An oxygen-releasing mouthwash shows species-dependent antibacterial and antibiofilm activity against periodontal pathogens

Abstract

Objectives

This study evaluated whether an oxygen-releasing mouthwash formulation (O₂MW) produces species-dependent antibacterial, antibiofilm, and gene expression responses in Aggregatibacter actinomycetemcomitans and Prevotella intermedia compared with 0.12% chlorhexidine under planktonic and biofilm conditions in vitro.

Materials and methods

Minimum inhibitory concentrations (MICs) were determined using a resazurin-based broth microdilution, and minimum bactericidal concentration (MBC) by agar subculture. Agar diffusion assessed growth inhibition. Biofilm biomass was quantified using crystal violet staining, while pre-formed biofilm structure and membrane integrity were examined by confocal laser scanning microscopy (CLSM). Quantitative PCR assessed adhesion-, toxin-, oxidative-stress-, and iron-associated genes. Molecular docking used sodium perborate as a representative oxygen-releasing constituent.

Results

O₂MW MICs were 3.13% for A. actinomycetemcomitans and 50% for P. intermedia. MBC was detected only for A. actinomycetemcomitans. Agar diffusion showed inhibition zones from 0.78% O₂MW for A. actinomycetemcomitans, whereas P. intermedia showed zones only at 50% and 100% O₂MW. O₂MW reduced biofilm biomass, with minimum biofilm inhibitory concentration (MBIC) values of 1.56% and 25%, respectively. CLSM showed reduced biomass and thickness at higher concentrations. Sub-MIC O₂MW increased expression of adhesion- and virulence-associated genes, particularly ltxA in A. actinomycetemcomitans and oxidative-stress and iron-associated genes, including ahpF, ftnA, and feoB in P. intermedia. Docking suggested pathway-level involvement only.

Conclusions

O₂MW showed species-dependent antibacterial and antibiofilm activity, with greater effects at lower concentrations against A. actinomycetemcomitans. The findings support further preclinical and host-cell evaluation before clinical translation.

Clinical relevance

O₂MW may have potential as an adjunct for periodontal biofilm control, but pathogen-specific responses warrant further evaluation.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1007/s00784-026-07203-0.

Introduction

Periodontitis is a dysbiotic, biofilm-driven inflammatory disease in which subgingival microbial communities interact with the host immune response and contribute to periodontal tissue destruction [1]. Mechanical plaque control and periodontal debridement remain central to treatment, with chemical adjuncts used where access is limited [2, 3]. Chlorhexidine (CHX) is a reference oral antiseptic with broad antimicrobial activity but is limited by tooth staining, taste disturbance, mucosal irritation, and concerns with repeated or prolonged exposure [3, 4]. These limitations support the evaluation of adjunctive formulations with different modes of action.

Oxygen-releasing oral-care formulations have gained attention because periodontal pockets and mature oral biofilms contain low-oxygen microenvironments that can support anaerobic periodontal pathogens [5–7]. O₂MW is an oxygen-releasing mouthwash formulation containing sodium perborate (SP), an oral-care ingredient used for biofilm control that can contribute to peroxide-mediated oxidative activity in aqueous environments [8, 9]. However, the response to oxygen-releasing formulations may differ among periodontal pathogens because susceptibility is influenced by oxygen tolerance, surface-associated virulence factors, biofilm phenotype, and oxidative-stress adaptation [10, 11].

Aggregatibacter actinomycetemcomitans and Prevotella intermedia were selected as periodontal pathogens with contrasting biological characteristics. A. actinomycetemcomitans is a facultative Gram-negative pathobiont with virulence mechanisms involving adhesion, immune evasion, and toxin production, including leukotoxin and cytolethal distending toxin [10, 12]. In contrast, P. intermedia is a black-pigmented anaerobe associated with haem acquisition, biofilm formation, and oxidative-stress tolerance [11, 13]. These differences provide a useful model for determining whether an oxygen-releasing formulation produces uniform or species-dependent antimicrobial responses.

Although oxygen-releasing formulations show antimicrobial activity against oral bacteria, it remains unclear whether their effects differ among periodontal pathogens with distinct physiology, oxygen tolerance, virulence mechanisms, and biofilm characteristics. Our previous work showed that O₂MW reduced P. gingivalis biofilm formation and suppressed selected adhesion-associated genes [14]. Therefore, this study compared the antibacterial and antibiofilm effects of O₂MW against A. actinomycetemcomitans and P. intermedia, using CHX as a reference comparator, and assessed selected virulence-, adhesion-, oxidative-stress-, and iron-associated genes with exploratory molecular docking of sodium perborate.

