Application of D-Amino Acids as Biofilm Dispersing Agent in Dental Unit Waterlines

Aim and Purpose Biofilms in dental unit waterlines (DUWLs) are extremely difficult to eliminate. Aim of this study is to evaluate the efficacy of a mixture of four D-amino acids on biofilm dispersion in DUWLs. Materials and Methods A mixture of four D-amino acids (D-methionine, D-tryptophan, D-leucine, and D-tyrosine, 10 mM each), distilled water (control), and 0.1 M hydrochloric acid (HCl) was used in the experiment. In laboratory, pieces of DUWLs covered with biofilms were submerged in different solutions for 5 days, flushed, and measured OD600 of the dispersed biofilms. Remnants of biofilms on the DUWLs were evaluated by SEM. In clinic, fifteen DCUs were incubated with test and control solutions, flushed, and measured OD600 of the dispersed biofilms. Microbial count of DUWL output water was enumerated twice a week for four weeks. Results There was a slight, but not significant, increase in OD600 of flushing water in D-amino acids group. D-amino acids effectively reduced bacterial plaque as demonstrated by SEM. Incubation with D-amino acids significantly reduced biofilms especially after the first day of flushing. Bacterial count in DUWL output water was significantly reduced after treatment with D-amino acids. Conclusion D-amino acids are applicable as biofilm dispersing agents in DUWLs.


Introduction
Bio lms, the term for a community of microorganisms, are resistant to physical and chemical stresses. Bio lms can be found in any aquatic environment including pipelines of medical devices [1]. In dentistry, dental unit waterlines (DUWLs) consist of narrow (approximately 2-3 mm internal diameter) plastic tubes, and they deliver cooling water for dental equipment [1]. DUWLs are prone to bio lm formation which results in a heavily contaminated water output. is problem has been rst identi ed almost 50 years ago and is still of signi cant concern since several reports indicate bio lm-related infection for both normal and immunocompromised patients [1,2].
Numerous approaches have been introduced to decrease the presence of DUWL bio lms. ese include both nonchemical and chemical methods. Regarding a nonchemical strategy, which includes ushing, drying, and applying an antimicrobial lter, using deionized, distilled, or even sterile water appears not to a ect existing bio lms [1,2]. e application of chemical agents has been proven to be e ective [3][4][5][6]. Since regrowth of bio lms takes only a short period of time, a continuous chemical treatment is necessary. However, chemical agents can have an adverse e ect on the patient's oral tissues, dental chair unit (DCU) components, as well as dental restorative materials. is is, in particular, the case for residual agents present in DUWL output water, which enter the patient's oral cavity and may even be swallowed or inhaled from aerosols generated by dental instruments [7]. Nowadays, searching a better approach for the elimination of pathogenic bacterial bio lms is still a very relevant issue in the eld of infection control and prevention.
Amino acids (except glycine) can occur in two isomeric forms; these are called L-and D-forms. Only L-amino acids are produced by cells and incorporated into proteins. erefore, most natural amino acids are in the L-form. D-amino acids are found in the cell walls of bacteria, but not in their proteins. D-amino acids have been proposed to be stress signaling and are released during depletion of nutrients to trigger the dispersion of "old" bio lms [8].
ese D-amino acids help bacteria adapt to environmental challenges by modulating the structure and composition of peptidoglycans which are one of the major components in the bacterial cell wall [8][9][10].
D-amino acids are also involved in inhibition of bacterial bio lm formation and dispersion. Kolodkin-Gal et al. demonstrated that D-amino acids could disperse B. subtilis bio lm by a ecting the function of amyloid bers. When noncanonical D-amino acids incorporate into peptidoglycan, they interfere with the anchoring of amyloid bers that normally assists in holding the bio lms together [11]. Bucher et al. reported D-leucine (D-Leu) as a noncanonical D-amino acid which competes with D-alanine (D-Ala) for the fth pentapeptide of B. subtilis cell wall and interferes with transpeptidation and transglycosylation. e disturbance of the composition of the bacterial cell wall speci cally interferes with bio lm formation [12]. D-amino acids including D-tyrosine (D-Tyr), D-methionine (D-Met), D-FFtophan (D-Tryp), and D-leucine (D-Leu) can prevent pellicle formation [13]. e minimal inhibitory concentration (MIC) of these D-amino acids that could inhibit bio lm formation depends on the type of D-amino acid. Testing these amino acids individually revealed that the MIC of D-Met was 2 mM, of D-Tryp was 5 mM, of D-Tyr was 3 µM, and of D-Leu was 8.5 mM. D-Tyr proved to be the most e ective among the four. Nonetheless, the mixture of all four amino acids is more e ective than D-Tyr alone. As for the dispersion of bio lms, D-Tyr (3 µM) or a mixture of D-Tyr, D-Met, D-Tryp, and D-Leu (2.5 nM each) breakdowns the bio lm pellicle of B. subtilis [13]. Not only this type of bio lm was a ected but D-amino acids can also a ect Staphylococcus aureus and Pseudomonas aeruginosa bio lms [13][14][15][16].
Recently, several studies have proposed D-amino acids as candidate molecules to be applied as a bio lm dispersing agent in endodontic treatment. Rosen et al. demonstrated that D-Leu was e ective in inhibiting E. faecalis bio lms grown on human dentin slabs [17]. Zilm et al. demonstrated that a mixture of D-amino acids containing D-Leu, D-Met, D-Tyr, and D-Tryp signicantly reduced bio lm formation of E. faecalis. e inhibitory e ect of D-amino acids on bio lm formation was concentration dependent. e authors of this study, therefore, proposed D-amino acids as a novel bio lm inhibitor in endodontic treatment [18]. ese e ects of D-amino acids on bio lm dispersion together with their safety support their use as a candidate for the removal of bio lms in DUWLs. erefore, the present study aims to investigate the e cacy of D-amino acid mixtures on the removal of bio lm in DUWLs.

