Clinical parameters of the participants

Table 1 shows clinical parameter of the participants. The median age was 53.5years (Interquartile range (IQR) 45.868.0), and the median number of teeth was 25.5 (IQR 23.028.0). Of the total number of 153 periodontal sites, the median number of pockets less than 3mm was 123.50 (IQR 101.80147.00), the median number of pockets 4mm was 16 (IQR 3.7528.25), and the median number of pockets greater than 5mm was 1.5 (IQR 0.008.00).

In this study, salivary microbiota composition of 16 patients was studied based on the sequencing of the 16S rRNA gene. The samples provided 2,092,625 quality reads corresponding to the V3V4 regions of the 16S rRNA gene sequences, which were subsequently assigned to 308 species-level operational taxonomic units (OTUs) based on~97% sequence similarity. We investigated the changes of alpha-diversity due to exposure to NBW. In observed features, O2-NBW and HOCl-NBW tended to decrease alpha-diversity relative to the control; however, the differences were not significant (P=0.85). Shannon index also did not show significant differences (P=0.79) (Supplementary Fig.1). Figure1 shows the scatter diagram of beta-diversity based on Principal Coordinate Analysis (PCoA). In the Unweighted UniFraq distance, there was no significant difference between control and O2-NBW (P=0.168) or between control and HOCl-NBW (P=0.916) (Fig.1A). Similarly, there was no significant difference between control and O2-NBW or HOCl-NBW at the Weighted UniFrac distance (Fig.1B, P=0.521; P=0.828, respectively).

Beta-diversity of unweighted UniFraq distance (A) and weighted UniFraq distance (B). Colored dots indicate individual sample groups: Black: Control; red: O2-NBW; green: HOCl-NBW. Colored circles indicate groups exposed to NBW; Black: Control; Red: O2-NBW; Green: HOCl-NBW.

Supplementary Fig.2 shows the relative frequencies of the different salivary bacteria. The bacterial genera, based on detection in 1% or more of the total population of the salivary microbiome, were composed of 71 OTUs (frequency>0.001) (Supplementary Fig.2A). Specifically, 14 major genera including Prevotella, Streptococcus, Veillonella, Neisseria, Haemophilus, Leptotrichia, Porphyromonas, Fusobacterium, Rothia, Graulicatella, Alloprevotella, Campylobacter, Atopobium, Saccharibacteria (TM7) [G-1] were detected. In similar analyses, bacterial species that were detected in 1% and more of the salivary microbiome, constituted 166 OTUs (frequence>0.001) and included 25 major species, namely Prevotella melaninogenica, Haemophilus parainfluenzae, Streptococcus salivarius, Neisseria spp., Porphyromonas pasteri, Veillonella dispar, Streptococcus spp., Rothia mucilaginosa, Fusobacterium periodonticum, Veillonella atypica, Leptotrichia sp. HMT417, Prevotella pallens, Veillonella parvula, Veillonella rogosae, Prevotella spp., Granulicatella adiacens, Leptotrichia sp. HMT221, Streptococcus parasanguinis clade411, Neisseria subflava, Prevotella sp. HMT313, Prevotella salivae, Campylobacter concisus, Leptotrichia sp. HMT215, Saccharibacteria (TM7) [G-1] bacterium HMT352, Atopobium parvulum.

We next investigated the relative abundance in the control and exposed groups by bacterial genera. Repeat measures ANOVA for the 14 bacterial genera with detection rates greater than 1% showed that only the genus Porphyromonas had a significant association among the three groups. Multiple testing also revealed significant associations between control and O2-NBW (P=0.044) and between control and HOCl-NBW (P=0.007) in the genus Porphyromonas (Table 2). Also, we investigated the relative abundance in the control and exposed groups by bacterial species. Repeated measures analysis of variance for the 25 bacterial species with detection rates greater than 1% showed that only P. pasteri was significantly associated among the three groups (P=0.008). Multiple testing also showed a significant reduction (1.066%) in P. pasteri (P=0.028) between control and HOCl-NBW (Table 3).