Materials and methods

Bacterial strains and culture conditions

A. actinomycetemcomitans ATCC 29522 and P. intermedia ATCC 25611 were obtained as KWIK-STIK™ preparations (Microbiologics, St. Cloud, MN, USA) and activated according to the manufacturer’s instructions. A. actinomycetemcomitans was cultured on Brain Heart Infusion (BHI) agar/broth (Oxoid Ltd., Basingstoke, Hampshire, UK) at 37 °C in 5% CO₂ for 48 h. P. intermedia was cultured in supplemented tryptic soy broth/agar containing hemin and vitamin K (Watin-Biolife, Riyadh, Saudi Arabia), and incubated anaerobically at 37 °C for 48–72 h using an anaerobic jar system and AnaeroGen™ gas-generating sachets (Thermo Scientific™, Oxoid Ltd., Basingstoke, Hampshire, UK). For anaerobic assays, fresh or pre-reduced medium was used when required.

Bacterial suspensions were prepared from fresh cultures grown to the mid-logarithmic phase and standardized spectrophotometrically at OD₆₀₀ before each experiment.

Test solutions

Blue^®m mouthwash/O₂MW (BlueM Europe, Zwolle, The Netherlands) was used as the oxygen-releasing formulation and diluted in organism-specific medium immediately before use.

Chlorhexidine mouthwash (0.20%, Parodontax^® Extra, Omega Pharma Manufacturing GmbH, Herrenberg, Germany) was diluted to 0.12% and used as a positive antimicrobial comparator. Untreated bacterial cultures served as growth controls, and uninoculated medium served as the sterility control. Concentration ranges reflected assay purpose and organism-specific susceptibility: broad dilutions for MIC determination, sub-inhibitory to higher concentrations for antibacterial and biofilm assays, and MIC-based concentrations for CLSM and qPCR analyses.

Minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) determination

MIC was determined by resazurin-based broth microdilution [15, 16]. Two-fold O₂MW serial dilutions (0.048-100%) were prepared in 96-well round-bottom plates (Greiner Bio-One GmbH, Frickenhausen, Germany), using BHI broth for A. actinomycetemcomitans and supplemented tryptic soy broth for P. intermedia.

Standardized bacterial suspensions were added and incubated under organism-specific conditions. After incubation, 0.015% resazurin solution was added, and plates were further incubated. MIC was defined as the lowest O₂MW concentration that prevented detectable resazurin color change compared with the untreated growth control [15].

For MBC determination, aliquots from wells at the MIC and higher concentrations were plated onto organism-specific agar plates and incubated under the corresponding conditions. MBC was defined as the lowest concentration showing no visible colony growth [16].

Agar diffusion assay

Standardized bacterial suspensions were spread onto BHI agar for A. actinomycetemcomitans or supplemented TSA containing hemin and vitamin K for P. intermedia. Agar diffusion was used as an in vitro inhibition-zone assay based on general disk-diffusion principles [17, 18]; no clinical breakpoints or susceptibility categories were applied.

Sterile 6-mm Whatman^® paper discs (Sigma-Aldrich, Darmstadt, Germany) were loaded with 10 µL test solution. The tested concentrations were 0.78–6.25% for A. actinomycetemcomitans and 12.5–100% for P. intermedia. CHX 0.12% served as the positive comparator, and medium-loaded discs served as growth controls. After incubation, inhibition zones were measured in millimeters; absence of a visible inhibition halo was recorded as 0 mm.

Biofilm biomass assay

For crystal violet microtiter plate assay, bacterial suspensions were adjusted to OD₆₀₀ = 0.08–0.10 and inoculated into 96-well U-bottom plates [19, 20].

Tested O₂MW concentrations were 0.39–12.5% for A. actinomycetemcomitans and 12.5–100% for P. intermedia. Untreated cultures served as biofilm growth controls, and CHX 0.12% was included as the positive comparator.

Plates were incubated for 72 h under organism-specific conditions to allow static biofilm formation [21–23]. After incubation, planktonic cells were removed, wells were gently washed with sterile distilled water, dried for 20 min at room temperature, stained with 0.5% crystal violet (Sigma-Aldrich, St. Louis, MO, USA) for 10 min, washed, and solubilized with absolute ethanol. Absorbance was measured at 490 nm using a Synergy HT microplate reader (BioTek Instruments, Winooski, VT, USA).

The minimum biofilm inhibitory concentration (MBIC) was defined as the lowest O₂MW concentration that produced ≥ 50% reduction in biofilm biomass compared with the untreated control [24, 25].

