Distinct functional profiles of oral neutrophils in molar incisor pattern periodontitis, generalized periodontitis and periodontal health

Abstract

This study aimed to compare oral neutrophil (oNeut) functions in molar–incisor pattern periodontitis (MIPP), generalized periodontitis (GP), and periodontally healthy subjects, and to explore how biofilm exposure shapes these functions. oNeut were isolated from healthy, GP, and MIPP volunteers (n = 10 per group) and challenged ex vivo with Aggregatibacter actinomycetemcomitans JP2. Reactive oxygen species (ROS) production, cell viability, and cytokine release were quantified post-infection. Separately, healthy oNeut were exposed to de novo biofilms modeling healthy, GP, or MIPP microbiomes, and their functional responses were assessed. Results showed that periodontitis patients (GP and MIPP) had higher baseline oNeut counts but exhibited reduced resistance to necrosis and lower ROS output after JP2 challenge than controls; JP2-stimulated ROS was significantly lower than both HOCl-treated and naïve controls. MIPP oNeut secreted more TNFα, CCL2, OPG, and RANKL than GP, whereas GP displayed a higher OPG/RANKL ratio. Except for TNFα and IL-1β, all measured mediators were elevated in healthy oNeut compared with those from periodontitis groups. Under dysbiotic versus symbiotic biofilm challenge, healthy oNeut produced less ROS but secreted higher levels of TNFα, OPG, and RANKL. Overall, oNeut from periodontitis patients exhibited distinct oxidative and cytokine responses to JP2, reflecting both host-specific and biofilm-driven priming.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-39112-3.

Introduction

Periodontitis is a condition characterized by inflammation-mediated alveolar bone loss. Periodontal inflammation is initiated and perpetuated by a chronic presence of the dysbiotic microbiome in the periodontal pockets. This condition has two main distinct phenotypes – generalized periodontitis (GP, previously termed chronic periodontitis) and molar-incisor pattern periodontitis (MIPP, previously termed aggressive periodontitis), with an ongoing debate as to whether these phenotypes represent two forms of the same pathology or distinct pathologies.

While conditions of periodontal diseases are associated with dysfunctional host-mediated tissue destruction, GP and MIPP differ in their dysbiotic microbiome; the GP dysbiotic microbiome is diverse, structurally disordered, enriched in pathogenic taxa, and characterized by host-destructive metabolic pathways¹. This pathogenic microbiota consists of different gram-negative species that outgrow health-associated taxa. Among the enriched species are the classical red-complex triad consisting of Treponema denticola, Porphyromonas gingivalis, and Tanneralla forsythia. Several other Treponema spp. also appear to be abundant components of periodontitis communities, in agreement with classic microscopy studies that indicate the abundance of spirochaetes². While the MIPP dysbiotic microbiome is also diverse, it is associated with the presence of Aggregatibacter actinomycetemcomitans^3–8. In MIPP, the JP2 clone of A. actinomycetemcomitans is present, with a strong prognostic association with the initiation and progression of the disease and tissue attachment loss^5,9.

Early studies on the function of neutrophils in MIPP suggest that peripheral neutrophils play a major protective role against periodontal infection, and that defects in chemotaxis may predispose individuals to periodontal disease^10,11. Still, Mizuno et al. showed that only some subjects with MIPP display chemotactic dysfunction¹². Another study on circulating neutrophils from different periodontal patients showed that the extracellular release of reactive oxygen species was higher in GP than in MIPP, and neutrophil elastase release was elevated in both periodontal groups compared with periodontally healthy controls¹³. Such conflicts led to a shift in the paradigm from the hyporesponsive to the hyperresponsive model of neutrophil dysfunction in periodontal etiopathogenesis¹⁴. Tapashetti et al. showed that the prevalence of neutrophil dysfunction, predominantly hypofunctional, was significantly high in GP patients, with a few having hyperactive respiratory burst function¹⁵. Johnstone et al. found larger receptor-independent respiratory bursts and higher phagocytotic activity in peripheral neutrophils derived from patients with recurrent MIPP when compared with GP and periodontally healthy patients¹⁶. Since then, ample studies have demonstrated that periodontal neutrophils are hyperactive and primed and release enhanced levels of oxygen radicals and inflammatory mediators, such as cytokines and matrix-degrading enzymes^13,17–19. The hyperactivity of neutrophils is also associated with the destruction of periodontal tissues; indeed, several studies have shown that impaired neutrophil functions, such as defective and depressed chemotaxis, decreased phagocytic function, and increased production of reactive oxygen species, which increases the activity of proteases and activates neutrophil-produced matrix metalloproteinases, contribute to tissue damage^14,18,20,21.

