Salivary metagenomic profiling of Neisseria, Dialister, and Filifactor species in periodontal health and disease using next-generation sequencing

Abstract

Background

Periodontal diseases represent a complex dysbiosis-driven inflammatory condition, where the transition from health to gingivitis and periodontitis is accompanied by distinct microbial shifts. Emerging evidence highlights the significance of less-studied genera such as Neisseria, Dialister, and Filifactor in shaping periodontal outcomes. This study aimed to investigate the salivary distribution of Neisseria, Dialister, and Filifactor species across periodontal health, gingivitis, periodontitis, and gingival recession using next-generation sequencing (NGS).

Methods

Whole saliva samples were collected from 40 participants (10 per group) classified according to the American Academy of Periodontology criteria. Microbial DNA was extracted and subjected to 16S rRNA sequencing (V3–V4 region, Illumina MiSeq). Species-level classification was performed using the Human Oral Microbiome Database. Frequency distributions were compared across groups using Fisher's exact test, with significance set at p < 0.05.

Results

Distinct patterns were observed. Several commensal Neisseria species, including N. subflava (p = 0.001), N. elongata(p = 0.015), and N. polysaccharea (p = 0.001), showed significantly reduced prevalence in periodontitis compared with health and gingivitis. In contrast, Dialister pneumosintes exhibited a sharp increase in all diseased groups (p = 0.002). Filifactor alocis was markedly enriched in gingivitis, recession, and periodontitis (p = 0.011), suggesting its strong association with disease states.

Conclusion

The findings demonstrate a characteristic microbial shift in saliva: health-associated Neisseria species decline with disease progression, while anaerobic taxa such as D. pneumosintes and F. alocis expand. These results align with the polymicrobial synergy and dysbiosis model and underscore the potential of these species as salivary biomarkers for early detection and monitoring of periodontal disease.

1. Introduction

Periodontal disease involves the destruction of the supporting structures of the tooth due to colonized microbes in the form of polymicrobial biofilm, which is a key factor in the initiation of the disease. However, the disease is multifactorial in nature, and microbial etiology alone is not enough to cause it. The subgingival biofilm presents a strong microbial challenge, which stimulates the host's immune and inflammatory response, leading to increased production of inflammatory mediators. These mediators, in turn, destroy periodontal structures and ultimately lead to tooth loss. Gingivitis, a reversible form of the disease, is considered a non-specific reaction to dental plaque, affecting a majority of the world's population across all age groups. Periodontitis, however, is a more severe form of the disease that can ultimately lead to tooth loss if left untreated. It affects over 11% of the global population and is regarded as the sixth most common chronic disease worldwide ^{1, 2, 3, 4}.

Both these conditions are characteristic traits of the same inflammatory cascade, although it is worth noting that not all gingivitis lesions progress to periodontitis. The transition to periodontitis and its further advancement seem to depend more on an individual's disease susceptibility. Factors influencing disease susceptibility include modifiable and non-modifiable risks, such as genetic, inflammatory/immunological, and environmental factors like age, race, gender, tobacco use, and systemic diseases such as diabetes mellitus ^{5, 6, 7}. The bacterial communities in the subgingival biofilm constitute the microbiome of the periodontal environment and could turn dysbiotic under favourable circumstances, which depend on a myriad of host factors as well. This, in turn, could set in the inflammatory tone in the periodontal milieu. So, the microbiome plays a pivotal role in adapting harmoniously and maintaining the ecological balance in the oral cavity niches.^8,9 Any disorganisation of this colonised community triggers the progression of periodontal disease.¹⁰

Several studies have demonstrated the presence of poly microbial communities in the sub-gingival microenvironment and have improved our understanding of their role in the etiopathogenesis of periodontal disease.¹¹ However, most microbiota remained unidentified in the early culture-based techniques of identification. Later, more advanced molecular methods based on microarray techniques and DNA-DNA hybridization improved the identification and understanding of microbiota. Currently, culture-independent techniques that target small subunit rRNA sequences, combined with high-throughput DNA sequencing, have become a promising approach for detailed microbial profiling from various sites of the human body. This method has revealed the complex nature of the microbial flora associated with health and disease through the amplified and deeply sequenced 16S rRNA gene segments.

