2.1 Light source samples

We used five different types of LED dental lights (Figures 1) as experimental samples. The LED modules had peak wavelengths at 390nm, 405nm, 440nm, and 630nm for testing the S. mutans LED-light inhibition effect, and 390nm, 405nm, 630nm, and 850nm LED modules were used for normal cell viability testing. Previous studies on LED light sources for skin care have shown that the low-wavelength LEDs have a high bactericidal effect ⁽¹⁴⁾. Therefore, in the S. mutans inhibition test, 390nm, 405nm, and 440nm wavelength light sources were selected as test samples. In addition, among the LED light sources of various wavelengths, the 630nm light source is the light source of the wavelength range expected to have the highest cell activation effect ⁽⁹⁾, and if the S. mutans inhibition effect is confirmed, it is judged to be suitable as the light source for oral care. So samples were added. LED light sources for cell activity tests were selected for 630nm and 850nm light sources⁽¹⁰⁾. The 390nm and 405nm light sources are expected to be superior to the S. mutans inhibition effect, but they were added to the test group of the cell activity test to determine the degree of inhibition of normal cell activation.

2.2 Measurement of optical properties

Table 2 lists the specifications of the goniophotometer used for measuring the optical properties of the LED lights (Figures 1). For the cell viability experiment, we used a cell culture plate measuring 3.5cm in diameter (Figure 2) and set the irradiation distance of the LED sources to 1.4cm, considering their use inside the oral cavity.

For the post-irradiation cell viability test (Figure 3), we measured the radiant flux (W) with a fixed detector, rotating the LED module embedded in the goniophotometer with the rotation conditions of a 2° step in the γ plane over the range 0∼90°, and in the C plane a 90° step over the range 0∼360°.

We calculated the half angle θ of the light irradiated by the LED from the radius of the cell culture plate and the irradiation distance, as expressed by Eq. (1).

Where θ is the half angle in degrees, X is the radius of the cell culture plate (cm), and Y is the irradiation distance (cm), Z is defined as the distance from the center point of the LED module to the bottom end of the cell culture plate (cm).

The half angle θ is half of the total cone apex angle (viewing or beam angle), i.e. the angle of irradiation onto the cell culture plate in the post-irradiation cell viability test. By inserting an irradiation distance of 1.4cm into Eq. (1), the half angle is calculated as 51.3°, and a final value of 52° was obtained under the goniophotometer configuration of the 2° step in the γ plane. If the irradiation distance is 2cm, the half angle is 42°. The radiant flux (Wθ) in the half angle was calculated using the goniophotometer, and the irradiance (Eθ) was calculated using Eq. (2).

where Eθ is measured in ㎽/㎠, Wθ is the total radiant flux in the half angle, in ㎽, and dA is the total area of the bottom of the cell culture plate, in ㎠. Finally, the radiant exposure, in J/㎠, was calculated by multiplying the irradiance, in ㎽/㎠, by the irradiation time, in seconds, on the cell plate.

2.3 Post-irradiation cell viability test

We evaluated the post-irradiation cell viability after exposing the cells in the cell culture plate (Figure 2) to the LED modules from a distance, Y, of 1.4cm ^(12,16-18). Two samples were prepared for each experimental group and measurements were repeated in triplicate, i.e., six measurements for each experimental group. Tables 3 and 4 present the S. mutans inhibition effect and fibroblast activation test-conditions, respectively, where CW stands for the continuous-wave mode, and PW1 and PW2 denote two different square-wave pulsed modes. The conditions PW1 and PW2 were configured to have pulse widths of 35ms and 18ms, respectively, the period in both cases was 36ms and the duty cycle in the latter case was 50% ^(10,19), which ensured that the blinking of the LED module was not perceptible to the naked eye.

