Impact of Chlorine Dioxide on Pathogenic Waterborne Microorganisms Occurring in Dental Chair Units

Bacterial contamination is a problem in dental unit water lines with the consequence of implementing regular disinfection. In this study, the short-term impact of chlorine dioxide (ClO2) treatment was investigated on the microorganisms Legionella pneumophila and L. anisa, Pseudomonas aeruginosa, Escherichia coli, and Staphylococcus aureus. The environmental background was proven as an important factor regarding the tolerance to 0.4 mg/L ClO2 as saline and phosphate-buffered saline resulted in a higher bacterial reduction than tap water. Gram-positive microorganisms demonstrated higher robustness to ClO2 than Gram-negative, and microorganisms adapted to tap water showed increased stability compared to cultured cells. At high densities, substantial numbers of bacteria were able to withstand disinfection, whereby the use of 4.6 mg/L ClO2 increased the inactivation rate. A massive cell decrease occurred within the first 5 minutes with subsequent plateau formation or slowed cell reduction upon further exposure. This biphasic kinetics cannot be explained by a ClO2 depletion effect alone, because the probability of bacterial subpopulations with increased tolerance should be taken into account, too. Our results prove high disinfection efficiency to microorganisms that were rather found in correlation to the level of bacterial contamination and background solutions than the chosen concentration for ClO2 treatment itself.


Introduction
The quality of water in dental unit water lines (DUWL) is of considerable importance for safety and health in dental practices, whereby a high microbiological load of DUWLs is a widespread, long-known, and still current problem [1][2][3]. The main reason for microbiological contamination of DUWLs is generally the presence of microorganisms in tap water [4,5], composed of non-pathogenic microorganisms and potentially opportunistic pathogens. Another risk factor for the contamination of DUWLs is retrograde contamination by aspiration of patients' saliva [6,7]. Stagnation conditions and biofilm formation promote an increase in cell numbers as well [5,8]. Both dental staff and patients are persistently exposed to water directly or to aerosols due to the regular and repeated use of water-bearing instruments such as ultrasonic scalers and high-speed dental handpieces [9,10]. The regular monitoring for microbiological contamination of dental chair units is recommended by the Robert Koch Institute s guideline for infection prevention in dentistry [11] with the determination of the total bacterial count and presence of Legionella species. The Gram-negative Legionella are mainly present in warm water lines [12]. They were detected in 68% of the water samples recovered from DUWLs and 8% were identified as Legionella pneumophila (L. pneumophila) [13]. Legionella anisa (L. anisa) was found in DUWLs as well [14]. Inhalation or aspiration of contaminated aerosols and water can cause Pontiac fever or legionellosis with life-threatening symptoms [15,16]. Both diseases are predominantly related to water

Bacterial Isolates and Strains
The monitoring of microbiological contamination of dental chair units aims at the detection of Legionella with L. pneumophila as a high-risk pathogen and L. non-pneumophila frequently determined in dental chairs. In case of increased bacterial counts, it is recommended by the Robert Koch Institute s guideline for infection prevention in dentistry [11] to check for the presence of P. aeruginosa as an opportunistic waterborne pathogen. The Gram-positive strain S. aureus as a classic human skin microorganism and the Gramnegative Escherichia coli (E. coli) as a fecal bacterium can be transmitted to dental units via patient contacts. Table 1 summarizes the isolate and strains used in this study. The reference strains, originating from American Type Culture Collection, were obtained from the German Collection of Microorganisms and Cell Cultures (Leibniz Institute DSMZ, Heidelberg, Germany). The environmental isolate of L. anisa was received from a water sample of a dental chair by routine analyses according to recommendations of the Robert Koch Institute Germany [11]. L. anisa was identified by cultivation on buffered charcoal yeast extract with glycine, vancomycin, polymyxin B, cycloheximide (GVPC) agar (Xebios, Düsseldorf, Germany), and the species was determined using the biotyping technique (Bruker MALDI Biotyper, Bremen, Germany).