Previous studies have demonstrated distinct oral microbiomes in the subgingival environment in periodontal health and disease. In health, several taxa have been identified, especially some members of the genus Actinomyces and Streptococcus, which have been undeniably assigned to periodontal health.¹² Others have demonstrated Rothia and Streptococcus predominantly in the healthy sites. In addition, Actinobacteria, Firmicutes, Synergistetes, and TM7 showed less predominance in healthy sites. Moreover, Micrococcaceae, Actinomycetaceae, Streptococcaceae, and Fusobacteriaceae demonstrated predominance at the family level.¹³ Certain other specific genera, including the Rothia, Corynebacterium, Veillonella, Neisseria, and Actinomyces, were significantly high in health, while distinctly other genera, Treponema, Tannerella, Eubacterium, and Campylobacter unquestionably correlated with periodontitis.¹² The diseased sites were also predominated by Fusobacterium and Porphyromonas.¹³ A previous study identified the presence of Synergistetes in gingivitis, and its absence in health; presence of the HACEK group species in both health and gingivitis, while Dialister and Aneroglobus dominated the gingivitis sites.¹⁴ Another previous study demonstrated that the oral microbiome in health was similar to that of disease at the genus level but exhibited a significant variation at the species level, which could be attributed to demographic, environmental, and other factors.¹⁵ Thus, the previous literature has demonstrated that the oral microbiome in health is represented to have remained in a well-balanced ecosystem, and any dysbiosis may lead to its translation to gingivitis, and eventually to periodontitis with an underlying microbial shift.^16,17

Though the association between specific groups of taxa, with health versus disease, has been extensively studied, little is understood about its succession from health to disease. For many members, distinguishing between health and disease has been challenging due to the ambiguity caused by their similar prevalence and relative abundance. This could be substantiated by the high prevalence of species, viz. Fusobacterium nucleatum, Veillonella parvula, while showcasing less abundance of Lautropia mirabilis, Campylobacter gracilis, or Granulicatella adjacens.^18,19 Based on the available literature, a more precise microbial measure of dysbiosis could address gaps in the data, enhancing our understanding and diagnosis of periodontitis. Additionally, defining microbial signatures specific to health and disease states may facilitate effective monitoring of periodontal disease, enabling tracking of disease progression or recovery and potentially identifying individuals at heightened risk of developing periodontitis.

Emerging evidence suggests that specific bacterial species within the genera Neisseria, Dialister, and Filifactor may contribute to periodontal disease. However, existing research has largely focused on a few well-studied species, while other subspecies and their roles remain poorly understood. The advantage of focusing on these genera lies in their contrasting ecological roles within the periodontal microbiome. Several Neisseria species are considered health-associated commensals that contribute to microbial homeostasis and nitrate metabolism, whereas Dialister pneumosintes and Filifactor alocis are emerging anaerobic taxa strongly linked with periodontal inflammation and tissue destruction. Studying these organisms together enables characterization of the microbial shift from a symbiotic, health-associated community towards a dysbiotic, disease-associated profile, consistent with the polymicrobial synergy and dysbiosis model of periodontitis. In addition, identifying disease-associated taxa such as F. alocis and D. pneumosintes may, in future, aid in refining targeted antimicrobial or host-modulatory strategies.

Understanding the distribution and prevalence of these bacterial species in both periodontal health and disease is essential for elucidating their potential role in oral microbial homeostasis and pathogenesis. By leveraging advanced techniques such as Next Generation Sequencing (NGS) with Illumina technology, this study aims to perform a comprehensive metagenomic analysis to map and quantify Neisseria, Dialister, and Filifactor species in whole saliva. This analysis could uncover significant insights into the microbial shifts that accompany periodontal health and disease, potentially paving the way for improved diagnostics and therapeutic strategies.