2.4 Test of the S. mutans inhibition effects

The test of the S. mutans inhibition effect was performed on the strain KCOM 1504, provided by the Korean Collection for Oral Microbiology (KCOM, Gwangju, Korea). S. mutans is a facultative gram-positive bacterium that forms biofilms (early dental plaques) in the human oral cavity. The cell culture plates inoculated with S. mutans were irradiated under the conditions outlined in Table 3 and incubated in an incubator (LIB-150M, DAIHAN Labtech Co., Korea) for 24hrs at 37℃.

After incubation, the culture medium was removed and the unattached cells were removed by washing, twice, carefully, using phosphate buffered saline (PBS). After adding 200㎕ of fluorescent dye (SYTOⓇ9, propidium iodide) to each sample to stain the cells, we wrapped the cell culture plates with aluminum foil to shield them from light and left the samples to react at room temperature for 15min. Next, we thoroughly washed away the residual staining solution using PBS and then observed the samples using a confocal laser scanning microscope (Leica TCS SP5 AOBS/tandem, Leica, Germany) and measured the S.mutans biofilm thickness using a dedicated analysis program (Leica LAS AF software, Leica Microsystems, Bensheim, Germany).

2.5 Fibroblast cell activation test

Mouse fibroblast cell line NCTC clone 929 (L929) was used for the fibroblast cell activation test with LEDs of different wavelengths. We incubated fibroblast cells in an incubator (Forma Series II 3111 Water Jacketed CO2 Incubator, Thermo Scientific, USA) in the culture medium RPMI-1640 (Roswell Park Memorial Institute 1640, Lonza, Basel, Switzerland) containing 10% fetal bovine serum (FBS) and 1% penicillin at 37℃ in an atmosphere with 5% CO2.

Mouse fibroblast cells were prepared at a density of 2×104cells/㎖, and aliquots of 2㎖ per well were seeded into a 96-well 3.5-cm-diameter culture plate. The cells were then incubated in the incubator in a chamber with a 5% CO2 atmosphere at 37℃. They were then irradiated under the conditions outlined in Table 4 and incubated again in the incubator for 24h. After irradiation, 200㎖ of EZ-Cytox (Itsbio, Korea) was added to each plate(Figure 4), and the EZ-Cytox-treated samples were left to react in the incubator for 30min. We then measured the absorbance at a wavelength of 450nm using an absorbance measurement device (Microplate (ELISA) reader, ELx 800UVⓇ, Bio-Tek Instrument. Inc., USA) ^(13-14).

3.1 S. mutans inhibition test results and discussion

Table 5 presents the irradiation test results relating tothe inhibitory effect of LED light on S. mutans. The mean values were obtained from six repeated measurements of the thickness (in µm)of the S.mutans biofilm. The biofilm thickness in the control group (no irradiation) is 1µm. The standard deviation (Std) of the measured biofilm thickness values is 0.18 for the E and K experimental groups and 0.09 for other experimental groups.

Three LED modules demonstrated inhibition of S. mutans biofilm formation of greater than 50%, with statistically significant differences (p < 0.05) in the S. mutans inhibition effect among the experimental groups: K* (630±1nm, 196s) < A* (390±3nm, 35s) < E* (405±1nm, 199s) (Figure 5). The mean inhibition rates of the modules of wavelengths 390±3nm, 405±1nm, 440±1nm, and 630±1nm were 48%, 29%, −38%, and 50%, respectively. It was found that the lower the wavelength, the higher is the inhibition rate for an irradiation time of 70s or shorter.

However, the experimental group with the lowest biofilm formation thickness was irradiated by the 405±1nm LED module, which suggests that a low peak wavelength is not necessarily a required condition for a strong S. mutans inhibition effect and that irradiation time greatly influences the inhibitory effect. However, the irradiation time is not proportional to the inhibition rate. The 390±1nm LED module showed the highest inhibition rate, 53%, at the shortest irradiation time, 35s, which decreased to 30% at 70s and increased again at 141s.