Most neutrophils circulate in the bloodstream and are recruited to peripheral tissue sites at times of infection^22,23. Oral neutrophils are circulating neutrophils that migrate from the bloodstream through the gingival tissue and gingival pockets into the gingival crevicular environment^23,24. They have a hyperactive phenotype characterized by increased potential for ROS production²⁵. In addition, oral neutrophils are more apoptotic, with increased levels of degranulation markers in periodontitis compared to periodontal health²⁶. Lakschevitz et al. demonstrated that oral neutrophils show a significant increase in T-cell receptor expression compared with circulating neutrophils, suggesting a role for oral neutrophils in crosstalk between the innate and adaptive immune systems in the mouth²⁷. Furthermore, when comparing oral and circulating PMNs, oral cells from patients with periodontal disease displayed an altered transcriptome after migration into the tissues, leading to a pro-survival phenotype in GP patients compared with healthy subjects, and consequently a longer neutrophil lifespan²⁸. Still, no study has compared differences in oral neutrophils stemming from patients with different periodontal diagnoses, and the function of oral neutrophils in MIPP is currently unknown.

Based on previous reports of altered neutrophil function in periodontitis, we hypothesized that oral neutrophils from patients with MIPP and GP would display impaired ROS generation and distinct cytokine expression patterns compared with healthy controls. We further hypothesized that the magnitude of these differences would approximate a 15% change in ROS production, consistent with prior studies, which provided the basis for our sample size calculation.

The present study aims to examine oral neutrophil function in MIPP, GP, and periodontally healthy subjects and the impact of the different dysbiotic microbiomes on their priming.

Results

Oral neutrophil extraction from whole saliva and their purity are shown in Supplementary Fig. 1. The mean age of each group was 63.3 years for GP, 26.4 years for MIPP, and 29.7 years for the periodontally healthy group, and the percentage of females was 70% in GP, 55% in MIPP, and 50% in the periodontally healthy group.

The average number of neutrophils obtained from each case was significantly higher in the periodontitis groups compared with the periodontally healthy group (mean counts: 3.4 × 10⁶ cells in MIPP, 3.1 × 10⁶ cells in GP, and 1.4 × 10⁶ cells in the periodontally healthy group; P < 0.01).

Periodontal diagnosis is associated with a functional change in oral neutrophils

In periodontally healthy individuals, substantial oxidative stress was observed in naïve oNeut, whereas both JP2 infection and the positive control showed lower levels than the naïve group (Fig. 1a). Moreover, the JP2 group exhibited lower ROS levels compared with the positive control. This pattern differed in both periodontitis groups: while naïve cells expressed ROS, the positive control showed similar values, and only the JP2 groups demonstrated reduced levels (Fig. 1a). Overall, periodontally healthy oNeut expressed higher ROS compared with the periodontitis groups under all tested conditions, with only minor differences in ROS kinetics (Fig. 2a). ROS measurements at the end of incubation revealed that in the healthy group, two subsets of oNeut were present—one with low ROS levels and another with high ROS levels (Fig. 1c). Interestingly, in the GP group, the high ROS subset was absent, whereas in the MIPP group, the low ROS subset was missing (Fig. 1c).

