Alterations in the feline oral microbiome in common oral diseases - A comprehensive review

Abstract

The most prevalent diagnostic conditions in domestic cats (Felis catus) are oral diseases, affecting up to 90% of older cats. Periodontal disease (PD), feline chronic gingivostomatitis (FCGS), and tooth resorption (TR) are the principal clinically relevant entities, each with distinct histopathological and microbiological features. Certain molecular techniques, including 16S rRNA sequencing, shotgun metagenomics, and metatranscriptomics, have substantially advanced our understanding of the feline oral microbiome alterations. This review summarizes the findings of the healthy oral microbiome and its disease-associated shifts in PD, FCGS, TR, and feline immunodeficiency virus (FIV)-associated pathology. The healthy oral cavity is dominated by Proteobacteria, Bacteroidota, Bacillota, Fusobacteria, and Actinobacteria, notably Porphyromonas, Moraxella, Capnocytophaga, and Fusobacterium. Dental disease is characterised by expansion of Bacteroidota and Spirochaetota, enrichment of Treponema, Peptostreptococcus, Filifactor, and Fusobacterium nucleatum, and depletion of commensals. The contributions of fungi, viruses, and host immunity are critically evaluated, alongside the development of microbiome-based diagnostics and therapeutics. We argue that dysbiosis is a unifying (albeit not monocausal) feature of feline oral pathology and identify gaps in current knowledge that require further investigation.

1. Introduction

Oral diseases are the single most prevalent group of health conditions encountered in feline veterinary practice. Cross-sectional surveys consistently report a prevalence between 50 and 90% in cats older than three years, with the variability in differences in case definition, the inclusion or exclusion of subclinical lesions, and whether assessments are performed under general anaesthesia with full-mouth dental radiography [1–3]. The three most clinically significant feline oral diseases are periodontal disease (PD), feline chronic gingivostomatitis (FCGS), and tooth resorption (TR; previously termed feline odontoclastic resorptive lesion, FORL). These conditions differ markedly in their pathobiology yet share a common dependence on the oral microbiota for initiation, perpetuation, or both [4]. PD alone affects about 70% of cats by three years of age and is the most commonly diagnosed disease in feline practice, regardless of breed or geographic region [2, 3].

The oral cavity hosts one of the most complex ecosystems in the mammalian body, following the colon. The oral microbial ecosystem combines a continuously shedding mucosa with a hard-tissue surface of teeth on which biofilms can develop for prolonged periods [5]. This unique anatomy has profound ecological consequences: dental biofilms support certain communities with stratified oxygen gradients, syntrophic metabolic exchanges, and quorum-sensing-mediated coordination. These processes are essentially impossible on continuously desquamating epithelial surfaces [6, 7]. The oral microbiome, comprising bacteria, archaea, fungi, viruses (including bacteriophages), and protozoa, is therefore not a single community but a set of overlapping, site-specific microbial assemblages that interact with each other and with the host [5, 8].

Under homeostatic conditions the oral microbiome forms a symbiotic relationship with the host and contributes to colonization resistance through competitive exclusion, immune education, mucosal barrier regulation, and metabolic processing of dietary and host-derived substrates [6]. Dysbiosis is not a single pathogen overgrowth but a complete microbiota reorganization in which certain taxa exert elevated influence and reshape the dental ecosystem [7]. The polymicrobial synergy and dysbiosis (PSD) model reported by Hajishengallis and Lamont (2012) for human periodontitis posits that key pathogens, such as Porphyromonas gingivalis, subvert host immune responses and allow pathogens to expand, combining their virulence [7]. Currently, a similar pathogen spread, with unique species-level modifications was reported in cats [9–11].

Recently, novel molecular techniques have transformed the field. Sequencing of cloned 16S rRNA gene amplicons, followed by 454 pyrosequencing and now Illumina-based short-read and PacBio long-read sequencing, has further expanded the diversity of the feline oral microbiota. Previously, only a few dozen taxa were detected by anaerobic cultures, which now expanded to over 200 species-level operational taxonomic units (OTUs) [5, 8, 12]. Metagenomic as well as metatranscriptomic approaches transformed the field beyond taxonomic enumeration, identifying metabolic pathways differentially encoded or expressed in certain dental diseases [13]. Still, the field remains heterogeneous: studies vary in the selection of hypervariable region (V1-V3, V3-V4, V4, or V1-V9), DNA extraction protocols, primer sets, sampling site, bioinformatic pipeline, and reference database, contributing to the variability between these studies [5, 8, 14].