In the analysis results, the effective energy means radiant exposure including irradiation time with the S. mutans inhibition effect. Analysis of the test results revealed the energies and irradiation times for each LED module that are advantageous for inhibiting S. mutans biofilm formation (p < 0.05) as follows :

390±3nm LED module :

① [effective energy] ≦ 2.34J/㎠ (35s) ≦ [effective energy]

② [effective energy] ≦ 9.41J/㎠ (141s) ≦ [effective energy]

405±1nm LED module :

① [effective energy] ≦ 13.86J/㎠ (199s) ≦ [effective energy]

② [effective energy] ≦ 13.22J/㎠ (255s) ≦ [effective energy]

③ [effective energy] ≦ 9.27J/㎠ (137s) ≦ [effective energy]

440±1nm LED module :

① [effective energy] ≠ 4.64J/㎠ (59s) : onset of toxicity

② [effective energy] ≦ 13.92J/㎠ (177s) ≦ [effective energy]

630±1nm LED module :

① [effective energy] ≦ 2.35J/㎠ (196s) ≦ [effective energy]

Looking at the difference in the biofilm formation inhibition effect depending on the irradiation distance, the inhibition rates at irradiation distances of 1.4㎝ and 2㎝ for the 405±1nm LED module were 59% and 30%, respectively, despite the difference in the radiant exposure input being less than 5%. This demonstrates that the shorter the irradiation distance, the stronger is the S. mutans inhibition effect of the LED light, even with similar radiant exposure.

Using the 405±1nm LED module, we compared the biofilm formation inhibition rates of the CW and PW LED lights with respect to the radiant exposure input. The S. mutans biofilm formation inhibition rate of PW1 (period = 36ms, pulse width = 35ms) was 34% at a radiant exposure of 9.27J/㎠. Upon comparing experimental groups E* and F*, the biofilm formation inhibition rate was found to decrease by 29% as their radiation distance increased from 1.4㎝ to 2㎝, despite the difference in the radiant exposure input being less than 5%. Incontrast, experimental group G* showed a biofilm formation inhibition rate that was 25% less than that of experimental group E* at a 34% lower radiant exposure input, when irradiated by PW LED light. This suggests that it is more efficient to use PW light than to adjust their radiation distance in order to improve the S. mutans inhibition effect of LED light.

In contrast to the finding of one study, according to which the disinfection or antimicrobial effect of the LED increases as the peak wavelength decreases ⁽¹⁷⁾, the 630±1nm LED module outperformed the 440±1nm LED module, exhibiting an inhibition rate greater than 50% at an irradiation distance of 1.4cm, irradiation time of 196s, and radiant exposure of 2.35J/㎠. In particular, the radiant exposure of the 630±1nm LED module was 83% lower than that of the 405±1nm LED module (E* experimental group) whose S. mutans effect inhibition was 59%. This proves that 630±1nm is the LED module wavelength with the greatest S. mutans inhibition effect per unit radiant exposure.

Figure 6 shows confocal laser scanning microscope images of S. mutans colonies in the control and experimental groups. Live and dead cells are displayed in green and red, respectively. It was confirmed that irradiation by the 390±3nm, 405±1nm, and 440±1nm LED modules for 118s or longer resulted in partial destruction of S. mutans colonies. A study on Propionibacterium acnes, a bacterium belonging to the normal flora of the human skin, verified the disinfection effect of 370∼470nm LED light ⁽¹⁴⁾. However, the findings of the present study, given the presence of live cells in most of the experimental groups, indicate that LED light of each wavelength band has a growth inhibition effect on S. mutans rather than a disinfection effect.

3.2 Fibroblast activation test results and discussion

Table 6 presents the post-irradiation mouse fibroblast (L929) activation test results. The mean values were obtained from six repeated measurements of the post-irradiation mouse- fibroblast cell viability (in %). Std indicates the standard deviation for the measured viability. The mean viability and Std of the control group are 1% and 0.06%, respectively.