Adaption of Microorganisms to Tap Water and Survival Duration
As microorganisms may survive in water for a long time accompanied by physiological changes, we investigated the stability of this distinctive water-adapted status to ClO 2 treatment. To obtain an intensive cell density, single colonies from a Luria lysogeny broth (LB) agar plate (Roth, Karlsruhe, Germany) were suspended in an LB culture medium (Roth, Karlsruhe, Germany) following incubation at 36 • C overnight under constant rotation at 180 rpm. For the successive adaption to tap water, the sterilized institute s tap water was used with the physical parameters of pH 8.4, the conductivity of 490 µS/cm, and oxidizability of 1.6 O 2 /mL, measured according to DIN EN ISO 10523, DIN EN 27888 and DIN EN ISO 8467, respectively. The overnight culture was mixed with the doubled volume of sterile tap water and incubation continued at 25 • C under low shaking conditions (96 rpm) for 4 days. Finally, microorganisms were harvested by centrifugation (2000× g for 20 min) and the pellet was resuspended in sterile tap water in half of the initial volume. Incubation continued for the experimental test period at 25 • C with constant low rotation (96 rpm) to avoid biofilm formation.
Colonies of L. pneumophila serogroup 1 (ATCC 33152) and L. anisa (DSM 17627), were grown on buffered charcoal yeast extract agar (BCYE) supplemented with cysteine (Xebios, Düsseldorf, Germany), were inoculated directly in sterile tap water resulting in a density of 9-10 log 10 cells/mL. Incubation followed at 25 • C under low shaking (96 rpm) for a prolonged time period.
Aliquots of tap water-adapted bacteria were periodically analyzed for cultivability by dropping the suspensions in tenfold dilutions in a volume of 10 µL as triplicates on LB agar or in the case of Legionella on GVPC agar plates. Following incubation at 36 • C, plates were inspected until growth was visible, and resulting colony-forming units (CFUs) were counted.

ClO 2 and Inactivation Solutions
The ClO 2 solution (0.6% (v/v); Clorious2; Brenntag, Essen, Germany) was stored at 4 • C in the dark and was pre-diluted freshly in filtered deionized water before use in each case. The appropriate volume ratios were used in the experiments so that the final ClO 2 concentrations resulted in 0.4 mg/L, 1.0 mg/L, and 4.6 mg/L. Samples were incubated for the time intervals of 1 min, 2.5 min, 5 min, 15 min, and 30 min together with the indicated ClO 2 concentrations. After time intervals, the ClO 2 reactions were stopped by the addition of sodium thiosulfate (Na 2 S 2 O 3 ) (Merck, Darmstadt, Germany) ranging in final concentrations about sevenfold excess of ClO 2 in the respective assays.

ClO 2 Treatment of Microorganisms and Re-Cultivation on Solid Media
The microorganisms were inoculated onto LB agar and on BCYE agar plates with cysteine followed by incubation at 36 • C for approximately 18 h for non-Legionella and 72 to 96 h for Legionella species, respectively. Well-grown colonies were serially suspended in either sterilized tap water (Institute for Hygiene, Münster, Germany), saline (0.85% (w/v) NaCl; Merck, Darmstadt, Germany), or phosphate-buffered saline (PBS) (30 mM Na 2 HPO 4 /KH 2 PO 4 ; pH 7.2) and each adjusted to a cell density of approximately 10 7 to 10 8 cfu/mL. As S. aureus tended to aggregate, cells were sonicated carefully in an ultrasonic bath (Bandelin, Berlin, Germany) for 5 min and 35 kHz. Cell suspensions were diluted serially tenfold in the respective suspension medium to 10 4 cfu/mL. The microorganisms, which were adapted to tap water, were diluted tenfold in tap water in turn. The ClO 2 aliquots were added in concentrations as indicated to each bacterial dilution. The reactions were left to act for indicated time intervals at room temperature prior to inactivation using the Na 2 S 2 O 3 stop solution.
For cell counting, samples were diluted again in steps of ten, and volumes of 10 µL were spotted as triplicates on agar plates, non-Legionella on R2A, and Legionella species on GVPC. The CFUs were counted finally after 48 h at 36 • C for non-Legionella and after 10 days for Legionella species.