Confocal laser scanning microscopy (CLSM) biofilm analysis

Biofilms were allowed to pre-form on sterile coverslips under organism-specific growth conditions to measure biofilm structure and membrane integrity [21, 22].

After biofilm formation, coverslips were transferred to fresh culture plates and exposed to O₂MW or CHX for 1 min, consistent with the recommended mouth-rinsing duration [26]. As CLSM evaluated pre-formed biofilms, MIC-based concentrations were selected to provide standardized exposure levels across species. Tested O₂MW concentrations corresponded to ¼ MIC, ½ MIC, MIC, and 2× MIC: 0.78%, 1.56%, 3.13%, and 6.25% for A. actinomycetemcomitans, and 12.5%, 25%, 50%, and 100% for P. intermedia. Untreated biofilms served as controls.

Coverslips were gently washed with phosphate-buffered saline (PBS) and stained using LIVE/DEAD™ BacLight™ stain (Molecular Probes, Eugene, OR, USA) to assess membrane-integrity patterns. SYTO 9 fluorescence was excited at 488 nm and detected in the green channel, while propidium iodide fluorescence was excited at 568 nm and detected in the red channel.

Z-stacks were acquired using a Nikon C2 confocal microscope (Nikon Instruments Inc., Tokyo, Japan), and biofilm biomass and thickness were quantified separately using COMSTAT2 (version 2.1) [27, 28].

Quantitative real-time PCR (qPCR)

For sub-inhibitory qPCR analysis on selected genes, A. actinomycetemcomitans was treated with 0.78% and 1.56% O₂MW, corresponding to ¼ MIC and ½ MIC, and the selected targets were cdtA, ltxA, aae, emaA, omp100/apiA, and omp64. P. intermedia was treated with 12.5% and 25% O₂MW, corresponding to ¼ MIC and ½ MIC, and the selected targets were oxyR, ahpC, ahpF, ftnA, feoB, and ompA. Untreated cultures served as time-matched controls, and CHX 0.12% served as an antimicrobial-stress comparator.

Bacterial cultures were exposed for 30 min and 2 h to assess early and short-term transcriptional responses; these extended exposures were mechanistic rather than clinically simulated rinsing [11, 29]. After exposure, cells were collected by centrifugation at 10,053 × g for 5 min. Total RNA was extracted using HiGene™ Total RNA Prep kit (BioFACT Co. Ltd., Daejeon, Korea) and RNA concentration and purity were measured using BioSpectrometer^® basic (Eppendorf, Hamburg, Germany). cDNA was synthesized using the RT Ace First Strand cDNA synthesis kit (Haven Scientific, Thuwal, Saudi Arabia).

qPCR was performed using EverGreen Universal Real-time PCR Master Mix on an ABI 7500 Real-Time PCR System (Applied Biosystems, Waltham, MA, USA). Primer sequences are provided in Supplementary Table S1. Gene expression was normalized to 16S rRNA as the reference gene, and relative fold changes were calculated using the 2^−ΔΔCt method [34]. Experiments were performed using three independent biological replicates, and each qPCR reaction was analyzed in technical triplicate.

Molecular docking

Molecular docking was performed as an exploratory and hypothesis-generating analysis of potential interactions between sodium perborate, a representative oxygen-releasing constituent of O₂MW and selected bacterial protein targets [9].

Target proteins included Aae, LtxA, CdtA, Omp64, EmaA, and Omp100/ApiA for A. actinomycetemcomitans, and AhpC, AhpF, FtnA, OmpA, FeoB, and OxyR for P. intermedia.

Protein sequences were retrieved from the NCBI Protein database using genome assemblies ASM2355964v1 and ASM195395v1; and predicted structures were obtained from AlphaFold [35, 36]; accession details are provided in Supplementary Table S2.

Sodium perborate ligand structure was generated using ChemAxon Marvin JS (version 22.11.1) and Avogadro software (version 1.2.0). Ligand and receptor PDBQT files were prepared using AutoDock Tools (version 1.5.7) [37], and blind docking was performed using GNINA (version 1.3.2), with whole-receptor search space, 4 Å padding, exhaustiveness of 64, and 20 output poses per target [38].

Protein-ligand complexes were visualized using PyMOL (version 3.0.3) [39] and ChimeraX (version 1.7.1) [40], and interaction maps were generated using LigPlot+ (version 2.3.1) [41]. Docking results were reported as predicted binding energies and interpreted as exploratory only.