Fig. 1.

Reactive oxygen species expression in oral neutrophils from periodontally healthy subjects, generalized periodontitis (GP), and molar–incisor pattern periodontitis (MIPP). Oral neutrophils were extracted from whole saliva and inoculated with A. actinomycetemcomitans JP2 or HOCl (positive control, PC) and compared with naïve cells (negative control, NC). ROS production was measured using DCFH and a fluorescence plate reader over 90 min and presented as (A) cumulative ROS (area under the curve [AUC] of relative fluorescence units [RFU]) ± SD and (B) ROS kinetics. At the end of incubation, cells were collected for analysis by flow cytometry (C). *P < 0.05, **P < 0.01, ***P < 0.0001; statistically significant differences.

Fig. 2.

Evaluation of necrosis and reactive oxygen species expression in oral neutrophils from periodontally healthy subjects, generalized periodontitis (GP), and molar–incisor pattern periodontitis (MIPP). Oral neutrophils were extracted from whole saliva and inoculated with A. actinomycetemcomitans JP2 or HOCl (positive control, PC) and compared with naïve cells (negative control, NC). ROS production was detected using DCFH, and necrosis was assessed using propidium iodide. (A) Representative flow cytometry dot plots of all groups showing ROS and necrosis levels in neutrophils. (B) Quantification of ROS/necrosis-positive cells ± SD. (C) Quantification of high ROS–positive or high necrosis–positive cells ± SD. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; statistically significant differences.

ROS expression is independent of JP2-induced cell necrosis

JP2 is known to induce necrosis in neutrophils, thereby affecting ROS production. To examine this, we stained JP2-infected cells for both ROS and necrosis. The results showed that in healthy oNeut, most cells were viable (PI-negative) and expressed ROS (DCFH-positive) across all tested groups (Fig. 2a). However, in the periodontally healthy group under JP2 infection, 25% of oNeut were negative for both necrosis and ROS, while 40% were positive for both necrosis and ROS (Fig. 2a). In the GP group, a clear difference was observed under JP2 infection: the majority of cells displayed low ROS expression, and only 10% expressed ROS. In the MIPP group, JP2 infection did not induce ROS production at all, and most cells remained PI-negative (viable). In the positive control, both periodontitis groups showed two populations of cells—one ROS-positive and the other ROS-negative.

Flow cytometry quantification revealed a reversed pattern in the proportions of viable ROS-positive (PI⁻DCFH⁺) and viable ROS-negative (PI⁻DCFH⁻) cells, with JP2 infection reducing the percentage of viable ROS-positive cells (Fig. 2b). Notably, in the periodontitis groups, JP2 infection induced a peak in necrotic cells lacking ROS expression, a finding not observed in the periodontally healthy group (Fig. 2b). When analyzing only viable or ROS-expressing cells, a reversed pattern was again observed in both periodontitis groups, but not in the periodontally healthy group, under JP2 infection (Fig. 2c).

MIPP oral neutrophils express pro-inflammation and pro-osteoclastogenesis phenotypes

While ROS is a key effector function of neutrophils, other traits such as cytokine expression may further elucidate differences in oNeut function between the tested groups. To this end, we examined cytokine and chemokine expression, including mediators associated with the hallmark phenotype of periodontitis—bone loss. The results showed that TNFα and CCL2 expression were reduced in GP, whereas in MIPP their levels were more similar to those of healthy controls (Fig. 3a). IL-1β and CXCL10 displayed an opposite pattern, with the periodontitis groups showing elevated IL-1β and reduced CXCL10. In both periodontitis groups, RANKL and OPG levels were reduced, resulting in a decreased OPG/RANKL ratio compared with the healthy group (Fig. 3b).

Fig. 3.