Effect of ClO 2 Depletion
To investigate a depletion effect under our conditions, two concentrations (10 3 and 10 6 cfu/mL) of heat-killed microorganisms (90 • C for 5 min) of E. coli and P. aeruginosa were mixed with ClO 2 at a final concentration of 0.9 mg/L. After 5 min pre-incubation, living cells of the respective species were added at a final concentration of 10 6 cfu/mL. Reactions were stopped with the application of Na 2 S 2 O 3 after a further 5 min incubation. Survival was analyzed by CFU determination as described above.

Analysis of the Data
Each series of tests was carried out at least two to three times. The plots given here refer to the minimum effect of ClO 2 in a test batch.
The cell counts of each dilution were determined in triplicates. The total number was determined from the countable dilutions as arithmetic means (±standard deviation).
Regarding the ClO 2 depletion experiment, analyses with E. coli and P. aeruginosa were performed twice each. In respective of the analyzed bacteria species, the arithmetic means for both test runs were calculated and finally, single approaches were statistically compared by the one-way analysis of variance (ANOVA) test using the Bonferroni post-test.

Impact of the Suspension Background for Bacterial Inactivation by ClO 2
The environmental background in which microorganisms are present may have an impact on cell stability and tolerance to ClO 2 . Therefore, E. coli, a well-characterized and fast-growing bacterium, was suspended in different media to investigate the biological stability on the one hand and the efficiency of the disinfection agent on the other hand. Based on a bacterial concentration of 10 5 cfu/mL and exposure to 0.4 mg/L ClO 2 , complete inactivation of all cultivable microorganisms was achieved, when E. coli was taken up in PBS ( Figure 1). In contrast, a successive decrease of living bacteria followed over time when suspended in physiologic saline. Counts dropped down gradually by 1.3 and 2.8 log 10levels within the first and after 2.5 min, respectively. After 5 min of ClO 2 incubation in saline, no growth could be observed. However, microorganisms solved in tap water showed high stability. In the first minute of incubation, the cell number demonstrated a 1.3 log 10 -reduction and remained stable during further incubation.
For the conduct of experiments, bacteria were suspended in tap water, because it is the natural environmental background for microorganisms in pipes and dental chair units.

Adaption of Microorganisms to Tap Water and Survival Monitoring
When microorganisms colonize as planktonic bacteria in the water supply system for a longer time period, their physiology may change as a reaction to stress conditions and starvation states. To simulate this condition, we adapted the microorganisms to tap water and monitored their survival over an extended time period using cultivation ( Figure 2). During the five weeks incubation period, the survival rate decreased by less than one log10level for all Gram-negative strains P. aeruginosa, E. coli, and the Legionella species. The survival curve for S. aureus decreased clearly over time. Figure 1. Impact of the medium of suspended microorganisms on ClO 2 treatment: E. coli cells, given as cfu/mL on a log 10 scale, were suspended in tap water (black columns), physiological saline (striped columns) and PBS (dotted columns) followed by treatment with 0.4 mg/L ClO 2 for several minutes as indicated. Additionally, the course of short-time inactivation in the respective suspension media is displayed as log 10 -reduction for tap water (solid line), saline (dashed line), and PBS (dotted line). The mean values are presented from one experiment; the cell counts of each dilution were determined in triplicates. The total number was determined from the countable dilutions as arithmetic means (±standard deviation).

Adaption of Microorganisms to Tap Water and Survival Monitoring
When microorganisms colonize as planktonic bacteria in the water supply system for a longer time period, their physiology may change as a reaction to stress conditions and starvation states. To simulate this condition, we adapted the microorganisms to tap water and monitored their survival over an extended time period using cultivation ( Figure 2). During the five weeks incubation period, the survival rate decreased by less than one log 10 -level for all Gram-negative strains P. aeruginosa, E. coli, and the Legionella species. The survival curve for S. aureus decreased clearly over time.