The objectives were to identify the presence of salivary Neisseria, Dialister, and Filifactor species and compare their frequency distribution across individuals with gingival health, gingivitis, periodontitis with pockets, and gingival recession, thus aiming to characterize the microbial shifts associated with each condition. This study will expand current knowledge on the microbial ecology of periodontal health and disease, particularly concerning less-studied subspecies within these genera.

2. Materials and methods

In this study, individuals seeking dental treatment at the Department of Periodontics, Ragas Dental College and Hospital, Chennai, were invited to participate. Ethical clearance for the study was obtained from the Institutional Review Board of Ragas Dental College. Participants were clinically examined and classified into four groups according to the American Academy of Periodontology (AAP) criteria,²⁰ based on their periodontal health status: clinically healthy gingiva, gingivitis, chronic periodontitis, and gingival recession. The study included a total of 40 subjects, with 10 individuals assigned to each group.

2.1. Study population

The control group consisted of 10 subjects with clinically healthy, non-inflamed gingiva, as indicated by a probing pocket depth (PPD) of 3 mm or less, no clinical attachment loss (CAL), and no bleeding on probing (BOP). Test Group 1 included 10 subjects diagnosed with gingivitis, characterized by bleeding on probing without clinical attachment loss and a PPD of 3 mm or less. Test Group 2 comprised 10 subjects with periodontitis with gingival recession, which involved clinical attachment loss of 3 mm or more, minimal PPD of 3 mm or less, and radiographic evidence of bone loss. However, subjects in this group with gingival recession due to faulty tooth brushing without other local factors or BOP were excluded. Lastly, Test Group 3 included 10 individuals with periodontitis, as evidenced by periodontal pocket formation (PPD >5 mm), clinical attachment loss of 3 mm or more, and radiographic evidence of bone loss.

2.2. Selection criteria

To be eligible for inclusion, participants had to be 20 years or older, in good systemic health, and meet the specified periodontal health or disease criteria. Exclusion criteria included the presence of systemic disorders like diabetes mellitus or immunological conditions, use of drugs potentially affecting microbial characteristics (e.g., immunosuppressants, steroids), history of tobacco use, recent periodontal treatment or antimicrobial therapy (within the past six months), history of malignancy or cancer treatment, presence of oral lesions, and pregnancy or lactation. Additionally, individuals who refused to provide informed consent were excluded from the study.

Before clinical examinations and saliva collection, the study protocol was explained to each participant, and written informed consent was obtained. Medical and dental histories were documented for all subjects to ensure compliance with inclusion and exclusion criteria.

2.3. Clinical examination

A single qualified dental surgeon conducted all clinical examinations to ensure consistency. Each subject underwent a comprehensive periodontal assessment, which included measurements of probing pocket depth (PPD), clinical attachment loss (CAL), and bleeding on probing (BOP), which was documented.

2.4. Saliva collection and sampling

A single calibrated examiner performed all saliva collection procedures to maintain methodological uniformity. Participants in both test and control groups were instructed to abstain from eating, drinking, and performing oral hygiene activities for at least 2 h before sample collection. Each participant was seated in a dental chair and shielded from gustatory or other sensory stimuli to prevent alterations in saliva composition. Saliva collection was conducted using the “passive spitting method” as described by Navazesh, 1993.²¹ Participants allowed saliva to accumulate in the floor of their mouth, then spit it into a sterile saliva collection tub every 60 s until approximately 10 ml was collected. Once collected, the saliva samples were immediately frozen and stored at −80°C to preserve their integrity until it was subjected to further analysis.

2.5. Microbial profile analysis

For microbial profiling, DNA extraction, 16S rRNA gene amplification, library construction, and sequencing were conducted on 40 saliva samples, consisting of 10 samples from each group: healthy controls, gingivitis, periodontitis patients with recession, and those with periodontal pockets. Genomic DNA was extracted using the Qiagen PowerSoil Kit following the manufacturer's protocol. The extracted DNA samples were quantified using a NanoDrop spectrophotometer, ensuring all samples passed quality control before proceeding with library preparation.