Statistical analysis

Data were analyzed using GraphPad Prism (version 6.00 for Windows, La Jolla, CA, USA). Agar diffusion was performed with six replicates per group (n = 6); other assays were performed in three independent experiments, each in triplicate. The triplicate readings from each independent experiment were averaged before statistical analysis. One-way ANOVA followed by Dunnett’s multiple-comparisons test was performed separately for comparisons with the untreated control and 0.12% CHX reference group. Two-way ANOVA followed by Tukey’s multiple-comparisons test was used when two factors were analyzed. Statistical significance was set at p < 0.05.

Results

Antibacterial activity

The resazurin-based broth microdilution assay showed organism-specific inhibitory responses to O₂MW (Fig. 1). The MIC was 3.13% for A. actinomycetemcomitans and 50% for P. intermedia. Growth control wells showed resazurin color change, whereas broth controls and 0.12% CHX controls showed no color change. The MBC was 3.13% for A. actinomycetemcomitans. For P. intermedia, no MBC was detected within the tested concentration range.

Fig. 1.

Representative resazurin-based broth microdilution assay showing O₂MW serial dilutions against (A) A. actinomycetemcomitans and (B) P. intermedia. Bacterial growth control, broth control, and 0.12% CHX were included as assay controls. The MIC was recorded as the lowest O₂MW concentration showing no resazurin color change after incubation. The MIC was determined as 3.13% v/v for A. actinomycetemcomitans and 50% v/v for P. intermedia, as indicated by the boxed wells

The in vitro agar diffusion assay showed measurable inhibition zones for selected O₂MW concentrations against both bacterial species (Fig. 2). For A. actinomycetemcomitans, no inhibition halo was detected in the untreated control. O₂MW produced measurable halos from 0.78% to 6.25%, increasing from 1.89 ± 0.56 at 0.78% to 4.51 ± 0.87 at 6.25% (Fig. 2A). CHX 0.12% produced the largest mean inhibition-halo width, measuring 10.58 ± 0.79 mm. O₂MW concentrations from 0.78 to 6.25% were significantly different from the untreated control and from CHX, while CHX was also significantly different from the untreated control.

Fig. 2.

In vitro agar diffusion inhibition zones of O₂MW against A. actinomycetemcomitans and P. intermedia. (A) Representative agar diffusion imagesRepresentative CLSM images of showing inhibition halos after exposure to O₂MW concentrations and 0.12% CHX. The upper row represents A. actinomycetemcomitans, and the lower row represents P. intermedia. (B) Quantitative inhibition-halo width for A. actinomycetemcomitans. (C) Quantitative inhibition-halo width for P. intermedia. Data are presented as box-and-whisker plots with individual data points (n = 6). After incubation, inhibition-halo widths were measured from the edge of the disc to the outer edge of the clear zone; absence of a visible halo was recorded as 0 mm. * p < 0.05 compared with the untreated control; # p < 0.05 compared with the 0.12% CHX

For P. intermedia, no inhibition zone was observed in the untreated control or at 12.5% and 25% O₂MW. Measurable halos were detected only at 50% and 100% O₂MW, at 2.36 ± 0.08 and 2.98 ± 0.08 mm, respectively. CHX produced a comparable halo of 2.78 ± 0.15 mm. Significant differences versus the untreated control were observed at 50% and 100% O₂MW, while the untreated control, 12.5%, and 25% O₂MW were statistically different from CHX.

Biofilm biomass inhibition

Crystal violet staining showed that O₂MW reduced biofilm biomass in both bacterial species (Fig. 3). For A. actinomycetemcomitans, biofilm formation was reduced from 0.39% O₂MW, with greater reduction at 0.78% (Fig. 3A). At 1.56% O₂MW, biofilm formation was reduced to 22.15 ± 3.30% of the untreated control and remained low at higher concentrations.

Fig. 3.

Crystal violet biofilm biomass assay showing the effect of O₂MW on biofilm formation by (A) A. actinomycetemcomitans and (B) P. intermedia. Biofilm formation was expressed as percentage relative to the untreated control. O₂MW reduced biofilm biomass in both species, with an MBIC of 1.56% v/v for A. actinomycetemcomitans and 25% v/v for P. intermedia. CHX was included as the reference antiseptic comparator. Data are presented as box-and-whisker plots with individual data points. * indicates significant difference from the untreated control; # indicates significant difference from 0.12% CHX; p < 0.05

For P. intermedia, O₂MW reduced biofilm formation in a concentration-related manner. Biofilm was reduced at higher concentrations, reaching 42.01 ± 7.20% at 25% O₂MW. CHX 0.12% also reduced biofilm formation in both species. Based on these, the MBIC was 1.56% for A. actinomycetemcomitans and 25% for P. intermedia.