Two LED modules resulted in mouse fibroblast viability values in excess of 10%, and the viability increased in the following order of experimental groups: U*(850±1nm, 669s) < T*(630±1nm, 1,570s) < S*(630±1nm, 807s) < P*(630±1nm, 196s) with statistical significance (p < 0.05). The 630±1nm LED module demonstrated the highest cell viability at an irradiation distance of 1.4㎝ and an irradiation time of 196s (an increase of 43%).

In the analysis results, the effective energy means radiant exposure including irradiation time with the fibroblast activation effect. Analysis of the test results revealed the energies and irradiation times for each LED module that are advantageous for mouse fibroblast viability (p < 0.05) as follows :

405±1nm LED module:

① [effective energy] < 13.22J/㎠ (255s)

② [effective energy] ≦ 2.3J/㎠ (33s) ≦ [effective energy]

630±1nm LED module:

① [effective energy] < 4.7J/㎠ (392s)

② [effective energy] ≦ 2.35J/㎠ (196s) ≦ [effective energy]

850±1nm LED module:

① [effective energy] ≦ 9.5J/㎠ (669s) ≦ [effective energy]

The 630±1nm LED module was found to have the highest cell viability increase among all the experimental groups. In contrast, the 405±1nm LED module caused the cell viability to decrease. This allows us to hypothesize that the 630nm LED light has a wide range of effective energies and irradiation times for enhancing cell viability, whereas the 405±1nm LED light has a narrower scope for cell viability, which would need to be reflected in oral-care device design. In the 405±1nm LED module experimental groups, cell toxicity was observed at the radiant exposure of 13.2J/㎠(p < 0.05). Whereas cell viability increased in all of the 630±1nm and 850±1nm LED experimental groups, it decreased for the 405±1nm LED module experimental group at lower input radiant exposure. That is, as the peak wavelength becomes lower than 405±1nm, the effective energy band for cell activation becomes narrower. Furthermore, cell viability decreased by 8%, albeit without statistical significance, when the 390±3nm LED module was irradiated for 141s, from which we can imply that cell toxicity tends to increase at peak wavelengths lower than 405nm. When comparing M* and N*, for the 405±1nm LED module experimental groups, N* with lower irradiance showed lower fibroblast cell viability. Increase in radiant exposure even at a low irradiance resulted in 9% cell toxicity, which suggests that a radiant exposure exceeding the abovementioned energy value can trigger the toxicity effect.

The fibroblast viability per unit radiant exposure increased in the order of 405±1nm < 850±1nm < 630±1nm for the respective LED modules. Accordingly, the 630±1nm LED module with the highest effective energy efficiency for cell activation is considered the best-suited LED module for a dental care device for cell activation. However, to avoid health risks, care should be taken not to exceed an irradiation time of 40min, given the result of a study that showed that irradiation with 660nm and 840nm LED lights for 20 and 40min, respectively, can induce cell toxicity ⁽¹¹⁾.

In the 405±1nm and 630±1nm LED modules, the PW mode was advantageous over the CW mode for improving fibroblast viability. The 630±1nm LED module showed cell viabilities of 7%, 21%, and 18% at the same radiant exposure of 9.42J/㎠ under the CW, PW1 (period 36ms, pulse width 35ms), and PW2 (period 36ms, pulse width 18ms, duty cycle 50%) irradiation conditions, respectively. In other words, cell viability was found to be higher in the PW mode than in the CW mode, and higher in PW1 (broader pulse width) than in PW2. Therefore, PW is more desirable in clinical applications for improving the mouse-fibroblast activation effect, given that CW light is known to have low transmissivity, whereas PW has been reported to have high transmissivity in tissue ⁽⁹⁾.

LED light therapy is generally recognized as advantageous for leaving no chemical residues and for being safe and non-invasive. However, the results of this study revealed that LED dental lights could induce activation or toxicity of normal cells, depending on the irradiation time and the radiant exposure. In particular, the fact that pulsed-mode LED operation allows the adjustment of radiant exposure within a specific irradiation time should be taken into account when designing photonic medical devices.