E cacy of D-Amino Acid Mixture on Dispersing Bio lm on DUWLs In Vitro.
DUWLs covered with bio lm, as was judged by eye, were obtained from in-use dental chair units (DCUs) and cut into small pieces with a length of 2.5 cm. A mixture of four D-amino acids, consisting of D-methionine (D-Met; #M9375, Sigma-Aldrich, St. Louis, MO), D-tryptophan (D-Tryp; #T9753, Sigma-Aldrich, St. Louis, MO), D-leucine (D-Leu; #L0027, TCI, Tokyo, Japan), and D-tyrosine (D-Tyr; TCI, Tokyo, Japan), was prepared by dissolving them in 0.1 M hydrochloric acid (HCl) to get a nal concentration of 10 mM (each). In a preliminary set of experiments, we assessed that 10 mM was the most e ective concentration of D-amino acids to be used for the removal of a bio lm.
e DUWLs were submerged in 4 mL of one of the following three solutions: (i) distilled water (negative control), (ii) D-amino acids, and (iii) 0.1 M HCl (dissolvent control). e samples were incubated at room temperature for 5 days before each sample was rinsed 3 times with 1 mL of distilled water.
e dispersed bio lm from each sample was collected to sonicate and to measure the OD at 600 nm. e DUWL samples were then xed with 2.5% glutaraldehyde for 24 hours and washed with 1 mL of PBS for 3 times. ey were then dehydrated, critical-point dried, gold-sputter coated, and examined using a scanning electron microscopy (JSM-5410 LV; JEOL, Tokyo, Japan) at a magni cation of ×2000. Bio lms found inside the reservoir bottle of DCUs were also incubated with test and control solutions for 5 days, ushed, and visualized with SEM.

E cacy of D-Amino Acid Mixture on Dispersing Bio lm on DUWLs in a Clinical
Setting. Fifteen dental chair units (DCUs) were randomly divided into 3 groups of 5 DCUs each: distilled water (negative control), D-amino acids, and 0.1 M HCl (dissolvent control). e study used DCUs of the dental school during summer break; a period during which the DCUs were not in use. Before the start of the experiment, water from the DUWLs was collected to evaluate the initial bacterial CFUs. e water samples were sonicated for 5 minutes to disperse clumps of microorganisms. en, a serial ten-fold dilution was performed before plating a 100 µL of each sample onto R2A agar plates, and these were then incubated at 35°C. e baseline CFUs were assessed after 7 days of incubation. In the D-amino acids and HCl groups, the water supplying the DCUs was replaced with a mixture of D-amino acids or 0.1 M HCl, respectively.
In order to limit the amount of D-amino acid used in each DCUs, the minimum amount of water that completely ushed to the outlet of the waterline was determined by ushing diluted gentian violet into the DUWL. Fifty-ve mL of the tested solution was determined as an appropriate amount to completely ll up the waterlines with testing solution. e test solutions were left in the tubing of DUWLs for 5 days. After 5 days of incubation, the waterlines were repeatedly ushed with 1.6 liters of water every other day for 10 days and the ushing water was collected to measure the OD at 600 nm. After ushing, 25 mL of water from each unit was collected and sonicated for 5 minutes. e samples were diluted ten-fold, plated on R2A agar, and incubated at 35°C for 7 days. e bacterial CFUs were then counted. Bacterial contamination in DCU output water samples was evaluated twice a week repeatedly for 4 weeks (Figure 1).
All data were analyzed using SPSS 18. e Kruskal-Wallis test and the Mann-Whitney U test were used for both in vitro and clinical experiments. Signi cance was set at p values < 0.05.