2.6. 16S metagenomic sequencing library preparation

The 16S ribosomal RNA gene, approximately 1500 base pairs in length, was selected for metagenomic analysis due to its variable regions, which facilitate phylogenetic classification of microbial populations. For this study, the V3 and V4 regions of the 16S rRNA gene were targeted, as these regions are commonly used in taxonomic identification down to the genus and species levels. The amplicon library was prepared following Illumina's protocol for the MiSeq system.

Amplicon Primer Design: Primers targeting the V3 and V4 regions were designed to produce a single amplicon of approximately 460 bp. Overhang adapter sequences were appended to the primer pairs to enable compatibility with Illumina index and sequencing adapters. The V3 and V4 regions were amplified, and Illumina sequencing adapters along with dual-index barcodes were attached using limited-cycle PCR. With Nextera XT indices, up to 96 libraries were pooled for sequencing in a single run. The libraries were sequenced on the Illumina MiSeq platform using 300-bp paired-end reads with MiSeq v3 reagents. By overlapping the ends of each read, high-quality, full-length reads of the V3 and V4 regions were obtained. The MiSeq run produced over 20 million reads, yielding more than 100,000 reads per sample—sufficient for comprehensive metagenomic analysis.

Following sequencing, the MiSeq system's onboard software, MiSeq Reporter (MSR), or the cloud-based BaseSpace platform, was used for secondary analysis. The Metagenomics workflow within MSR or BaseSpace facilitated taxonomic classification using the Greengenes database, enabling the identification of organisms at various taxonomic levels (kingdom, phylum, class, order, family, genus, and species). Data analysis was performed using the 16S Metagenomics tool from BaseSpace Onsite, with operational taxonomic units (OTUs) assigned based on the Human Oral Microbiome Database (HOMD).

2.7. Library clean-up and quantification

PCR products were purified with Agencourt AMPure XP beads (Beckman Coulter Genomics) according to the manufacturer's protocol. Dual indices and Illumina sequencing adapters were attached using 5 μL of amplicon PCR product, combined with Nextera XT Index primers (5 μL each of N7xx and S5xx), 25 μL of 2x KAPA HiFi HotStart Ready Mix, and 10 μL of PCR-grade water. The thermocycling conditions included an initial denaturation at 95°C for 3 min, followed by eight cycles of 95°C for 30 s, 55°C for 30 s, and 72°C for 30 s, with a final extension at 72°C for 5 min.

Library quantification was carried out using a fluorometric quantification method with dsDNA binding dyes, following Illumina's recommendations. The libraries were then normalized to a concentration of 4 nM and pooled by allotting 5 μL from each sample. Depending on coverage requirements, up to 96 libraries were pooled for a single MiSeq run, providing >100,000 reads per sample to ensure a thorough metagenomic survey of bacterial composition.

2.8. Library denaturation and MiSeq loading

Before cluster generation and sequencing, pooled libraries were denatured with NaOH, diluted with a hybridization buffer, and heat-denatured. To maintain sequencing quality, a minimum of 5% PhiX control was added to each run, as recommended by Illumina for low-diversity libraries. MiSeq v3 reagent kits were used to enhance run metrics.

2.9. Data analysis

The MiSeq system performed on-instrument secondary analysis using the Metagenomics workflow in MiSeq Reporter (MSR). This workflow classified organisms in the V3 and V4 amplicon regions using the Greengenes database. The data output provided a detailed taxonomic classification, including kingdom, phylum, class, order, family, genus, and species levels. Statistical analysis was conducted on the individual bacterial frequency distributions, and intergroup comparisons were made using the Chi-square test, with a significance level set at P < 0.05 using IBM SPSS® version 20.0 software.

In this study, presence versus absence was defined based on detection in the sequencing data. A taxon was considered present if it showed any detectable signal (non-zero reads) after quality filtering and taxonomic assignment, and absent, if no reads were detected. Thus, presence–absence reflects detection versus non-detection, and not, graded expression levels.

Because of this binary definition, the heatmap analyses do not test whether expression or abundance levels are statistically different between groups. The observed differences are therefore differences in prevalence (how often a taxon is detected) rather than differences in quantitative expression or relative abundance. While a taxon may be frequently detected in one group and rarely detected in another, this does not necessarily imply that its expression level, when present, is significantly different. Accordingly, the results are interpreted as evidence of community-level shifts in microbial occurrence, consistent with a dysbiosis model, rather than as statistically significant differences in microbial expression or abundance levels.