Still the feline oral microbiome is less investigated than the human or canine oral microbiome. The feline oral microbiome is phylogenetically distinct from both; for example, Porphyromonas gulae rather than P. gingivalis is the dominant Porphyromonas species, and the relative abundance of Proteobacteria (renamed Pseudomonadota under the 2021 International Code of Nomenclature of Prokaryotes (ICNP) revision in healthy cats appears higher than in either humans or dogs [5, 9, 10]. These inter-species differences indicate that a species-specific research and therapeutic strategies are needed. Furthermore, the contributions of non-bacterial components such as the mycobiome, the virome, and the archaeome are remain to be explored in cats [15].

This narrative review includes the peer-reviewed literature on changes in the feline oral microbiome associated with oral disease. We describe the composition of the healthy feline oral microbiome, including the influences of diet, sampling site, age, and methodology, before considering in turn the dysbiotic shifts associated with PD, FCGS, TR, and FIV-associated oral pathology. We then examine the roles of the mycobiome, the virome, and the host immune system, and conclude with an evaluation of microbiome-based diagnostic and therapeutic strategies, including the limitations and future directions. Moreover, Table 1 provides an overview of the reported bacterial and fungal taxa in association with feline oral health and disease.

Table 1.

Bacterial and fungal taxa reported in association with feline oral health and disease states. Direction (↑/↓) is given relative to clinically healthy controls