Figure2 shows the results of the hierarchical cluster analysis by Wards method based on the results of the PCoA, which revealed two subclusters in terms of both Unweighted UniFraq distance (Fig.2A) and Weighted UniFraq distance (Fig.2B). In Fig.2A, CL1 and CL2 were formed, with CL1 having 10 subjects and CL2 having 6 subjects. In Fig.2B, CL3 and CL4 were formed, with CL3 comprising 9 subjects and CL4 7 subjects.

Results of cluster analysis of relative abundance in oral microbiome (N=16). (A) Unweighted cluster. (B) Weighted cluster. Stratified cluster analyses were performed according to the Ward method based on the results of PCoA. Numbers indicate sample ID. Clustering was performed using the Ward method with Euclidian Distance.

Supplementary Fig.3 shows the results of the principal coordinates analysis of the Unweighted UniFraq distance (A, B), and the Weighted UniFrac distance (C, D). Supplementary Fig.3A shows the results between Control and O2-NBW, and Supplementary Fig.3B shows the results between Control and HOCl-NBW. In Supplementary Fig.3A, there was no significant difference between the two groups in CL1 (P=0.536), while in CL2 there was a significant difference between the two groups (P=0.033). On the other hand, in Supplementary Fig.3B, there was no significant difference between CL1 and CL2.

In contrast, Supplementary Fig.3C,D show the results of the principal coordinates analysis of the Weighted UniFraq distance. Supplementary Fig.3C shows the results for control and O2-NBW, and Supplementary Fig.3D shows the results for control and HOCl-NBW. There were no significant differences between the two groups for both CL3 and CL4 in Supplementary Fig.3C,D.

We investigated the relative abundance of bacterial genera in CL1 and CL2 in the Unweighted cluster; no bacterial genera were significantly different in both CL1 and CL2. Also, in bacterial species, no bacterial species were found to have a significant difference between CL1 and CL2 (Supplementary Table 1).

On the other hand, in the relative abundance of bacterial genera in CL3 in the weighted clusters, the only significant reduction (1.186%) between Control and HOCl-NBW was observed in the genus Porphyromonas (Table 4). However, no bacterial genus showed significant differences in CL4. In the relative abundance of bacterial species, only P. pasteri showed significant reduction (0.921%) among the bacterial species in the CL3. On the other hand, no significant differences were found among the bacterial species in the CL4 (Table 5).

Tables 6 and 7 show the clinical parameters of the subjects according to cluster. Table 6 shows the Unweighted results; the categories that showed significant differences between CL1 and CL2 were the number of probing pocket depth (PD)s less than 3mm, the number of PDs 4mm, and the number of PDs greater than 5mm. No significant differences were found in the other categories. Table 7 shows the weighted results, where the category that showed a significant difference between CL3 and CL4 was the number of PDs of 4mm. No significant differences were found in the other categories.

Figure3 shows a scatter plot between PD values and difference in relative abundance in CL3 (N=9), the cluster where a significant association between Control and HOCl-NBW was observed in Tables 4 and 5. As shown in Fig.3B, t=2.45 at PD=4mm, indicating that the higher the number of PD=4mm, the higher the effect of HOCl-NBW exposure on P. pasteri. On the other hand, no significant association was found for PD=3mm or less and PD=5mm or more. These results suggest that relative abundance of P. pasteri is associated with clinical signs of early stage of periodontitis.

Scatter plots and correlation coefficient tests in CL3 group (N=9). Spearmans rank correlation coefficient. Alternative hypothesis: true is greater than 0. The significance level was set at alpha=0.05. (A) Spearmans rank correlation coefficient0.0667 (P=0.58). (B) Spearmans rank correlation coefficient 0.653 (P=0.028). (C) Spearmans rank correlation coefficient 0.131(P=0.37).

Read the original post:
The effects of exposure to O2- and HOCl-nanobubble water on ... - Nature.com