3. Results

The sequencing output for this study provided reliable information on taxon detection across samples, enabling presence–absence and prevalence-based analyses. However, numerical abundance or read-count data suitable for robust normalization and differential abundance testing were not available from the sequencing pipeline. Therefore, we focused on prevalence-based heatmap analyses to capture how consistently specific taxa are detected across periodontal groups, rather than making quantitative claims about expression or abundance. This conservative approach avoids biases related to sequencing depth and compositional effects, and is appropriate for describing community-level microbial shifts associated with periodontal disease.

Whole-saliva sequencing revealed distinct prevalence patterns for Neisseria, Dialister, and Filifactor across periodontal health and disease. Neisseria spp. were common in all groups, but several species declined markedly in advanced disease (Table 1). N. subflava was detected in 80% of healthy and nearly all gingivitis/recession samples but only 30% of periodontitis samples (p = 0.001). Likewise, N. flava was ubiquitous in healthy saliva (100%) and frequent in periodontitis (90%) but absent in gingival recession and gingivitis (0% in both, p = 0.001). N. elongata peaked in both recession and gingivitis (90%) but fell to 30% in periodontitis (p = 0.015). N. polysaccharea showed an inverse pattern: low in health (20%), high in both recession and gingivitis (90%), and low again in periodontitis (10%, p = 0.001). By contrast, other Neisseria (e.g., N. sicca, N. flavascens, N. macacae, N. mucosa, N. perflava, N. bacilliformis, N. oralis, N. meningitidis, N. lactamica) showed no significant differences across groups (p > 0.05). These data indicate that several commensal Neisseria species are significantly diminished in severe periodontal disease (Table 1, Fig. 1).

Table 1.

Frequency distribution of Nisseria species across the study groups.

Organisms	Health %	Gingival Recession %	Gingivitis %	Periodontitis %	Fisher's Exact statistic	P value
Nisseria Elongeta	60	90	90	30	10.45	0.015∗
Nisseria sicca	30	0	0	10	4.85	0.167
Nisseria subflava	80	100	100	30	15.86	0.001∗
Nisseria flavescens	80	90	90	60	3.26	0.45
Nisseria macacae	100	100	90	90	2.214	1.0
Nisseria polysaccharea	20	90	90	10	22.90	0.001∗
Nisseria mucosa	100	100	90	80	3.14	0.59
Nisseria flava	100	0	0	90	39.72	0.001∗
Nisseria perflava	80	90	90	70	1.85	0.80
Nisseria bacilliformis	30	80	60	30	6.96	0.087
Nisseria Oaralis	10	30	20	20	1.37	0.94
Nisseria meningitidis	60	20	20	40	4.85	0.23
Nisseria lactamicia	100	100	90	90	2.214	1.0

Fig. 1.

Distribution of 11 Neisseria in Health, Gingivitis, Periodontitis, and Gingival Recession across the study groups.

Dialister spp. were rare in health. D. pneumosintes showed a striking association with disease: it was detected in only 30% of healthy samples versus 100% of recession, 90% of gingivitis, and 80% of periodontitis samples (p = 0.002). In contrast, D. invisus and D. micraerophilus remained infrequent and did not differ significantly by group (Table 2, Fig. 2). Thus, D. pneumosintes prevalence rose sharply in all disease states, whereas the other two species showed no clear pattern.

Table 2.

Frequency distribution of Dialister species across the study groups.

Organisms	Health %	Gingival Recession %	Gingivitis %	Periodontitis %	Fisher's Exact statistic	P value
Dialister invisus	30	0	10	40	5.90	0.14
Dialister pneumosintes	30	100	90	80	13.30	0.002∗
Dialister micraerophilus	10	0	10	20	0.96	1.0

Fig. 2.

Distribution of Dialister species in Health, Gingivitis, Periodontitis, and Gingival Recession across the study groups.