Taxon	Methodology	Direction	Functional/virulence relevance	Refs
HEALTH (BASELINE)
Bacterial taxa
Phylum Proteobacteria (∼33%); Bacteroidetes/Bacteroidota (∼25%); Firmicutes/Bacillota (∼18%)	16S V4 (Illumina); near full-length Sanger	●	Predominant phyla in feline oral cavity; Proteobacteria relative abundance higher than in healthy humans or dogs	[5, 9, 12]
Genera Moraxella, Capnocytophaga, Bergeyella, Neisseria, unclassified Pasteurellaceae	16S, multiple regions	●	Aerobic/microaerophilic commensals; competitive exclusion of obligate anaerobes; immune education	[5, 8, 9, 12]
Fusobacterium spp. (genus level)	16S	●	Commensal at genus level in health; species-specifically (F. nucleatum) enriched in PD/FCGS - see disease sections	[5, 8, 9, 12]
Porphyromonas spp. (genus level)	16S; species-specific PCR	●	Genus contains both commensals and putative keystone pathogens; P. gulae † assumes pathogenic role in disease	[5, 9, 10, 12]
Capnocytophaga canimorsus (most prevalent species in clinically healthy cats)	16S species-level	●	Commensal in cats; recognised zoonotic agent of fulminant sepsis in immunocompromised humans following cat bites	[15]
Fungal taxa
Saccharomyces cerevisiae	ITS amplicon	●	Most prevalent fungal species in healthy cats; non-pathogenic; depleted in FCGS (see below)	[15]
PERIODONTAL DISEASE (PD)
Bacterial taxa — concordant findings
Phylum Bacteroidetes/Bacteroidota (gradient: HEALTH ∼25% → CP 33.2% → AP 35.6% → FCGS 40.6%)	16S V3–V4; pyrosequencing	↑	Anaerobic Gram-negative; many proteolytic; central to dysbiotic “bloom” model rather than diversity collapse	[9, 14, 17]
Phylum Spirochaetes/Spirochaetota	16S	↑	Motile, tissue-invasive spirochetes; activate TLR2-dependent inflammatory responses	[9, 17]
Phylum Proteobacteria (gradient: HEALTH 32.9% → CP 21.3% → AP 13.4% → FCGS 9.8%)	16S	↓	Progressive loss of aerobic/microaerophilic commensals across disease severity	[9]
Peptostreptococcus, Filifactor, Peptococcus	16S pyrosequencing	↑	Gram-positive anaerobic asaccharolytic proteolytic genera; produce cytotoxic SCFAs (butyrate, propionate), polyamines, ammonia	[17]
Treponema spp. (incl. T. denticola †)	16S; species-level resolution limited	↑	Tissue-invasive spirochetes; chymotrypsin-like dentilisin protease; surface lipoproteins activate TLR2	[9, 17]
Porphyromonas gulae † (principal feline Porphyromonas periodontal pathogen; ∼70% of affected cats)	qPCR; 16S	↑	Gingipain orthologues (Kgp, RgpA/B); FimA fimbriae; immunomodulatory LPS; keystone-pathogen role per PSD model	[10, 11]
“Porphyromonas gingivalis” (PCR-positive in 96% of affected cats - but species-level discrimination from P. gulae technically demanding; result likely reflects cross-detection)	species-specific PCR	↑	Contested: most authorities now attribute the high apparent positivity rate to P. gulae; species identity uncertain pending validated, target-specific primers	[10]
Prevotella nigrescens	species-specific PCR	↑	Black-pigmented anaerobe; proteolytic; periodontal pathogen of humans and animals	[10]
F. nucleatum	16S	↑	Bridging organism between early and late biofilm colonisers; pro-inflammatory; well-characterised role in human periodontitis	[9, 17]
Contested/heterogeneous findings
Ochrobactrum, Odoribacter denticanis, T. denticola, Porphyromonas macacae, Fretibacterium fastidiosum	16S V3–V4 (Illumina)	↓	Apparent inverse pattern relative to earlier studies; possibly reflects compositional-data effects, sampling site, severity stratification, geography, or case definition (see Section 3.2)	[14]
FELINE CHRONIC GINGIVOSTOMATITIS (FCGS)
Bacterial taxa
Phylum Bacteroidetes/Bacteroidota (highest of any disease group, 40.6%)	16S	↑	See PD entry; FCGS represents the most extreme expansion of this phylum	[8, 9]
Phylum Proteobacteria (lowest of any disease group, 9.8%)	16S	↓	Most extreme depletion of health-associated commensal phylum across all feline oral diseases studied	[8, 9]
Treponema spp.	16S	↑	See PD entry; consistent enrichment across multiple cohorts and geographies	[9, 17, 20]
Fusobacterium spp. (incl. F. nucleatum)	16S	↑	Bridging/pro-inflammatory role; documented in Chinese FCGS cohort (Dai 2024)	[9, 20]
Family Peptostreptococcaceae; Filifactor spp.; anaerobic Gram-negative bacteria	16S	↑	Proteolytic; polyamine and ammonia production; metabolic phenotype unfavourable to mucosal homeostasis	[9, 17, 20]
Pasteurella multocida subsp. multocida	16S; culture-confirmed	↑	Paradoxical enrichment: commensal of healthy oropharynx (70–90% prevalence) yet enriched in FCGS - possibly opportunistic expansion under local immune dysregulation; PMT toxin, capsular polysaccharide, LPS as candidate virulence factors	[20]
Moraxella spp.; family Pasteurellaceae	16S	↓	Health-associated commensals lost in FCGS - concomitant with Bacteroidetes expansion	[8, 9]
Multi-site dysbiosis across ten oral sites (not localised to lesions)	full-length 16S V1-V9 (PacBio)	site-dependent	Indicates whole-cavity ecosystem perturbation, not lesion-restricted; supports rationale for full-mouth or premolar/molar extraction therapy	[8]
Fungal taxa
S. cerevisiae	ITS	↓	Loss of dominant healthy mycobiome member	[15]
Malassezia restricta, M. arunalokei	ITS	↑	Lipophilic basidiomycetous yeasts; obligate lipid scavengers; secrete lipases/phospholipases liberating free fatty acids and indoles; in skin disease activate aryl hydrocarbon receptor and induce IL-6, IL-17, TSLP - analogous oral mechanism plausible	[15]
Cladosporium penidielloides/salinae	ITS	↑	Environmental fungi; pathogenic role in oral cavity unclear	[15]
Family Aspergillaceae	ITS	↑	Includes potential opportunistic pathogens; mechanistic significance in FCGS pending	[15]
TOOTH RESORPTION (TR/FORL)
Bacterial taxa
Six most abundant genera in lesion samples: Bergeyella, Capnocytophaga, Lampropedia, Moraxella, Porphyromonas, Treponema	16S	mixed	Three of six (Bergeyella, Capnocytophaga, Moraxella) are also dominant health-associated commensals; lesion microbiome not dominated by classical periodontal pathogens to the same degree as PD/FCGS	[22]
Lampropedia spp.	16S	↑	Atypical Gram-negative; flat square cellular organisation; mineralisation-modulating metabolism (alters local Ca:P equilibria); proposed but unproven mechanistic link to odontoclastic activity	[22]
Distinct microbial signatures in FORL-1 vs FORL-2 clusters	16S, unsupervised cluster analysis	n/a	Possible parallel to Type 1/Type 2 histopathological dichotomy; suggests subtype-specific microbial environments - hypothesis-generating, requires validation	[22]
TR microbiome distinguishable from PD and FCGS in beta-diversity ordinations (with overlap)	full-length 16S V1-V9	n/a	Disease-specific microbial signatures support potential for microbiome-based diagnostics; awaits prospective validation	[8]
FIV-ASSOCIATED ORAL DISEASE
Microbial and host findings
Multi-site oral dysbiosis; cART-treated cats more closely resemble uninfected controls than untreated FIV + cats	16S, longitudinal (4 time points)	n/a	Restoration of CD4+ T-cell function, mucosal Th17/IL-17 axis, secretory IgA, and innate antimicrobial peptides plausibly underlies microbiome partial normalisation	[27]
Increased prevalence of opportunistic colonisers, including Candida spp.	fungal screening; clinical observation	↑	Parallels human HIV/AIDS oral candidiasis; reflects compromised mucosal Th17-IL-17 axis required for Candida control	[18, 27]
Higher gingival inflammation in untreated FIV + vs cART-treated and uninfected controls	clinical scoring +16S	n/a	cART partially reverses both microbiome- and immunity-mediated oral inflammation; supports causal role of immunodeficiency in oral dysbiosis	[27]