Cytokine expression in oral neutrophils from periodontally healthy subjects, generalized periodontitis (GP), and molar–incisor pattern periodontitis (MIPP). Oral neutrophils were extracted from whole saliva and inoculated with A. actinomycetemcomitans JP2 or HOCl (positive control, PC) and compared with naïve cells (negative control, NC). mRNA was collected for qRT-PCR analysis. Results are expressed as fold change ± SD. (A) Inflammatory cytokines. (B) Osseomodulatory cytokines. *P < 0.05, **P < 0.01, ***P < 0.001, ***P < 0.0001; statistically significant differences.

Reduced ROS generation and increased cytokine/mediator secretion by oral neutrophils in response to periodontal biofilms

Oral neutrophils are immune cells that migrate from the bloodstream into the periodontal pocket tissue, then through infected pockets, and are eventually shed into the oral cavity. This trajectory may affect neutrophil priming and effector function. To test this, we collected neutrophils from periodontally healthy individuals and exposed them separately to de novo biofilms representing health, GP, or MIPP, and examined their functionality.

ROS production by oNeut was higher in the healthy biofilm than in either of the periodontal biofilms (Fig. 4a), resembling the pattern observed in primary oNeut (Fig. 1). In contrast, expression of the inflammatory cytokine TNFα, the osteoclastogenic cytokine RANKL, and the osteoprotective cytokine OPG was elevated in the periodontal biofilms compared with the healthy biofilm (Fig. 4b), which was opposite to the pattern observed in primary oNeut (Fig. 3).

Fig. 4.

ROS production and cytokine expression in oral neutrophils from periodontally healthy subjects exposed to different biofilms. Oral neutrophils were extracted from whole saliva and incubated separately with three biofilms, each representing a distinct periodontal condition: health, generalized periodontitis (GP), or molar–incisor pattern periodontitis (MIPP). (A) ROS production during 90 min of incubation with the biofilms, expressed as area under the curve (AUC) ± SD. (B) Cytokine levels after incubation with the biofilms, expressed as fold change ± SD. **P < 0.01, ***P < 0.001, ***P < 0.0001; statistically significant differences.

Discussion

This study describes a unique phenotype of oral neutrophils that depends on periodontal diagnosis and the associated dysbiotic microbiome. On the one hand, the high oNeut counts, resistance to necrosis, low ROS production, and distinct cytokine expression profile reflect the complex phenotype of periodontitis and its association with bone loss. On the other hand, examining oNeut from healthy individuals exposed to different biofilms showed a consistent effect on ROS production but not on cytokine profiles. This discrepancy may be explained by microbiome-dependent neutrophil priming (reflected by ROS) that is independent of the immunomodulatory functions of neutrophils (cytokine expression).

Oral neutrophils migrate through the connective tissue and pocket epithelium to kill pocket microbes continuously, even without clinical inflammation or tissue damage²⁹. In periodontitis, inflammation at the periodontal pockets³⁰ causes chemokine paralysis²⁹ and massive recruitment of neutrophils into the pocket and subsequently into saliva. Thus, oral neutrophils in periodontal health and periodontitis reflect the complete neutrophil journey, thereby exhibiting functions that are linked to the destructive/protective host response. Several studies have compared circulatory and oral neutrophils (cNeut vs. oNeut) during bacterial insult; Rijkschroeff et al. showed that in healthy subjects, oNeut were more activated and in a more mature state than cNeut, although when challenged with bacteria, both cell types had a similar response³¹. Nicu et al. compared oNeut and cNeut in healthy and untreated periodontitis cases and showed a higher count, more apoptosis, higher expression of activation markers, and higher ROS production in oNeut in the periodontitis group²⁶. Still, as in the previous study, both infected cells showed a similar response. These data suggest a hyperactive state of oNeut when compared to circulatory neutrophils in healthy and periodontal subjects. Hashai et al. examined circulatory neutrophils from healthy and MIPP subjects using the periopathogen A. actinomycetemcomitans JP2 clone and showed MIPP hypoactivation³². Indeed, oNeut levels correlate with the extent of oral inflammation and periodontal severity (Khoury, 2020) compared with healthy controls²⁶. Similar to Nicu et al., our study showed that periodontitis cases (both GP and MIPP) have higher oNeut counts than healthy patients.