CLSM pre-formed biofilm analysis

CLSM imaging showed treatment-associated changes in biofilm structure and membrane-integrity staining. Untreated A. actinomycetemcomitans biofilms showed dense biofilm architecture with dominant green fluorescence, whereas CHX-treated biofilms showed increased red/orange fluorescence and reduced green signal. O₂MW treatment produced a concentration-related reduction in green fluorescence and increased red/orange staining, most evident at higher tested concentrations, particularly at 6.25% (Fig. 4A).

Fig. 4.

(A) Representative CLSM images of A. actinomycetemcomitans biofilms after treatment with O₂MW or 0.12% CHX. For each group, three-dimensional biofilm reconstructions are shown above corresponding two-dimensional views. Green fluorescence indicates cells with intact membranes, while red/orange fluorescence indicates membrane-compromised cells. O₂MW treatment reduced biofilm structure and increased red/orange fluorescence in a concentration-dependent pattern. Scale bars = 100 μm. Quantitative analysis of (B) biofilm biomass and (C) biofilm thickness showed reductions after O₂MW and CHX treatment. Data are presented as box-and-whisker plots with individual data points. * indicates significant difference from the untreated control; φ indicates significant difference from 0.12% CHX; p < 0.05

Quantitative CLSM confirmed reduced A. actinomycetemcomitans biofilm biomass and thickness. Untreated biofilms had the highest biomass (24.47 ± 2.88 μm³/µm²), while O₂MW and CHX reduced biomass. Biofilm thickness decreased with increasing O₂MW concentration, most at 6.25% (39.08 ± 5.07 μm); CHX also reduced thickness.

For P. intermedia, untreated biofilms also showed dense green fluorescence, while CHX and O₂MW increased red/orange signal and reduced green fluorescence (Fig. 5A). The strongest red/orange pattern among O₂MW-treated groups was observed at 100%. Quantitative analysis showed that untreated biofilms had the highest biomass, 33.80 ± 6.04 μm³/µm², while O₂MW and CHX reduced biomass to 11.56 ± 2.68 to 23.02 ± 5.41 μm³/µm². Thickness followed a similar pattern, with untreated biofilms reaching 51.33 ± 9.42 μm and O₂MW-treated groups ranging from 23.58 ± 3.80 to 38.17 ± 5.36 μm.

Fig. 5.

(A) Representative CLSM images of P. intermedia biofilms after treatment with O₂MW or 0.12% CHX. Three-dimensional reconstructions and corresponding two-dimensional views are shown for each group. Green fluorescence indicates cells with intact membranes, while red/orange fluorescence indicates membrane-compromised cells. O₂MW reduced biofilm structure and increased red/orange fluorescence, particularly at higher concentrations. Scale bars = 100 μm. Quantitative analysis showed reduced (B) biofilm biomass and (C) thickness after O₂MW and CHX treatment. Data are presented as box-and-whisker plots with individual data points. * indicates significant difference from the untreated control; φ indicates significant difference from 0.12% CHX; p < 0.05

qPCR analysis

Relative expression of cdtA, ltxA, aae, emaA, omp100/apiA, and omp64 in A. actinomycetemcomitans was assessed after sub-inhibitory O₂MW exposure (Fig. 6A; Supplementary Fig. S1). At 30 min, 0.78% O₂MW increased expression of all tested genes, with the strongest induction observed for ltxA. Increased expression was also observed for toxin-, adhesion-, and outer-membrane-associated genes including cdtA, aae, emaA, omp100/apiA, and omp64. At 1.56% O₂MW, expression values were generally lower than those observed at 0.78%, with most genes closer to the untreated control range, although ltxA remained elevated. At 2 h, the overall response was lower than the 30-min response, while ltxA remained above control levels in both O₂MW-treated groups. CHX differed from O₂MW, with reduced expression of several genes, including cdtA, aae, emaA, and omp100/apiA, while ltxA remained above the untreated control but below the 0.78% O₂MW group.

Fig. 6.

qPCR heatmap of gene-expression changes after O₂MW and CHX exposure. (A) Relative expression of cdtA, ltxA, aae, emaA, omp100/apiA, and omp64 in A. actinomycetemcomitans after 30 min and 2 h exposure to sub-MIC O₂MW concentrations or 0.12% CHX. (B) Relative expression of oxyR, ahpC, ahpF, ftnA, feoB, and ompA in P. intermedia after 30 min and 2 h exposure to sub-MIC O₂MW concentrations or 0.12% CHX. Values represent relative fold change calculated using the 2^−ΔΔCt method, with time-matched untreated controls normalized to 1. Color intensity indicates fold-change magnitude. O₂MW, oxygen-releasing mouthwash; CHX, chlorhexidine