Symbols: ● = baseline/characteristic of healthy state; ↑ = relative-abundance enrichment in disease vs. health; ↓ = relative-abundance depletion in disease vs. health; † = putative keystone pathogen per the polymicrobial synergy and dysbiosis (PSD) model (Hajishengallis & Lamont 2012, [7]); n/a = not reported as a directional comparison.

Methodology abbreviations: 16S = 16S rRNA gene amplicon sequencing (specific hypervariable region indicated where known; “full-length” = V1-V9, PacBio or Sanger); ITS = internal transcribed spacer fungal amplicon sequencing; PCR/qPCR = species-specific polymerase chain reaction.

Notes: All directional calls derive from relative-abundance data; absolute abundances are largely unavailable. Inter-study comparison is constrained by variation in sampling site, hypervariable region, sequencing platform, OTU/ASV pipeline, reference database, and clinical case definition (see Section 12, Limitations). Phyla appear under both old and new GTDB-2021 names where applicable. Citations correspond to the manuscript reference list.

2. The healthy feline oral microbiome

2.1. Taxonomic composition and diversity

The foundational characterization of the feline oral microbiome was established by Dewhirst et al. (2015), who developed a provisional 16S rRNA-based taxonomy of 171 feline oral taxa, each represented by a near full-length reference sequence (∼1,500 bp) [5]. Their study, which combined Sanger sequencing of cloned amplicons, examined twenty domestic short-hair cats (ten clinically healthy and ten with periodontitis) and identified bacterial taxa distributed across ten phyla, with the bulk belonging to Proteobacteria, Bacteroidetes, Firmicutes, Spirochaetes, Fusobacteria, and Actinobacteria (now Actinomycetota) [5]. One third of the detected phylotypes did not match previously described species, indicating the unexplored diversity in the feline oral cavity. This manuscript provided a critical resource for subsequent feline-specific microbiome research.

Sturgeon et al. (2014) used Illumina-based V4 amplicon sequencing in healthy cats and identified Proteobacteria as the predominant phylum, followed by Bacteroidetes (now Bacteroidota) and Firmicutes (now Bacillota). The most abundant bacterial families were Pasteurellaceae, Moraxellaceae, and Porphyromonadaceae [12]. The dominant taxa were Porphyromonas, Moraxella, and Fusobacterium across the studies [8, 9, 12]. The differences in these studies were mainly in the DNA extraction, hypervariable region, and bioinformatic pipeline.

The high prevalence of Proteobacteria in the healthy feline oral cavity is a consistent characteristic of the feline microbiome. Rodrigues et al. (2019) reported that Proteobacteria constituted approximately one third of the microbiota in periodontally healthy cats, followed by Bacteroidetes and Firmicutes [9]. In addition, the predominant genera in the healthy otal cavity include Moraxella, Capnocytophaga, Bergeyella, Neisseria, and unclassified Pasteurellaceae [5, 9, 12]. Importantly, Capnocytophaga canimorsus, a recognised cause of severe sepsis in immunocompromised humans following animal bites, has been identified as the most prevalent species at the species level in clinically healthy cats [15]. Its high prevalence in the asymptomatic feline oral cavity has significant zoonotic implications, indicating that a conventional “pathogen/commensal” classification is rather context dependent.

When comparing results, some methodological issues need to be addressed. For example, the choice of hypervariable region affects how precisely organisms can be identified: V4 amplicons reliably identify bacteria at the family and genus level, but species-level identification is less consistent. Full-length V1-V9 sequencing with PacBio or Oxford Nanopore can resolve species, but it costs more per sample [8]. Differences in the bioinformatic clustering, for example OTU clustering, also affect diversity estimates. Databases (e.g. Greengenes, SILVA, Ribosomal Database Project (RDP), the Human Oral Microbiome Database, and the curated feline-specific dataset of Dewhirst et al.) differ in their coverage of feline-relevant taxa and can result in different classifications of the same sequence [5, 8, 14].

