially available and experimental dental adhesive resins. Therefore, the objective of the present study was to investigate the metabolism of S mutans biofilms grown on orthodontic adhesive resins for 1, 3, 5, and 7 days using a high-throughput and nondestructive bioluminescence assay in a real-time manner.
Material and methods
Disk-shaped specimens (n = 18 adhesive/d) having diameters of 6.0 mm and heights of 0.5 mm were fabricated in a single increment using 6 commercially available orthodontic adhesive products (BracePaste [BP], American Orthodontics, Sheboygan, Wis; GrēnGloo [GG], Ormco, Orange, Calif; GoTo [GT], Reliance Orthodontic Products, Itasca, Ill; NeoBond [NB], Dentsply Sirona, York, Pa; Opal Bond MV [OB], Ultradent Products, South Jordan, Utah; and Transbond XT [TX], 3M Unitek, Monrovia, Calif). The specimens were fabricated in custom-made polyvinyl siloxane (Reprosil, Dentsply Sirona) molds with mylar strips and clean glass slides placed on the top to ensure smooth and uniform surfaces. The specimens were then photopolymerized using a light-emitting diode light-curing unit (Model 5919; VALO LED, Ultradent Products), and any excess adhesive was mechanically removed from the edges of polymerized specimens. After that, specimens were sterilized using ultraviolet-C irradiation (254 nm, 0.8 J/cm 2 , UVP Crosslinker CL-1000; Analytik Jena US, Upland, Calif) and were transferred to individual wells of sterile, transparent 24-well microtiter plates (Falcon; Corning Inc, Corning, NY) containing 1.0 mL of sterile ultrapure water. Sealed plates containing the specimens were incubated at 37°C for 24 hours to extract unreacted monomers after a previously published protocol in preparation for biofilm growth.
Colonies of JM10 (UA159 derivative) were grown in anaerobic conditions on TH agar plates (Todd-Hewitt; BD Difco, Franklin Lakes, NJ) containing 0.3% of yeast extract (EMD Millipore, Chicago, Ill) and Spectinomycin (Spec) (800 μg/mL; MP Biomedicals, Solon, Ohio) at oral temperature for 48 hours. After 2 passages, a single colony was inoculated into Todd-Hewitt with Yeast (THY) extract broth (4.0 mL) supplemented with spectinomycin (32 μL; concentration 100 mg/mL). Liquid cultures were then incubated at oral temperature (16 hours; static, anaerobic culture). Planktonic cultures displaying optical densities at 600 nm of at least 0.9 (corresponding to ≥6.43 ×10 colony-forming unit [CFU]/mL) were used for biofilm growth. Aliquots (1.0 mL) of inoculated growth media (0.65× THY with 0.1% [w/v] sucrose; 1:500 dilution) were dispensed into separate wells of sterile, transparent 24-well microtiter plates containing sterile specimens (previously described above). Plates were then incubated in anaerobic conditions as static cultures for either 1, 3, 5, or 7 days at oral temperature, with media being replaced every 24 hours during each growth period. After each biofilm growth period, specimens were manually washed with phosphate-buffered saline (PBS; pH 7.4; 2 washes, 15 s/wash at 25°C) to remove nonadherent cells after a previously published protocol. An additional set of specimens (n = 3 specimens/adhesive/d) served as the negative control group. Control specimens were subjected to the same fabrication, sterilization, monomer extraction, and biofilm growth procedures for 7 days and were subsequently treated with 2% chlorhexidine gluconate solution (1.0 mL/well) for 2 minutes. After each growth period, biofilms were aseptically transferred to sterile, white 24-well plates and replenished with fresh 1× THY + 1% (w/v) glucose (recharge medium). Biofilms were then incubated at oral temperature for 60 minutes in preparation for bioluminescence analysis.
A computer-controlled system coupled to a Synergy HT microplate reader (Biotek Instruments, Winooski, Vt) was used to add aliquots (0.5 mL) of D-luciferin (LUCNA-300, GoldBio, St Louis, Mo) aqueous solution in a 1:2 ratio at a concentration of 100 mM in 0.1 M citrate buffer (pH 6.0) to individual wells containing specimens. , After that, individual plates were placed on an orbital shaker (20 s/plate at 100 rpm). Bioluminescence measurements (relative luminescence unit [RLU]) were performed immediately after (ie, 0 minutes) and in 2-minute increments up to 6 minutes after the addition of D-luciferin at 590 nm. Bioluminescence measurement procedures were obtained after the 1, 3, 5, and 7 day periods of biofilm growth.
A linear mixed-effects model in SAS software (version 9.3; SAS Institute, Cary, NC) was used to analyze the logarithm of RLU. This model included fixed effects of products, days, minutes, and all 2-way interactions. The factors of products, days, and minutes were treated as categorical variables. Values of the control group specimens were excluded from the statistical analysis because their primary purpose was to confirm the killing of the biofilms with 2% chlorhexidine gluconate solution. The covariance among 4 measurements on the same specimen each day was modeled using an unstructured covariance pattern. A backward model selection strategy was adopted for model selection such that interaction terms with P <0.05 were sequentially dropped from the model. A Tukey-Kramer post-hoc analysis was performed on the significant predictors: days and minutes.
The objective of the present study was to assess the metabolism of S mutans biofilms grown on 6 commercially available orthodontic resins (BracePaste, GrēnGloo, GoTo, NeoBond, Opal Bond MV, and Transbond XT) using a newly developed nondestructive and real-time high-throughput bioluminescence assay. Such analysis was possible because a previous study demonstrated the presence of strong and positive correlation levels between the metabolism of biofilms and the viability (colony counts) of sonicated bacteria of the model microorganism used in the present study. The log-transformed RLU (log RLU) was used in the final analysis after model diagnostics. All 2-way interactions were not significant, and consequently, the final model included the main effects of products, days, and minutes without any 2-way interactions. According to the linear mixed-effects model, days ( P <0.0001) and minutes ( P <0.0001) but not products ( P = 0.5869) were independent predictors of RLU.
The results reported in Figure 1 , A – F illustrate the mean and standard deviation values of log RLU categorized by time after exposure to D-luciferin solution of S mutans biofilms grown for 1, 3, 5, and 7 days against the surfaces of the 6 adhesives investigated. No significant differences were observed among mean log RLU values of the orthodontic adhesives. After adjusting for the duration of biofilm growth (including all products), all differences in time after exposure to D-luciferin solution were significant, as shown in the Table . The largest difference was observed between minute 6 and minute 0, with log RLU values on average being 0.3784 times lower at minute 6 than at minute 0 ( P <0.0001). The smallest difference was observed between minute 4 and minute 6 with an average log RLU of 0.1058 times lower at minute 6 than at minute 4 ( P <0.0001).