Relative expression of oxyR, ahpC, ahpF, ftnA, feoB, and ompA was assessed in P. intermedia following O₂MW and CHX treatment (Fig. 6B; Supplementary Fig. S2). At 30 min, 12.5% O₂MW increased expression of several oxidative-stress, iron-homeostasis, and outer-membrane-associated genes, including ahpC, ftnA, feoB, and ompA, whereas oxyR remained below the untreated control. The 25% O₂MW group showed lower expression than 12.5% for several genes, although ftnA remained elevated. At 2 h, 12.5% O₂MW increased all tested genes, while 25% O₂MW showed gene-dependent expression, with sustained elevation of ahpF, ftnA, and feoB. CHX differed from O₂MW, with particularly high expression of feoB and increased expression of selected oxidative-stress and iron-homeostasis genes.

Docking analysis

Molecular docking was performed as an exploratory analysis of sodium perborate interactions with selected protein targets (Supplementary Table S3; Supplementary Fig. S3). For A. actinomycetemcomitans, mean docking scores ranged from − 3.83 to − 6.74 kcal/mol. Omp64 showed the most negative mean score (− 6.74 ± 0.03 kcal/mol), followed by Aae, LtxA, CdtA, EmaA, and Omp100.

For P. intermedia, mean docking scores ranged from − 4.70 to − 6.71 kcal/mol. OxyR showed the most negative mean docking score (− 6.71 ± 0.02 kcal/mol), followed by FeoB, AhpF, OmpA, FtnA, and AhpC.

Docking visualizations showed predicted sodium perborate interactions with selected protein targets, with the most favorable interaction scores observed for Omp64 in A. actinomycetemcomitans and OxyR in P. intermedia.

Discussion

This study evaluated O₂MW against A. actinomycetemcomitans and P. intermedia. O₂MW showed differential activity, with A. actinomycetemcomitans showing lower MIC and MBIC values than P. intermedia. In agar diffusion testing, inhibition zones appeared at lower O₂MW concentrations for A. actinomycetemcomitans, whereas P. intermedia required higher concentrations. Although CHX produced larger inhibition halos than O₂MW, agar diffusion does not measure antimicrobial potency alone. Inhibition halos depend on antimicrobial activity as well as solubility, molecular diffusion through agar, release kinetics from the disc, and stability of active components during radial diffusion [18, 42, 43]. Therefore, the smaller O₂MW halos may reflect restricted diffusion or localized active-oxygen release rather than lower intrinsic antibacterial activity. This distinction is important because CHX and O₂MW differ mechanistically; CHX is a cationic bisbiguanide that binds to bacterial surfaces and disrupts membrane integrity, whereas O₂MW is an oxygen-releasing formulation [4, 44].

Fernandez y Mostajo et al. reported a different selectivity pattern for Ardox-X^® technology, an oxygenating sodium-perborate system. At the highest AX concentration, inhibition zones were 9 ± 1 mm for P. intermedia and 2 ± 1 mm for A. actinomycetemcomitans. Similarly, MIC/MBC values were lower for P. intermedia than for A. actinomycetemcomitans (80/80 mg/L SP versus 159/319 mg/L SP, respectively) [45]. In contrast, the present O₂MW formulation showed lower MIC and MBIC values for A. actinomycetemcomitans than for P. intermedia. This difference in findings may reflect formulation-specific selectivity, strain background, growth conditions, exposure format, and biofilm model rather than a uniform sodium-perborate class effect. Similarly, a hydro-carbon-oxo-borate compound showed antibiofilm activity in a multispecies subgingival biofilm model, although CHX produced stronger reductions for several periodontal species, including P. intermedia [46].

The species-dependent responses are likely multifactorial, reflecting strain-specific physiology, growth requirements, biofilm characteristics, formulation-dependent exposure, and oxidative-stress handling. The contrasting selectivity reported for another sodium-perborate system [45] suggests that formulation, experimental context, and oxidative-stress responses may also influence activity. Hydrogen peroxide has been shown to inhibit A. actinomycetemcomitans at relatively low concentrations [8, 9, 29], whereas haem-associated pigment in P. intermedia may contribute to protection against peroxide [47, 48]. MBC was detected only for A. actinomycetemcomitans within the tested concentration range.

For A. actinomycetemcomitans, biofilm biomass was reduced across the tested concentrations, with an MBIC of 1.56%. Because this was below the planktonic MIC, the antibiofilm effect of O₂MW may involve interference with surface attachment, colonization, or matrix accumulation rather than planktonic growth inhibition alone. This is supported by the role of surface-associated adhesins, including Aae, EmaA, and ApiA/Omp100 in A. actinomycetemcomitans attachment and biofilm formation [23, 49].