2.2. Influence of diet on the oral microbiome

Adler et al. identified in 2016 that diet has a measurable influence on the composition of the feline oral microbiome [16]. In this study cats were fed dry or only canned food, and the dry-food regimen resulted in higher bacterial alpha-diversity (Shannon index, P < 0.001). Furthermore, the abundances of Porphyromonas spp. (P < 0.01) and Treponema spp. (P < 0.01) were significantly elevated relative to wet-food-fed cats [16]. Possible explanations would be that the potential mechanical abrasion by the dry food may alter biofilm architecture; or the differences in carbohydrate and water content may affect the fermentative metabolic niches [16]. Finally, the salivary flow rates between the two feeding regimens may alter the stomach pH. However, the higher prevalence of Porphyromonas and Treponema in dry-food-fed cats does not indicate elevated disease risk, as many of those species are commensal in the feline oral cavity. Thus, the relationship between diet, microbial composition, and clinical disease is still not fully defined and a prospective longitudinal study would be necessary. These findings are in line with the previous literature in humans, linking dietary composition to oral microbial structure and caries risk. However, the strictly carnivorous diet characteristic of cats, the differences in pH levels measured in the oral cavity, and the differences in salivary amylase activity allow only for an indirect comparison.

2.3. Site-specific variations

Anderson et al. (2023) described that microbial community composition varies markedly between the oral sampling sites in cats, including the buccal mucosa, sublingual region, hard and soft palates, gingival sulcus, and various tongue and tonsillar locations [8]. These studies detected by full-length 16S rRNA gene sequencing (V1-V9), that microbial communities vary within these sites. Moreover, beta-diversity analyses revealed site-driven clustering, exceeding inter-individual variation. Comparable site-specificity has been described in the human oral cavity, where the communities are shaped by salivary exposure, mucin composition, and surface topography [5, 8]. When comparing studies investigating different sample sites (supragingival plaque, pooled buccal swabs, sulcular paper-point samples) it is important to consider this variable. Therefore, the indication of exact anatomical location, instrument type (brush or sterile swab), pooling strategy, and storage conditions, would substantially improve cross-study comparability.

2.4. Effects of age, breed, and other host factors

Although it has not yet been systematically studied, age also appears to exert a measurable influence on the feline oral microbiome. Dental biofilms accumulate progressively along the age, with an increase in the relative abundance of anaerobic and proteolytic taxa even in clinically healthy animals [5, 9, 17]. This age-associated alteration further complicates the interpretation of cross-sectional comparisons between younger healthy controls and older diseased cats, raising the possibility that the previously diagnosed disease-associated dysbiosis may be age-related. There has been only limited research conducted on different breeds, although the prevalence of feline chronic gingivostomatitis (FCGS) in certain breeds, such as the Maine Coon, Ragdoll, and Siamese suggests that genetics play also a role [18, 19]. Further environmental factors, including multi-cat household status, outdoor access, and exposure to feline upper respiratory tract pathogens should be also further evaluated.

3. Microbiome changes in periodontal disease

3.1. Overview of feline PD

The pathological changes in feline periodontal disease extend from gingivitis, which is reversible inflammation, to and irreversible periodontitis with a loss of the periodontal attachment apparatus, including alveolar bone and the periodontal ligament. [4, 17]. PD is the most common feline oral disease, with a prevalence near to 70% by a three-year-old to over 80% in older cats [2, 3]. The pathogenesis reflects a complex interplay between microbial biofilm accumulation, host immune and inflammatory responses, environmental factors (diet, husbandry, oral hygiene), and host-genetic factors [6, 7, 17]. Infiltrating neutrophil granulocytes, macrophages, and mononuclear cell subsets were detected during histopathological evaluation upon feline periodontitis. Moreover, a progressive collagen degradation, Receptor Activator of Nuclear factor κB (RANK)/RANK Ligand (RANKL)-mediated osteoclastogenesis, and apical migration of the junctional epithelium were described [4, 17].