Kinetics of the Disinfection Efficacy of ClO2 on Waterborne Microorganisms
The inactivation kinetics of several bacterial species, taken directly from agar plates During this time interval, the cell count decreased by less than one log 10 stage for P. aeruginosa, E. coli, and the Legionella species. The survival rate of S. aureus decreased continuously over time. Plate counting was carried out in triplicates (±standard deviation) at time points as indicated.

Kinetics of the Disinfection Efficacy of ClO 2 on Waterborne Microorganisms
The inactivation kinetics of several bacterial species, taken directly from agar plates or after adaption to tap water, were analyzed over time when suspended in tap water and treated with 1.0 and 4.6 mg/L ClO 2 ( Figure 3). Higher concentrations of the disinfectant caused more intense inactivation of all microorganisms. The impact over time played a lesser role (Figure 3).
E. coli ( Figure 3A,B) and P. aeruginosa ( Figure 3C,D), both cultivated on plates overnight and adapted to tap water, respectively, with an initial cell density of 10 7 cfu/mL and treated with 1 mg/L ClO 2 , showed a low decrease within the first minutes following plateau formation (maximum decrease of 1.3 log 10 -levels). When adjusted to 10 6 cfu/mL, bacterial counts were reduced by maximal 3.5 log 10 -levels for the overnight culture and maximal 1.6 log 10 -levels for the water-adapted culture after 30 min. At the lower density of 10 5 cfu/mL and after 30 min, the counts were reduced by maximal 3.5 log 10 -levels for the overnight culture. In contrast, tap water-adapted strains were eradicated after this time. When ClO 2 was added with a high concentration of 4.6 mg/L, a maximum decrease of 5.5 log 10 -levels was achieved for a cell density of 10 7 cfu/mL with overnight plate-cultured cells. In contrast, tap water-adapted cells proved high tolerance to this treatment with a maximal decline of 1.9 log 10 -levels. At lower densities of 10 6 and 10 5 cfu/mL and after 5 and 2.5 min contact time, agar-cultivated E. coli failed to grow entirely, whereas P. aeruginosa showed no growth after 1 min. Tap water-adapted cells failed to grow under these conditions. Both serogroup representatives of L. pneumophila at 10 7 cfu/mL from different cultivations demonstrated a low reduction of cell numbers with plateau formation when treated with 1 mg/L ClO 2 ( Figure 3E,F,I). With an initial cell density of 10 6 cfu/mL, agar-cultivated L. pneumophila serogroup 1 was eradicated after 2.5 min but the tap water-adapted showed elimination only after 15 min of incubation. L. pneumophila serogroup 5 achieved disinfection standard with >3 log 10 -levels cell reduction. Using 4.6 mg/L ClO 2 , only tap water-adapted L. pneumophila serogroup 1 at 10 7 cfu/mL was able to survive. Interestingly, we could detect different growth behavior patterns with rapid and delayed growth and atypical morphology (Figure 4). Using 10 7 cfu/mL, the inactivation kinetics of the L. anisa microorganisms showed plateau formation and cell reduction among 0.6 and 2.0 log 10 -levels ( Figure 3G,H,J) after treatment with 1 mg/L ClO 2 . Lower densities of L. anisa from the plate failed in growth starting after 5 min and 2.5 min of incubation. On the other hand, L. anisa both tap wateradapted and as an environmental isolate showed growth and plateau formation at an initial cell density of 10 6 cfu/mL during the treatment with 1 mg/L ClO 2 . Growth failed at lower cell densities after 1 min for the tap water-adapted strain and after 15 min for the dental chair isolate. Treated with 4.6 mg/L ClO 2 , only the tap water-adapted L. anisa at a cell density of 10 7 cfu/mL was able to survive and demonstrated growth in association with a decrease of >4 log 10 -levels after 1 min.
The Gram-positive S. aureus demonstrated marking higher stability to ClO 2 -treatment compared with the Gram-negative microorganisms ( Figure 3K,L). Regardless of the cultivation manner, treatment with 1 mg/L ClO 2 caused a reduction of S. aureus by 0.1 and 1.3 log 10 -levels only that were independent of incubation time and the initial cell density. Similar to other findings in this work, S. aureus in high density demonstrated plateau formation when treated with 4.6 mg/L ClO 2 , whereas disinfection or elimination occurred only with a low initial cell density. The Gram-positive S. aureus demonstrated marking higher stability to ClO2treatment compared with the Gram-negative microorganisms ( Figure 3K,L). Regardless of the cultivation manner, treatment with 1 mg/L ClO2 caused a reduction of S. aureus by 0.1 and 1.3 log10-levels only that were independent of incubation time and the initial cell density. Similar to other findings in this work, S. aureus in high density demonstrated plateau formation when treated with 4.6 mg/L ClO2, whereas disinfection or elimination occurred only with a low initial cell density.