For P. intermedia, O₂MW reduced biofilm biomass in a concentration-related manner, with an MBIC of 25%. Although below the planktonic MIC, this was substantially higher than the MBIC for A. actinomycetemcomitans and may reflect biofilm-associated tolerance to oxidative stress, including haem-associated surface protection and iron(III) protoporphyrin IX pigment formation, which may reduce peroxide susceptibility [13, 50]. Because crystal violet measures total attached biomass, including bacterial cells and extracellular matrix, these MBIC values should be interpreted as reduced biofilm accumulation rather than direct bacterial killing [24, 25]. Clinically, the stronger low-concentration in vitro effect against A. actinomycetemcomitans may be relevant because this organism contributes to periodontal colonization through adhesion- and biofilm-associated mechanisms, while its pathogenicity is also linked to major virulence factors such as leukotoxin and cytolethal distending toxin [10, 12, 23, 49]. Sub-MIC reduction of A. actinomycetemcomitans biofilm biomass may therefore suggest potential value in limiting surface colonization or early biofilm accumulation. However, these findings were obtained using single-species in vitro models and do not establish clinical efficacy. The higher concentrations required for P. intermedia further suggest that O₂MW warrants evaluation as a potential adjunct to mechanical plaque control and biofilm disruption, rather than as a standalone strategy for suppressing mature anaerobic biofilms [2, 51].

CLSM analysis showed treatment-associated changes in pre-formed biofilm structure and membrane-integrity staining. In both organisms, O₂MW reduced biofilm biomass and thickness and increased red/orange fluorescence, with stronger effects observed at higher concentrations. These findings suggest that O₂MW affected both the quantity and architecture of pre-formed biofilms. The LIVE/DEAD signal was interpreted as a membrane-integrity endpoint and not as evidence of bacterial eradication or culturability. The 1-min CLSM exposure reflects clinical mouth-rinsing duration, whereas the longer qPCR exposures were mechanistic rather than clinically representative.

Our previous O₂MW study showed reduced P. gingivalis biofilm formation with suppression of fimA and hagA expression [14]. In contrast, this study showed distinct organism-specific responses in A. actinomycetemcomitans and P. intermedia, including greater low-concentration activity against A. actinomycetemcomitans and stress-associated transcriptional responses in both organisms under sub-inhibitory exposure.

In A. actinomycetemcomitans, 0.78% O₂MW increased ltxA, cdtA, aae, emaA, omp100/apiA, and omp64 expression, with the strongest response observed for ltxA. The increased ltxA transcription should not be interpreted as increased leukotoxin production or activity, which was not assessed in this study. Because crystal violet and CLSM showed reduced biofilm biomass and structure, this increase should not be interpreted as enhanced biofilm activity. Rather, it may reflect a sublethal oxidative-stress response among residual cells exposed to low-concentration O₂MW. This is based on the roles of the induced genes where aae, emaA, and omp100/apiA contribute to epithelial adhesion, serum resistance, and biofilm formation, whereas ltxA and cdtA encode major toxin-associated factors [10, 12, 49]. Therefore, O₂MW may have reduced biofilm accumulation while remaining cells increased colonization- and virulence-associated gene expression. At 1.56% O₂MW and at 2 h, expression was lower for most genes, although ltxA remained elevated. This is consistent with reports that sub-inhibitory antimicrobial exposure can alter bacterial gene expression, while stronger exposure may reduce the proportion of metabolically active or responsive cells [52, 53]. CHX differed from O₂MW by reducing several colonization-associated genes, while ltxA transcription remained stress-responsive.

The docking results for A. actinomycetemcomitans were consistent with surface-colonization involvement but remain exploratory. The most favorable predicted scores involved Omp64, Aae, and LtxA, which are associated with surface structure, adhesion, or virulence. Together with the qPCR data, these findings suggest that surface-colonization and virulence-associated pathways may be involved in the response of A. actinomycetemcomitans to O₂MW. However, docking does not prove direct binding, inhibition, or regulation by sodium perborate.