3.2. Shifts in microbial community composition

Harris et al. (2015) described the first overall investigation of the feline subgingival microbiota across health, gingivitis, and mild periodontitis [17]. Using 16S rRNA sequencing, they investigated the plaque samples of 82 cats and identified 267 OTUs at 97% sequence identity [17]. The findings indicated a graded shift in community structure across the disease continuum, with increased presence of Bacteroidetes and Spirochaetota in diseased sites and corresponding contractions in Proteobacteria. Also, the Gram-positive anaerobic Peptostreptococcus and Filifactor, and Peptococcus were enriched in periodontitis compared with health [17]. These findings paralel the pattern of human periodontitis, in which proteolytic anaerobes derive energy from peptide and amino-acid catabolism, generating cytotoxic short-chain fatty acids, ammonia, and volatile sulphur compounds that drive tissue destruction and the characteristic halitosis [7].

Rodrigues et al. (2019) extended these observations in 44 cats stratified into four groups: periodontally healthy (n = 12), chronic periodontitis (CP, n = 11), aggressive periodontitis (AP, n = 9), and FCGS (n = 12) [9]. Proteobacteria decreased progressively from health (32.88%) to CP (21.27%), AP (13.42%), and FCGS (9.82%), with concomitant Bacteroidetes expansion in all three disease groups (CP: 33.17%; AP: 35.62%; FCGS: 40.56%) [9]. Counter-intuitively, alpha-diversity (Shannon, Chao1) was higher in disease groups than in healthy controls. This finding is the opposite of the pattern frequently observed in gut dysbiosis associated with inflammatory bowel disease, but consistent with several reports from human periodontitis [7]. Feline periodontal dysbiosis therefore appears to involve opportunistic colonisation by additional anaerobic taxa as the inflammatory environment becomes permissive, rather than simple loss of diversity, fitting a “bloom” rather than a “collapse” model of dysbiosis.

Thomson et al. (2025) used V3-V4 16S rRNA amplicon sequencing in a case-control design (15 healthy versus 15 periodontitis cats) and confirmed that Bacteroidota and Bacillota (the recently renamed Bacteroidetes and Firmicutes, following the 2021 Genome Taxonomy Database (GTDB)-driven taxonomic revisions) dominate both groups [14]. Interestingly, cats with periodontitis exhibited reduced relative abundances of several taxa, including Ochrobactrum, Odoribacter denticanis, Treponema denticola, Porphyromonas macacae, and Fretibacterium fastidiosum [14]. At first glance this contradicts earlier findings of Treponema enrichment in disease [9, 17], and warrants careful interpretation. Several non-exclusive explanations are possible: (I) the relative abundance of any taxon depends on the absolute abundances of all others, so a true increase in Treponema may be obscured by an even greater expansion of other taxa; (II) compositional data analysis methods (centred log-ratio transformation, Analysis of Compositions of Microbiomes with Bias Correction (ANCOM-BC)) are not yet uniformly applied across studies, and untransformed proportional comparisons can yield misleading inferences [14]; (III) sampling site, disease severity stratification, and case definitions differ between studies; and (IV) the geographic and dietary context of the study population may shape microbial composition. The Thomson study does not necessarily contradict earlier findings, but it does show that the feline periodontitis microbiome is not a fixed, stereotyped community: individual taxon abundances vary substantially between cohorts.

3.3. Functional implications of microbial shifts

Rodrigues et al. (2021) reported the first functional analysis of the feline subgingival microbiota, using Phylogenetic Investigation of Communities by Reconstruction of Unobserved States 2 (PICRUSt2) to predict metabolic capacity from 16S rRNA relative-abundance data [13]. Pathways for amino-acid metabolism, polyamine biosynthesis, and nitrogen metabolism were significantly enriched in periodontitis and FCGS, consistent with the proteolytic, asaccharolytic phenotype of dysbiotic communities [13]. Polyamines (putrescine, spermidine, cadaverine) are produced by bacterial decarboxylation of arginine, lysine, and ornithine, and contribute to the characteristic malodour of periodontitis. They also act as immunomodulators: they can induce apoptosis in epithelial cells, alter neutrophil function, and support biofilm formation [13]. Ammonia, a metabolic byproduct of urea hydrolysis and amino-acid deamination, increases the local pH, thereby favouring the persistence of species that are able to tolerate alkalinity. In contrast, cytotoxic short-chain fatty acids (butyrate and propionate) may impede fibroblast proliferation and can promote osteoclast differentiation, in part through inhibition of histone deacetylase activity.

PICRUSt2-based predictions are informative but inherently limited. They infer functional content from 16S rRNA-derived taxonomic profiles via reference genomes, and their accuracy is constrained by the phylogenetic distance between observed and reference taxa. This is particularly problematic for the many uncultured feline-oral taxa lacking a reference genome [13]. Validation and extension will require direct functional interrogation by shotgun metagenomics, metatranscriptomics, and metabolomics.