Analysis for ClO 2 Depletion on Selected Microorganisms
Our results give proof of a first fast and efficient microbiological inactivation phase and a subsequent long stagnation phase without a considerable decline of cultivable cells, whereby the effect was recognizable in all microorganisms examined. In addition to the different compositions of highly tolerant and less tolerant subspecies in the bacterial solutions, a loss of the effect of the disinfectant could also play a role. In the following experimental set-up, we investigated the influence of inactivated microorganisms on ClO 2 regarding a reduction in the efficiency of the disinfectant by the organic burden of microorganisms already present in tap water. Two different concentrations of heat-killed E. coli and P. aeruginosa cells were mixed with ClO 2 prior to the application of living cells ( Figure 5). Starting from 10 6 living cfu/mL, the log 10 -levels of reduction amounted to 4.7 and 3.7 for E. coli ( Figure 5A) and to 5.0 and 3.4 for P. aeruginosa ( Figure 5B) after pre-incubation with 10 3 and 10 6 heat-killed cfu/mL, respectively. Controls without an application of any heatkilled cells resulted in a 5.0 log 10 -levels reduction of E. coli cells whereas P. aeruginosa was fully eradicated. These results prove a ClO 2 -depletion effect during a short contact time.

Analysis for ClO2 Depletion on Selected Microorganisms
Our results give proof of a first fast and efficient microbiological inactivation phase and a subsequent long stagnation phase without a considerable decline of cultivable cells, whereby the effect was recognizable in all microorganisms examined. In addition to the different compositions of highly tolerant and less tolerant subspecies in the bacterial solutions, a loss of the effect of the disinfectant could also play a role. In the following experimental set-up, we investigated the influence of inactivated microorganisms on ClO2 regarding a reduction in the efficiency of the disinfectant by the organic burden of microorganisms already present in tap water. Two different concentrations of heat-killed E. coli and P. aeruginosa cells were mixed with ClO2 prior to the application of living cells ( Figure 5). Starting from 10 6 living cfu/mL, the log10-levels of reduction amounted to 4.7 and 3.7 for E. coli ( Figure 5A) and to 5.0 and 3.4 for P. aeruginosa ( Figure 5B) after preincubation with 10 3 and 10 6 heat-killed cfu/mL, respectively. Controls without an application of any heat-killed cells resulted in a 5.0 log10-levels reduction of E. coli cells whereas P. aeruginosa was fully eradicated. These results prove a ClO2-depletion effect during a short contact time.