Our data showed that oNeut from healthy individuals had higher ROS level than the periodontitis groups (GP and MIPP patients), whether stimulated or unstimulated. This difference may be due to the exhaustion of the oNeut ROS potential in periodontal subjects, as they are constantly exposed to periopathogenic bacteria and remain chronically active. In contrast, Nicu et al. (2018) did not find a difference in the ROS production of oral neutrophils between periodontal and healthy patients²⁶. A possible explanation for this difference might be that Nicu et al. based their study on circulating neutrophils. In periodontitis patients, the JP2-inoculated oNeut expressed lower ROS when compared to naïve oNeut, positive control, or JP2-inoculated oNeut in periodontally healthy patients. Similar results were shown in circulatory neutrophils, where MIPP PMNs exhibited lower ROS production in response to JP2 infection compared to neutrophils from the periodontally healthy group³². Rijkschroeff et al. showed higher ROS production in infected oNeut originating from healthy patients versus naïve cells³¹, which contradicts our findings, which show similar ROS production in the infected and naïve oNeut of healthy patients. This difference may be due to the fact that F. nucleatum was used as a stimulator in Rijkschroeff’s paper, while our study and that of Hashai and colleagues used Aggregatibacter actinomycetemcomitans JP2 clone infection.

oNeut from healthy individuals expressed higher cytokine levels and a higher OPG/RANKL ratio compared with periodontal patients, reflecting a low-inflammation, bone-protective phenotype. The OPG/RANKL ratio did not differ between the GP and MIPP groups, which reflects an osteoresorptive phenotype. On the other hand, IL-1β and CCL2 showed augmented levels in the MIPP group compared with the GP group, which may indicate a different inflammatory phenotype among the periodontitis groups. This is aligned with Galbraith et al., who showed that the cytokines released by oNeut were higher in periodontal patients³³. In addition, other evidence has shown a strong expression of IL-1β and TNF-α in gingival specimens of periodontal patients³⁴. TNFα plays a major role in the regulation of bone homeostasis by stimulating osteoclastogenesis, which requires the synergistic effect of RANKL and TNFα^35,36. This might explain the low OPG/RANKL ratio, which is pro-resorption and a less protective state of bone. Furthermore, this is compatible with the clinical features of MIPP patients where there was site-specific yet dramatic bone loss.

In conclusion, periodontal patients present an imbalanced cytokine profile and a low OPG/RANKL ratio when compared to healthy subjects, which might explain the bone resorption pattern in periodontitis. The difference between GP and MIPP patients was evident in TNFα levels in cytokines and protein-based analyses, exhibiting higher TNFα in MIPP patients in naïve and JP2-infected patients, which may contribute to the rapid bone loss associated with such a condition.

This study has several limitations. First, our focus was restricted to neutrophils; a broader assessment of additional immune cell populations could provide a more comprehensive picture of host susceptibility in periodontitis. Second, the biofilm models, although based on well-characterized species combinations, represent simplified versions of the complex native oral microbiome and, like any in vitro system, cannot fully replicate its ecological complexity or clinical relevance. Third, although neutrophils are not the predominant source of all mediators measured, prior work supports their ability to release molecules such as OPG and RANKL in inflammatory contexts, suggesting a potential role in linking innate immune activation with bone remodeling. Finally, the biofilm consortia were not independently verified for microbial composition or structural organization; future studies should address this limitation using quantitative approaches such as species-specific qPCR, 16S rRNA sequencing, or CLSM imaging to validate species representation and biofilm architecture.