Filifactor alocis was strongly linked to disease (Table 3, Fig. 3). It was found in only 30% of healthy samples but was present in 90% of gingivitis and recession samples and 80% of periodontitis samples (p = 0.011). This difference (healthy vs. diseased groups) was statistically significant. No other Filifactor species were detected.

Table 3.

Frequency distribution of Filifactor species across the study groups.

Organisms	Health %	Gingival Recession %	Gingivitis %	Periodontitis %	Fisher's Exact statistic	P value
Filifactor alocis	30	90	90	80	10.52	0.011∗

Fig. 3.

Distribution of Filifactor species in Health, Gingivitis, Periodontitis, and Gingival Recession across the study groups.

The high prevalence of F. alocis in diseased saliva underscores its emergence as a periodontitis-associated pathogen. Thus, the disease groups showed a significant shift: health-associated commensals (e.g. several Neisseria spp.) were depleted in periodontitis, whereas obligate anaerobes (D. pneumosintes, F. alocis) were enriched.

3.1. Heatmap-based microbial profiling and visualization

Salivary microbial taxa were profiled using next-generation sequencing and annotated via the Human Oral Microbiome Database (HOMD). To visualize distribution and disease-associated patterns, a multi-tiered heatmap approach was employed. First, a presence–absence heatmap at the individual sample level captured inter-individual variability using binary encoding, avoiding abundance-based normalization. Next, group-wise frequency heatmaps summarized the percentage of samples per clinical group (healthy vs. periodontitis), highlighting prevalence differences while minimizing individual-level noise. Finally, a targeted dysbiosis heatmap focused on taxa with the largest prevalence differences between groups; values were Z-score normalized to illustrate relative enrichment and depletion. All heatmaps were generated using Python's Seaborn library (Fig. 4, Fig. 5, Fig. 6).

Fig. 4.

Presence–absence heatmap illustrating the distribution of Neisseria, Dialister, and Filifactor species across individual salivary samples. Rows represent bacterial species and columns represent individual samples grouped by clinical periodontal status. Presence is indicated by colored cells, while absence is indicated by blank cells. The heatmap highlights inter-individual variability and provides an overview of species-level distribution without imposing abundance-based assumptions.

Fig. 5.

Group-wise frequency heatmap showing the prevalence (%) of Neisseria, Dialister, and Filifactor species in healthy and periodontitis groups. Values represent the percentage of samples within each clinical group in which a given species was detected. The heatmap highlights depletion of health-associated Neisseria species and enrichment of disease-associated anaerobic taxa in periodontitis.

Fig. 6.

Targeted dysbiosis heatmap. The Z-score–normalized heatmap of discriminatory salivary taxa between healthy and periodontitis groups. Red indicates relative enrichment and blue indicates relative depletion within each clinical group. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

4. Discussion

The pathogenesis of periodontal disease is the result of a complex interaction between a dysbiotic microflora and the immune-inflammatory response of the host, which ultimately leads to the destruction of the attachment apparatus of the tooth. The entire microorganisms present in the subgingival plaque constitute the microbiome of that environment. This is further divided into a core microbiome, which is consistent across individuals, and a variable microbiome, which evolves based on the phenotypic, genotypic, and environmental factors.

The polymicrobial synergy and dysbiosis (PSD) model has been recently hypothesized to implicate the role of microorganisms in the pathogenesis of periodontal disease. As the composition of a microbial community begins to turn from that of a synergistic one to a dysbiotic community due to a multitude of perturbations that have been brought about in the community, this leads to the sequence of events leading to the destruction of the periodontium. So it becomes inevitable to analyse the microbial community as a whole to understand its role in the progression and management of periodontal disease.