Discussion
Disinfection with ClO2 has been proven and evaluated in a variety of applications [30,32,33,34,35,43,44,46,47,48,49]. In this study, we focused on disinfection conditions in an aqueous environment with a high hygienic background level namely DUWLs in dental practices. For this purpose, the effect of ClO2 was studied on various relevant bacteria in the planktonic state present in the aqueous environment of DUWLs considering various microbial aspects and factors influencing the effectiveness of ClO2 in aqueous solutions.
One approach was to examine the impact of the environmental background solution, in which bacteria exist. Our results indicate that the microbiological background environment had a decisive impact on the successful reduction of bacterial counts. The effect of disinfectants on microorganisms is often investigated in saline and PBS associated

Discussion
Disinfection with ClO 2 has been proven and evaluated in a variety of applications [30,[32][33][34][35]43,44,[46][47][48][49]. In this study, we focused on disinfection conditions in an aqueous environment with a high hygienic background level namely DUWLs in dental practices. For this purpose, the effect of ClO 2 was studied on various relevant bacteria in the planktonic state present in the aqueous environment of DUWLs considering various microbial aspects and factors influencing the effectiveness of ClO 2 in aqueous solutions.
One approach was to examine the impact of the environmental background solution, in which bacteria exist. Our results indicate that the microbiological background environment had a decisive impact on the successful reduction of bacterial counts. The effect of disinfectants on microorganisms is often investigated in saline and PBS associated with centrifugation steps [45,50]. Since the centrifugation step could be a stress factor for the bacteria leading to cell surface damage and affecting surface-sensitive properties causing significant reductions in viability [51,52], centrifugation was omitted in this study. ClO 2 has a high bactericidal effect on E. coli when present in PBS as it also causes a continuous decrease of CFU over time when suspended in saline. The lowest effect in bacterial reduction was determined for ClO 2 in tap water. Our findings indicate that either the effect of ClO 2 in tap water is reduced by naturally occurring organic substances of dead microorganisms or bacteria in aqueous environments have different physiological states. The presence of inorganic substances in tap water has an impact on ClO 2 depletion, too [53]. As tap water represents the natural background environment of microorganisms existing in pipes and DUWLs, we have chosen tap water as a suspension matrix for bacteria for a better comparison of laboratory conditions and conditions in DUWLs and pipes.
High bacterial contaminations with maximum values between 10 5 -10 6 cfu/mL were identified in probed DUWLs [1,2]. For this reason, we have analyzed the ClO 2 efficiency on high cell densities ranging from very high (10 7 cfu/mL) to quite low (10 4 cfu/mL). ClO 2 proved to be highly effective in cell reduction against numerous microorganisms. Regarding our experimental setup, the level of inactivation was higher when supplemented doses were increased. Further, the inactivation effect was also dependent on the density of tested bacteria, which is in contrast to former publications [50,54]. Bacteria in high cell numbers were reduced to a small extent only. The lower the initial cell count, the more cell numbers decreased. This phenomenon was confirmed using various bacteria species. Conspicuously, the main effect of ClO 2 occurred within the first five minutes of exposure to tested bacteria, which was indicated by low numbers of cultivable cells on agar plates. In the following course of our experiments, the cell counts hardly changed, resulting in a tailing [34,50,54]. This tailing effect at prolonged exposure was observed in various microorganisms studied in this work whereby a ClO 2 depletion effect should not be disregarded.
Treatment with 1 mg/L ClO 2 was not sufficient to eradicate E. coli, taken from a plate, with a density of 10 5 cfu/mL in total, which concurs with the result of another study [55]. P. aeruginosa from the plate demonstrated this stability as well. However, the Legionella species from the plate failed to survive under these conditions, thus after 15 min at the latest, no more growth was observed. These results imply species-related stability or susceptibility to the disinfection procedure. Differences in susceptibility between Gram-negative and Gram-positive microorganisms were observed. S. aureus showed higher stability to ClO 2 treatment than E. coli. Our results are consistent with another study, which explained the enhanced stability of the membrane structure and the mechanical stability of S. aureus [30]. Gram-negative and Gram-positive bacteria have different membrane structures. Grampositive bacteria have a thicker peptidoglycan layer than Gram-negative bacteria, which have a double membrane layer [56]. This different susceptibility was evidenced as more than twice the ClO 2 dose was needed for S. aureus to obtain the same log reduction as for P. aeruginosa [54].