Conclusions

Oral neutrophils demonstrate distinct functional phenotypes that reflect periodontal health status and the influence of a dysbiotic microbiome. Periodontitis patients exhibit increased oral neutrophil counts, altered reactive oxygen species production, and a cytokine profile characterized by a lower OPG/RANKL ratio and elevated inflammatory mediators, particularly TNFα in MIPP patients. These findings indicate that neutrophil dysfunction and altered immunomodulation contribute to bone resorption and the rapid tissue destruction characteristic of aggressive periodontal conditions. Collectively, this work establishes oral neutrophils as key cellular mediators linking dysbiotic biofilms to periodontal tissue breakdown.

Methods

Accordance statement

All experimental procedures were approved by the Hadassah Medical Center Institutional Review Board and were performed in accordance with relevant guidelines and regulations.

Study population

The study was approved by the Hadassah Medical Organization Institutional Helsinki Committee (approval number 0033–19-HMO), and all participants provided written informed consent prior to enrollment.

The study was designed as a parallel-arms study including three groups of participants: periodontally healthy, generalized periodontitis (GP), and molar–incisor pattern periodontitis (MIPP). Diagnosis was established according to the 2017/2018 World Workshop classification of periodontal and peri-implant diseases and conditions³⁷ and based on a comprehensive periodontal examination. Clinical parameters recorded included probing pocket depth (PPD), clinical attachment level (CAL), bleeding on probing (BOP), and radiographic bone loss. Periodontally healthy participants presented with PPD ≤ 3 mm at all sites, no CAL, no radiographic evidence of bone loss, and < 10% of sites with BOP. GP was defined as CAL at non-molar/non-incisor sites affecting ≥ 30% of teeth, while MIPP was defined as interproximal attachment loss localized to first molars and incisors with radiographic evidence of bone loss.

Upon signing informed consent, all participants were invited for oral neutrophil collection. Inclusion criteria were: age ≥ 18 years, systemically healthy, and willingness to participate. Participants were considered systemically healthy if they reported no history of cardiovascular disease, diabetes, autoimmune or inflammatory disorders, cancer, or ongoing systemic infection, and had not received antibiotics, immunosuppressive drugs, or long-term anti-inflammatory medications in the preceding three months.

Exclusion criteria included: diagnosis of diabetes, cardiovascular disease, coagulation disorders, chronic use/abuse of drugs or alcohol, pregnancy, smoking more than 10 cigarettes/day, antibiotic use within the last three months, or regular use of systemic anti-inflammatory medications. Participants smoking ≤ 10 cigarettes/day were permitted in the healthy group, in line with epidemiologic definitions of light smoking, and smoking status was recorded for all groups.

Oral neutrophil collection

The patients were instructed to refrain from eating/drinking one hour prior to collection. Each patient was given two 15 ml tubes of Hanks Balanced Salt Solution (HBSS) to rinse their mouths for 30 seconds³⁸, each time with a different tube. The two suspensions were pooled and filtered through a 40 μm pore nylon mesh and then through a 10 μm pore mesh. The samples were then centrifuged at 1250 rpm/10 °C/10 min and the cellular pellet was suspended in HBSS^39,40.

All isolation steps were performed under cold conditions using pre-chilled HBSS and plasticware; cells were processed on ice and centrifuged at 10 °C. Total handling time from collection to plating was kept under approximately 45 min to minimize unintentional priming.

Cell purity and viability

Cell purity and viability were assessed using CD16 antibody staining and Annexin V/propidium iodide (PI) staining, respectively, followed by flow cytometry. Briefly, collected cells were stained with anti-human CD16 antibody (clone 3G8, mouse IgG1κ, BioLegend, San Diego, CA, USA) at a 1:100 dilution and incubated for 30 min on ice. After staining, cells were washed in HBSS and analyzed by flow cytometry. Isotype-matched controls (mouse IgG1κ, BioLegend) were included to assess background fluorescence.