The salivary flow is a mechanism by which periodontopathic microorganisms are translocated from various oral niches. Saliva contains bacteria from various ecological niches of the oral cavity, such as the supragingival and subgingival plaque, tongue, tooth surface, and oral mucosa. Consequently, salivary testing has been adopted for both qualitative and quantitative assessment of periodontopathogens, serving as a predictive measure of periodontitis risk.^22,23 It has been stated that the whole saliva and tissues of the periodontium tend to harbor relative levels of pathogens and higher salivary counts of periodontal pathogens, which imply risk of acquiring periodontitis.²⁴

Various methods to identify and study the microbes have been used since the time they were implicated in the pathogenesis of various infectious diseases. But newer open-ended technologies seem to have ushered in a whole new paradigm in the study of infectious diseases. In this study, too, Next Generation Sequencing (NGS) was utilized to analyse the salivary presence of microorganisms, viz. Neisseria, Dialister, and Filifactor alocis. These Gram-negative bacteria are common colonizers of mucosal surfaces and are also found in the oral cavity of humans.^{25, 26, 27} Despite the growing body of literature, there remains limited knowledge regarding the habitats, clinical prevalence, and pathogenic potential of these bacteria in the context of periodontal disease, particularly in conditions of gingival recession. This study aimed to bridge this knowledge gap by examining the salivary distribution of Neisseria, Dialister, and Filifactor species in individuals with healthy periodontium and those with gingivitis, chronic periodontitis, and gingival recession.

In the genus Neisseria, N. macacae, N. mucosa, and N. lactamica were almost ubiquitously present across all four groups, suggesting that they could be a part of the core microbiome irrespective of the periodontal conditions and a commensal in the periodontal environment. On the contrary, the species N. sicca and N. oralis were detected only in negligible numbers across all study groups, suggesting that they are unlikely to constitute a significant component of the periodontal biofilm. The gingivitis group exhibited a distinct salivary microbial profile compared to the healthy group, with all nine Neisseria species detected at varying frequencies. In particular, N. elongata, N. macacae, N. mucosa, N. flava, and N. lactamica were found in 100% of gingivitis samples, followed by N. subflava, N. flavescens, and N. perflava in 80% of samples. In the periodontitis group, N. macacae, N. flava, and N. lactamica showed the highest prevalence (90% detection frequency). Vieira Colombo AP et al., in their study evaluating periodontal-disease-associated biofilm, have demonstrated that the most prevalent species in the biofilm were Neisseria, along with certain others, namely, peptostreptococcus anaerobius, Enterobacteria, Clostridium difficile, Candida albicans, and so on (>40%). In addition, their study predominantly detected Neisseria species in the gingivitis and chronic periodontitis groups compared to the control, which was in accordance with our study results.²⁸ Similarly, Yamashita and Meuric et al. have also reported that a predominance of Neisseria correlates with periodontal health.^12,29 Changchang Ye et al. similarly found Neisseria to be negatively associated with bleeding on probing (a disease indicator) and reduced in periodontitis samples.³⁰ Meanwhile, in the gingival recession group of our study, N. subflava, N. macacae, N. mucosa, and N. lactamica were present in all samples. To the best of our knowledge, this is the first study to evaluate these Neisseria species in association with gingival recession conditions and to compare them with health and periodontitis. Our data extend these findings: species like N. subflava and N. elongata were more common in health than in periodontitis. This supports the idea that commensal Neisseria confer resilience to the oral ecosystem, and their loss may permit overgrowth of pathogens.

For Dialister species, which include D. invisus, D. pneumosintes, and D. micraerophilus, variations were observed across groups. In the healthy group, D. invisus and D. pneumosintes were each detected in 30% of samples, while D. micraerophilus was present in 10% of samples. Although there were no significant differences in the presence of D. invisus and D. micraerophilus across all the clinical groups, D. pneumosintes was significantly higher in all the disease groups compared to health. A previous study by Shaikh et al., in 2024, reported significantly higher prevalence of D. pneumosintes in patients with chronic periodontitis compared to healthy individuals.³¹ It has been postulated that one of the key virulence factors of D. pneumosintes is the presence of lipopolysaccharides in its cell wall, which can stimulate the production of proinflammatory cytokines and matrix metalloproteinases, thereby contributing to the breakdown of periodontal tissues and the resorption of alveolar bone.³² Moreover, Vieira Colombo AP et al. have demonstrated significantly higher levels of D. pneumosintes in chronic periodontitis in comparison with health, gingivitis, and aggressive periodontitis.²⁸