3.4. Specific periodontal pathogens in cats

The identification of specific periodontal pathogens in cats has been advanced by both targeted molecular detection studies and hypothesis-generating microbiome surveys. Ozavci et al. (2019) used species-specific PCR to detect periodontal pathogens in feline plaque samples and reported P. gingivalis in 96% of cats with periodontal disease, Prevotella nigrescens in 90%, and P. gulae in 70% [10]. The high prevalence of P. gingivalis should be interpreted with caution: species-level PCR detection within Porphyromonas is technically demanding because of close inter-species sequence similarity, and discrimination between P. gingivalis and P. gulae requires carefully validated, target-specific primers. Most current authorities consider P. gulae (not P. gingivalis) to be the principal Porphyromonas periodontal pathogen of cats and dogs [10, 11]. P. gulae expresses orthologues of the major P. gingivalis virulence factors, including the cysteine proteinases known as gingipains (Kgp, RgpA, and RgpB), the major fimbrial protein FimA, and lipopolysaccharide (LPS) variants with distinct Toll-like receptor (TLR) 4 agonist activity. These factors enable degradation of complement components, immunoglobulins, and host extracellular matrix proteins and manipulation of dendritic-cell and neutrophil signalling, activities that, in the P. gingivalis paradigm, support the keystone-pathogen role articulated in the polymicrobial synergy and dysbiosis model [7].

Booij-Vrieling et al. (2010) further showed that periodontal pathogens, including P. gulae, are sometimes shared between cats and their human owners, raising zoonotic considerations that remain under-investigated [11]. The bidirectional exchange of cats and humans microbiota is still not fully understood and has particular importance for immunocompromised owners.

T. denticola, a member of the so-called “red complex” in human periodontitis, has extensive virulence factors including the chymotrypsin-like protease dentilisin, motility apparatus enabling tissue penetration, and certain surface lipoproteins that can activate TLR2-dependent inflammatory responses [7]. In felines, Treponema spp. are often enriched in disease [9, 17], although species-level identification has been hindered by the limitations in their cultivation and the technical difficulty of resolving closely related Treponema phylotypes by short-read amplicon sequencing.

4. Microbiome changes in feline chronic gingivostomatitis

4.1. Clinical and pathological features of FCGS

FCGS is a severe oral inflammatory condition affecting an estimated 0.7–12% of cats, with marked geographic and breed-related variation [18, 20]. Clinically it is associated with ulcerative mucosal lesions, lateral to the palatoglossal folds (“caudal stomatitis” or “faucitis”), with frequent extension to the marginal gingiva, alveolar mucosa, buccal mucosa, and tongue [4, 18]. Affected cats develop chronic oral pain, dysphagia, ptyalism, weight loss, halitosis, and behavioural alterations (irritability, hiding, reduced grooming) and in advanced cases they are not able to eat [4, 18]. Histopathological examination indicates dense plasmacytic-lymphocytic infiltrates with neutrophil involvement, epithelial ulceration, and predominance of CD8⁺ cytotoxic over CD4⁺ helper T cells, a pattern consistent with a sustained antiviral or anti-intracellular-pathogen response [18, 21, 22].

4.2. Bacterial dysbiosis in FCGS

Converging evidence links FCGS to a multi-site oral dysbiosis [8, 9, 20, 23, 24]. It was previously shown that the microbial communities of cats with FCGS differed from those of healthy cats, indicating that FCGS dysbiosis extends throughout the oral cavity rather than being confined to the visible lesions [8]. The direct clinical implication is, that a surgical extraction of premolars and molars (currently the most effective therapy for refractory FCGS) may work not only by removing tooth-associated bacterial reservoirs but also by perturbing a microbial ecosystem reaching beyond the close dental area.

A study by Rodrigues et al. (2019) indicates that FCGS samples had the lowest Proteobacteria (9.82%) and highest Bacteroidetes (40.56%) relative abundances of any disease group studied [9]. In an epidemiological investigation conducted by Dai et al. (2024) in Xi'an, China, confirmed that the relative abundances of Treponema and Fusobacterium were greater in cats with FCGS when compared to healthy [20]. The FCGS sample exhibited an enrichment of taxa from the Peptostreptococcaceae family, Filifactor, and various anaerobic Gram-negative bacteria, while taxa associated with health, such as Moraxella and Pasteurellaceae, were depleted [8, 9, 17, 20]. Functionally, this remodelling is associated with increased proteolytic capacity, higher polyamine and ammonia production.