Cell identity was confirmed by anti-CD16 antibody staining. Since the oral rinse–based method reproducibly yields > 95% neutrophil purity⁴¹. No additional CD16-based sorting was performed to avoid unnecessary manipulation that could compromise neutrophil viability and function.

Bacteria cultivation

Aggregatibacter actinomycetemcomitans strain JP2 was grown in a medium containing 0.5 g yeast extract, 1.5 g Tryptone, 0.74 g D-glucose, 0.25 g NaCl, 0.075 g L-cysteine, 0.05 g sodium thioglycolate, and 4% NaHCO₃ (Sigma-Aldrich, Rehovot, Israel) in double-distilled water at 37 °C and 5% CO₂. Quantification of the bacteria was done by optical density (OD) measurement³². Fusobacterium nucleatum PK1594, Porphyromonas gingivalis ATCC 33,277, Streptococcus sanguis NC02863, and Actinomyces naeslundii 17,233 were separately grown in Wilkins-Chagren broth (Oxoid, Basingstoke, Hampshire, UK) and incubated at 37 °C for 24 h under anaerobic conditions (N₂ 85%, H₂ 5%, CO₂ 10%). S. sanguis and A. naeslundii were transferred to Wilkins-Chagren broth enriched with 2% sucrose (Sigma, Rehovot, Israel) and cultured under anaerobic conditions for an additional 24 h. F. nucleatum and P. gingivalis were transferred to Wilkins-Chagren broth and incubated for a further 24 h under anaerobic conditions. The bacterial suspensions of S. sanguis, A. naeslundii, and P. gingivalis were adjusted spectrophotometrically to 10⁹ cells/mL, and that of F. nucleatum was adjusted to 10⁸ cells/mL⁴².

The JP2 clone of A. actinomycetemcomitans was used as a high-virulence strain linked to MIPP, while an acapsular P. gingivalis strain was chosen for its genetic stability and reproducibility in in vitro models.

In vitro biofilm model

The bacterial suspensions were centrifuged (4000 rpm, 15 min) and suspended in a gingival crevicular fluid (GCF)-simulating medium (60% RPMI medium, 40% donor horse serum (Biological Industries, Beit Ha’emek, Israel)) enriched with 5 µg/mL hemin and 0.5 µg/mL menadione (both from Sigma). A suspension of S. sanguis and A. naeslundii (1:1 ratio in a total volume of 100 µl GCF-simulating medium) was inoculated onto 96-well plates and incubated for 24 h at 37 °C under anaerobic conditions. The wells with the newly formed biofilm were then washed with phosphate-buffered saline (PBS) and received the next preparation steps, as follows: For the GP biofilm, a suspension of P. gingivalis and F. nucleatum (1:1 ratio in a total volume of 100 µl GCF-simulating medium) was inoculated and incubated for an additional 48 h at 37 °C under anaerobic conditions. For the MIPP biofilm, a suspension of A. actinomycetemcomitans in a total volume of 100 µl GCF-simulating medium was inoculated and incubated for an additional 48 h at 37 °C under anaerobic conditions. For a healthy periodontal biofilm, no additional bacteria were added.

These biofilm models were designed as simplified constructs using well-characterized species previously associated with symbiotic (health) or dysbiotic (disease) states, rather than full representations of the in vivo oral microbiome.