Results
Bio lm mass was shown to cover the inner surface of DUWLs as shown by SEM (Figure 1(b)). After 5 days of incubation, the bio lms on the inner surface of the tubes were seen to become loosened, some were even detached in the D-amino acids and HCl groups (Figure 2(a)). Fragments of bio lm collected in rinsing water were used to measure OD at 600 nm. ere was a trend of an increase of OD of dispersed bio lms in the D-amino acids group compared with control, though this did not prove to be statistically signi cant (Figure 2(b)). Evaluation by SEM showed DUWL bio lms consisting of bacterial plaque on top of which smear-like structures were seen (Figure 3). Treatment the DUWL with D-amino acid mixtures was able to disperse the bacterial plaque that covered the bio lm (Figure 3(a)). Bio lms dispersing effects were also found on bio lms on the inner surfaces of water reservoir bottle of the DCUs that were incubated with D-amino acids for 5 days (Figure 3(b)).
In the clinical setting, the highest OD of dispersed biolms was observed on the rst day of ushing in the group treated with the D-amino acid mixture (Figure 4). After treating the DUWL with control and test solutions, bacterial contamination recovered from DCU output water was analyzed twice a week for approximately 4 weeks ( Figure 5). A statistically signi cant lower number of average CFUs was found in the DUWL treated with D-amino acids compared with the distilled water control group ( Figure 6).

Discussion
Our results demonstrated that a mixture of D-amino acids consisting of D-methionine, D-tryptophan, D-leucine, and D-tyrosine at a nal concentration of each amino acid of 10 mM was e ective in dispersing bio lm in DUWLs. is cleaning e ect was found both in an in vitro setting and in a clinical setting. SEM analysis revealed that bacterial bio lms inside the DUWL were eliminated after incubation with the D-amino acid mixture for 5 days. Treatment of the DUWLs with the D-amino acid mixture also signi cantly reduced bacterial contamination in output water of DCUs. In previous studies, the e ciency of D-amino acids in dispersing bio lms has been demonstrated in vitro with single species bio lms or microbial aggregates formed by mixed cultures [13,14,19]. Our study is, for the rst time, proposing a clinical application of a mixture of D-amino acids in the removal of bio lms in medical devices like DCUs.
Many methods both mechanical and chemical have been utilized to reduce bacterial contamination in dental unit waterlines, but none of these methods were able to disperse  us, the e ectiveness of D-amino acid in the removal of existing bio lms in the complex dental unit tubing system   International Journal of Dentistry 5 appears to be an attractive way to remove bio lms from the inner parts of medical devices. Our ndings demonstrate the e ciency of D-amino acids in this process. It should be noted that also treatment of the waterlines with HCl alone partially removed bio lms.
is e ect might be due to the acidity of hydrochloric acid and its e ect on bio lm dispersion [20,21]. According to Tam et al., bio lm detachment of Streptococcus mutans occurs more rapidly as the pH drops [21]. However, HCl might corrode the metal part of the dental tubing system if it is used at a high concentration for a longer period. Although the recommended diluent for D-amino acids is 0.1 M HCl, a diluent with a neutral pH might be more suitable.
e concentration of D-amino acid mixture used in our study is slightly higher than the one used in previous studies [13,14]. Yet, Hochbaum et al. reported a dose-dependent e ect of D-amino acids with a minimum concentration of 10 mM to disperse the Staphylococcus aureus bio lm [14]. Kolodkin-Gal et al. proposed 3 µM of D-tyrosine or a mixture of D-Tyr, D-Met, D-Tryp, and D-Leu (2.5 nM each) to disrupt the B. subtilis bio lm [13]. e concentrations of D-amino acids used in previous studies were e ective against single species bio lms. In our study, a more complex model of naturally occurring bio lms was utilized, and therefore, a higher concentration of D-amino acids was employed.
Besides a bio lm-dispersing property, D-amino acids have also been demonstrated to inhibit or slow down the growth of bio lms [16,22,23]. However, there are some contradicting data regarding the e cacy of D-amino acids on inhibiting bio lm formation. Some studies demonstrated that the e ect of D-amino acids was rather strain speci c [15,22,24,25]. One of the possible mechanisms by which D-amino acids act on bio lms was proposed to trigger the disassembly of matrix-associated amyloid bers (see Introduction). is e ect was observed in B. subtilis bio lms which are known to produce this type of matrix [8,13]. However, also with organisms that do not seem to produce amyloid bers, for example, S. aureus and P. aeruginosa, D-amino acids appear to trigger the disassembly of bio lms. e exact mechanisms involved in how D-amino acids disperse bio lms need further examination [8,26].

Conclusion
A mixture of D-amino acids was shown to partly remove bio lms in dental unit waterlines.
is study, therefore, proposes the use of such a mixture as a new approach to decontaminate bacterial bio lms in dental unit waterlines. International Journal of Dentistry