Microorganisms may change their physiology when living in tap water for an extended time [57]. To simulate the conditions in DUWLs being contaminated with planktonic bacte-ria, the microorganisms were adapted to tap water and left to survive under a tap water stress situation for several weeks. For E. coli and P. aeruginosa, the cell count decreased by about 1 log 10 -levels after 11 weeks. Surprisingly, the Gram-positive S. aureus was found to survive and be cultivable in tap water over the whole incubation period, although its cell numbers decreased most strongly compared to other species. The long-term survival in tap water allowed a comparison of different physiological conditions between overnight culturing and tap water adaption regarding a change in susceptibility to ClO 2 treatment. The presence of planktonic cells in tap water for several weeks proves a long-term survival of microorganisms in tap water that might be associated with biofilm formation and therefore, underlines the importance of efficient disinfection of DUWLs. The microbiological stability patterns of ClO 2 treatment changed when adapted to tap water. High cell numbers were no longer reduced as much as with cultured cells during ClO 2 treatment but tailing occurred again after a few minutes of exposure. Tap water-adapted cells seemed to be more robust than cultured ones. Adaptation to tap water can cause a change in the cell, such as a less permeable membrane [57] and other physiological changes [58].
As it is shown, ClO 2 inactivation kinetics corresponded to a two-phase model with a rapid decay in the first minutes followed by plateau formation as described in other studies for bacteria and viruses [38,50,54,59,60]. The occurrence of this tailing phenomenon may be due to several factors at the biological and physical-chemical levels of the disinfectant and microorganisms. Bacteria and viruses in particularly high densities tend to aggregate, resulting in inadequate exposure [50,60]. However, tailing was observed, although samples were intensively vortexed to prevent clumping [50]. One reason for the two-phase model in viruses may be the presence of a ClO 2 -resistant subpopulation [59]. Another study does not exclude the existence of a resistant subpopulation but explains the two-phase model with the change in the properties of the virus during disinfection forming a protective layer on the virus capsid [60]. The presence of a resistant subpopulation could not be excluded as a cause for the two-phase model in bacteria either [50]. Another reason for the appearance of the biphasic kinetics could be a depletion of the disinfectant by organic substances reducing the bactericidal impact of ClO 2 in solutions [39,48,61]. The pre-load of the ClO 2 solution with low amounts of heat-killed microorganisms did not show any significant difference for the elimination of E. coli and P. aeruginosa, but high amounts of heat-killed microorganisms caused a significant reduction of the ClO 2 effect on the bacterial inactivation level. Even if the effect of ClO 2 was diminished, an effective part of the disinfectant remained active. The results do not exclude the presence of a ClO 2 tolerant and stable subpopulation as it was equally reported elsewhere [50,59]. Other reasons, such as the change in the characteristics of the bacteria could also be a reason. In this work, the bacterial density was shown to have a high influence on the sensitivity to ClO 2 , which seems to provide protective measures for bacteria against disinfection. Thus, the tailing effect cannot be attributed to a single factor but it is the result of various interacting aspects. To what extent the change in bacterial properties leads to a tailing effect should be subject to future studies.

Conclusions
ClO 2 is a very effective disinfectant against pathogenic and waterborne microorganisms occurring in water systems of dental practices and chair units, where Gram-negative bacteria are more susceptible than Gram-positive. However, in the case of massive contamination with high cell numbers, a substantial number of bacteria can withstand ClO 2 treatment. When disinfection measures are carried out it should be considered that microorganisms living in tap water might have more robust properties proved by high resistance to ClO 2 than overnight-cultured bacteria. Furthermore, the environmental background in which the bacteria are suspended might have an impact on the effectiveness of ClO 2 and the susceptibility of bacteria to the disinfectant. ClO 2 is an effective disinfectant to waterborne microorganisms in the planktonic state. Further studies aim to investigate conditions for efficient inactivation of sessile and in biofilms living microorganisms in the water systems.