In P. intermedia, O₂MW mainly affected oxidative-stress and iron-homeostasis genes. At 30 min, 12.5% O₂MW increased ahpC, ahpF, ftnA, feoB, and ompA, whereas oxyR was not increased. By 2 h, 12.5% O₂MW increased all selected genes, including oxyR, consistent with a time-dependent redox-stress response. This is because OxyR regulates oxidative-stress survival in P. intermedia, and OxyR disruption affects survival under oxygen and peroxide stress together with genes such as ahpC, ahpF, and ftnA [11]. Increased ahpC and ahpF suggest peroxide-detoxification activity, while increased ftnA and feoB suggest involvement of iron storage and uptake. This is important because free iron can promote radical-mediated damage, whereas ferritin-like proteins can reduce oxidative stress by limiting available iron [54, 55]. The 25% O₂MW group showed a less broad response than 12.5% O₂MW, although ahpF, ftnA, and feoB remained elevated, suggesting that peroxide-detoxification and iron-homeostasis pathways remained active under stronger sub-inhibitory exposure. Accordingly, the higher MIC and MBIC required for P. intermedia may reflect lower direct susceptibility to O₂MW and activation of redox and iron-protective pathways in surviving cells.

CHX produced a stronger iron-stress-associated response in P. intermedia, particularly through marked induction of feoB and ftnA, with increased ahpF also observed. This suggests that CHX and O₂MW both triggered stress-associated transcriptional changes in P. intermedia, but the response to CHX was more dominated by iron-homeostasis and oxidative-stress-associated genes.

The docking findings for P. intermedia supported this interpretation but remain exploratory. Favorable predicted scores involved OxyR, FeoB, AhpF, and OmpA, corresponding to oxidative-stress, iron-homeostasis, and membrane-associated pathways. However, oxyR induction is more likely to reflect redox sensing than direct ligand binding. Therefore, the qPCR and docking data should be interpreted as supportive evidence of pathway involvement, not proof of direct molecular targets.

The effects of O₂MW should also be considered in the context of the broader oral microbiome. Oxygenating formulations have shown selective effects on oral streptococci and altered the composition and metabolism of multispecies biofilms [45, 56]. Such effects on early colonizers may influence subsequent biofilm development and ecological succession. Future studies should therefore examine O₂MW in multispecies community models.

This study has limitations. Single-species in vitro models do not fully reproduce multispecies periodontal biofilms, saliva, host factors, or subgingival microenvironments. Bacterial inocula were standardized by OD₆₀₀; therefore, equivalent OD values may not represent identical viable cell densities between the two species. The qPCR data represent transcriptional responses only and require protein-level or functional validation. Oxygen release, peroxide generation, reactive oxygen species, redox changes, and protein activity were also not directly measured after O₂MW exposure. Future studies should also determine whether the antibacterial and antibiofilm concentrations identified here remain compatible with host-cell viability and periodontal tissue repair. Despite these limitations, O₂MW showed organism-specific antibacterial and antibiofilm effects, with lower MIC and MBIC values for A. actinomycetemcomitans and higher concentration requirements for P. intermedia. These findings provide a preclinical basis for further evaluation of O₂MW for periodontal biofilm control, while clinical efficacy remains to be established.

Conclusion

In conclusion, O₂MW demonstrated species-dependent activity against A. actinomycetemcomitans and P. intermedia, with greater activity at lower concentrations against A. actinomycetemcomitans. O₂MW reduced biofilm biomass and structure and induced distinct transcriptional responses in genes associated with adhesion, virulence, oxidative stress, and iron homeostasis. Overall, these findings support further preclinical evaluation of O₂MW as a potential adjunct for periodontal biofilm control.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Fatemah M. AlAhmari (FMA): Conceptualization, Methodology, Validation, Formal analysis, Resources, Writing – Original draft, Writing – Review & editing, Supervision, Funding acquisition. Rhodanne Nicole A. Lambarte (RNAL): Formal analysis, Investigation, Data Curation, Writing – Original draft, Writing – Review & editing, Visualization, Project administration. Terrence S. Sumague (TSS): Software, Formal analysis, Investigation, Data Curation, Writing – Review & editing, Visualization. Mary Grace B. Vigilla (MGBV): Investigation, Data Curation, Writing – Review & editing. Marwa Y. Shaheen (MYS): Conceptualization, Methodology, Validation, Formal analysis, Resources, Writing – Review & editing. Sumaiah Ajlan (SA): Conceptualization, Methodology, Validation, Resources, Writing – Review & editing. Lamees Alssum (LA): Conceptualization, Methodology, Validation, Resources, Writing – Review & editing. Amani M. Basudan (AMB): Conceptualization, Methodology, Validation, Resources, Writing – Review & editing. Abdurahman A. Niazy (AAN): Methodology, Formal analysis, Investigation, Resources, Writing – Review & editing, Supervision. All authors have read and agreed to the final version of the manuscript.

Funding

This research was funded by the Ongoing Research Funding Program (ORF-2026-2076), King Saud University, Riyadh, Saudi Arabia.

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Declarations

Competing interests

The authors declare no competing interests.