Filifactor alocis was present in all groups but with varying prevalence. There was a significant difference in F. alocis distribution between healthy and all the disease groups. These findings strongly indicate that F. alocis is associated with periodontal disease progression. Similarly, a previous study has demonstrated higher levels of Filifactor. alocis in periodontitis groups compared to health.³¹ F. alocis exhibits distinct virulence traits, including the release of extracellular vesicles, the presence of lipoteichoic acid, resistance to oxidative stress, and secretion of proteases, all of which collectively stimulate a pro-inflammatory cytokine response in periodontal tissues. In addition, F. alocis is capable of invading epithelial cells like Porphyromonas gingivalis, either through filopodial extensions or vesicle-mediated uptake. As a strict anaerobe, F. alocis shows a preference for colonizing deeper periodontal sites and is frequently detected in the middle and apical thirds of periodontal pockets, where the low-oxygen environment favors its growth ^{33, 34, 35, 36, 37}.

Broader shifts in the salivary microbiome also support our specific findings. Periodontal disease is associated with a transition from a Streptococcus/Neisseria-dominated community towards one, enriched in anaerobes. The bacteria Porphyromonas, Treponema, Tannerella (the “red complex”), and other anaerobes often rise in periodontitis.³⁰

Additionally, in this study, to explore microbial community structure and its association with periodontal status, heatmap-based analyses were performed at multiple levels of resolution, encompassing individual-level presence–absence patterns, group-wise prevalence differences, and targeted dysbiosis visualization. The presence–absence heatmap highlighted substantial inter-individual variability in microbial composition, while also revealing a broadly conserved framework of shared taxa across healthy and diseased participants. Building on this, group-wise frequency heatmaps aggregated presence–absence data within clinical groups, reducing individual-level noise and highlighting coordinated shifts in taxa preferentially associated with health or periodontitis. Finally, a targeted dysbiosis heatmap focused on taxa showing the largest differences in prevalence between groups; Z-score normalization illustrated relative enrichment and depletion, revealing a clear reciprocal pattern between health-associated and disease-associated microbes. Collectively, these multi-tiered analyses provide a comprehensive visualization of salivary microbial dysbiosis, demonstrating that periodontal disease is characterized by coordinated community-level shifts rather than isolated changes in individual taxa, thereby reinforcing a dysbiosis-driven model of pathogenesis.

It is important to recognize that non-detection of a species in salivary metagenomic data reflects limits of sequencing detection rather than true biological absence, while taxa detected at very low abundance likely represent minor ecological contributors. While the presence of bacteria in saliva does not directly confirm their etiopathogenic role in specific periodontal sites, salivary microbial profiles can serve as valuable screening tools for identifying patients at risk for future periodontal disease activity. The findings from this study highlight the potential of certain bacterial species as biomarkers for periodontal disease progression, particularly Neisseria lactamica, N. macacae, N. mucosa, N. flava, Dialister pneumosintes, and Filifactor alocis. Further longitudinal studies with larger sample sizes are necessary to validate these results and clarify the potential role of these bacterial species in periodontal health and disease. Such studies could contribute to the development of enhanced risk management strategies, preventive measures, and patient education aimed at reducing periodontal disease onset and progression.

5. Conclusion

Key commensal Neisseria species dropped out as disease severity increased, while D. pneumosintes and F. alocis became predominant. These results mirror independent studies and reinforce the role of these taxa in periodontal pathology. Monitoring such shifts: ‘from Neisseria-rich to Filifactor/Dialister-rich profiles’, may have diagnostic value. Future studies could build on this by integrating functional metagenomics or longitudinal data to track how these microbial changes progress with disease and treatment.

Ethical clearance for Journal of Oral Biology & Craniofacial Research

Ethical clearance was obtained from the Institutional Review Board.

Patient's/Guardian's consent form for Journal of Oral Biology & Craniofacial Research

Patient consent was obtained before the start of the study.

Source of funding for Journal of Oral Biology & Craniofacial Research

Nil.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.