Reactive oxygen species production assay

The 96-well black plates were coated with 1% BSA in PBS (100 μl/well) overnight at 4 °C. ROS generation was quantified using 2′,7′-dichlorofluorescin diacetate (DCFH-DA, 10 µg/ml), a sensitive and widely applied method for assessing neutrophil oxidative burst^43,44. HOCl (0.0006%) was used as a positive control. Oral neutrophils were prepared in PBS containing calcium and magnesium at 1 × 10⁶ cells/ml, and 50,000 cells/well were seeded, a number optimized in pilot experiments to provide robust fluorescent signals without assay saturation. Cells were divided into three groups: neutrophils alone, neutrophils + HOCl, and neutrophils + A. actinomycetemcomitans JP2 (MOI 10). Each well received 50 μl of DCFH-DA and 50,000 cells. Plates were incubated for 90 min, a time point selected based on prior reports and preliminary testing showing stable ROS accumulation within this window. Fluorescence was recorded on a fluorescent plate reader (Tecan inc Grödig, Austria) at excitation/emission wavelengths of 488/525 nm. Following incubation, cells were harvested, stained with propidium iodide (2% in HBSS, 10 min at RT), and analyzed by flow cytometry.

qRT-PCR

Cell RNA was extracted using TRIzol reagent (Thermo Fisher Scientific, Waltham, Massachusetts, USA) according to the manufacturer’s instructions. Double-stranded cDNA was synthesized with 1 μg of total RNA using a qScript cDNA synthesis kit (Quantabio, Beverly, MA) according to the manufacturer’s instructions. SYBR Green quantitative real-time PCR was performed (PCR BIOSYSTEMS, London, UK). The primer sets used in this study are shown in the table below. All the reactions were carried out in duplicates, and the data were analyzed using the 2^−ΔΔCT method.

Gene name	Forward	Reverse
TNFα	CCTCTCTCTAATCAGCCCTCTG	GAGGACCTGGGAGTAGATGAG
IL1β	AGCTACGAATCTCCGACCAC	CGTTATCCCATGTGTCGAAGAA
CXCL10	GTGGCATTCAAGGAGTACCTC	TGATGGCCTTCGATTCTGGATT
CCL2	CAGCCAGATGCAATCAATGCC	TGGAATCCTGAACCCACTTCT
RANKL	TCGTTGGATCACAGCACATCA	TATGGGAACCAGATGGGATGTC
OPG	CGTCAAGCAGGAGTGCAATC	CCAGCTTGCACCACTCCAA
β-actin	CATGTACGTTGCTATCCAGGC	CTCCTTAATGTCACGCACGAT

The mediator panel included both classical neutrophil products (TNFα, IL-1β) and additional molecules (e.g., OPG, RANKL) that, while not predominantly produced by neutrophils, have been reported to be expressed by these cells under inflammatory or biofilm-driven conditions and are central to periodontal pathogenesis.

Data analysis

The sample size (n = 10/group) was determined based on an anticipated 15% change in ROS measurements, guided by effect sizes reported in prior studies of oral neutrophil function²⁸. Using these estimates, a power calculation (two-tailed, α = 0.05, β = 0.20) indicated that 10 participants per group would provide > 80% power to detect significant differences. Participants were recruited consecutively from the clinic population and allocated into groups according to their clinical periodontal diagnosis (MIPP, GP, or healthy controls). Data were analyzed using SigmaStat (Jandel Scientific, San Rafael, CA, USA). A one-way repeated measures analysis of variance (RM ANOVA) was applied to test for overall group differences in within-sample treatment comparisons (Figs. 1, 2 and 4). For gene expression data (Fig. 3), which involved comparisons between independent groups, between-group differences in ΔCt values were evaluated using one-way ANOVA followed by Student’s t-tests with Bonferroni correction for multiple comparisons.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization, DE and DP; Methodology, KJ, HB, and DP.; Investigation, KJ, HB, MN, HK, and DP; Writing—Original Draft, DE, and DP; Writing—Review & Editing, KJ, HB, MN, HK, ED and DP; Funding Acquisition, DP; Resources, KJ, HB, MN, HK, ED and DP; Supervision, ED and DP.

Funding

The study was self-funded.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Suitability for scientific reports

Scientific Reports values rigorous, mechanistic studies that advance broad biological understanding. Our work combines quantitative neutrophil assays with clinically relevant biofilm models to reveal novel insights into host–microbe interactions in periodontitis, making it a strong fit for the journal’s scope.