4.3. The role of Pasteurella multocida

Surprisingly, the P. multocida subsp. multocida was enriched in some FCGS samples [20]. P. multocida is usually considered a commensal of the feline oropharynx, with prevalence rates of 70–90% in healthy cats [25]. This bacterium is clinically relevant for its role as the primary causative agent of cat-bite wound infections in humans [25]. The predominance of this bacterium in FCGS may be indicative of either an opportunistic expansion upon local immune dysregulation and mucosal barrier breakdown, or an unrecognised pathogenic contribution. P. multocida has several virulence factors of potential relevance, including the dermonecrotic toxin (a deamidase that constitutively activates Gαq, Gαi, and Gα12/13 signalling) [25], and lipopolysaccharide variants, activating TLR4. Whether these factors directly contribute to FCGS, or are simply associated with the broader inflammatory milieu, remains unclear. Further functional studies using in vitro infection models would provide more valuable insights [20].

4.4. Multifactorial etiology of FCGS

A review by Sanchez-Vallejo et al. (2025) focused on the factors associated with FCGS and concluded that the disease is multifactorial, with no single sufficient cause [19]. When comparing the previous studies, microbial dysbiosis and viral infection, especially feline calicivirus (FCV), were the most consistently reported associated factors, alongside additional contributors such as genetic susceptibility, immune status, and chronic antigenic exposure [19]. This concept parallels the models of human oral mucosal disease, where genetic, environmental, microbial, and immune factors converge on a common pathway of dysregulated mucosal inflammation.

It is therefore necessary to distinguish between predisposing factors (host genetics, immune competence, breed), triggering factors (microbial dysbiosis, FCV or feline herpesvirus-1 infection, mechanical or chemical mucosal injury), and perpetuating factors (chronic immune activation, persistent antigenic stimulation, inadequate immune regulation). Ideally a novel effective therapy has to target all three layers, to avoid an incomplete or transient response.

4.5. Heterogeneity within FCGS

Clinically labelled “FCGS” probably encompasses a heterogeneous group of conditions sharing a common phenotype but differing in underlying mechanism. Clinical sub-classifications based on lesion distribution (gingivostomatitis with or without caudal involvement), histopathological pattern (plasmacytic-predominant versus mixed infiltrate), and viral status (FCV-positive versus FCV-negative) all suggest underlying mechanistic heterogeneity [18, 19, 24]. Pooling clinically heterogeneous cases under a single FCGS label may obscure genuinely distinct dysbiotic signatures, a limitation that only larger, more carefully phenotyped cohorts can resolve.

5. Microbiome changes in tooth resorption

5.1. Overview of feline TR

Tooth resorption is a progressive, painful, destructive condition affecting 20–75% of cats, with prevalence rising markedly with age [4, 26]. Histologically the disease shows progressive destruction of dental hard tissues (enamel, dentine, and cementum) by odontoclasts: multinucleated cells of the monocyte/macrophage lineage sharing core osteoclast biology (RANK/RANKL/Osteoprotegerin (OPG) signalling, integrin αvβ3, cathepsin K, tartrate-resistant acid phosphatase) but operate on dental rather than skeletal substrate [26]. The lesions are conventionally classified into Type 1 (inflammatory, associated with periodontal disease and characterised by preservation of the periodontal ligament and apparent radiographic distinguishability of the tooth from alveolar bone), Type 2 (non-inflammatory, with replacement resorption and progressive loss of the radiographic distinction between tooth and bone), and Type 3 (mixed). Despite decades of research, the aetiology of feline TR remains unclear. Proposed contributors include chronic inflammation, mechanical stress, vitamin D dysregulation, diet, and microbial factors. However, no single mechanism has been convincingly established for all forms [26].

5.2. Microbiome composition in TR

Thomas et al. (2021) reported the first 16S rRNA-based microbiome analysis of feline odontoclastic resorptive lesions [26]. These genera across FORL samples were Bergeyella, Capnocytophaga, Lampropedia, Moraxella, Porphyromonas, and Treponema [26]. Unsupervised cluster analysis indicated two distinct FORL groups (FORL-1 and FORL-2), with FORL-1 showing a microbial signature different from healthy controls as well as from FORL-2 [26]. This subgrouping was in line with the histopathological Type 1/Type 2 dichotomy, raising the possibility that the radiographic and histological subtypes correspond to mechanistically distinct microbial environments.

An important finding indicated the presence of Lampropedia species in TR samples [26]. Lampropedia is an atypical Gram-negative bacterium with a square cellular organisation that is able to eventually mineralise its environment by altering local calcium-phosphate equilibria. The hypothesis of Thomas et al., that Lampropedia metabolic activity may shift the local Ca:P ratio in a way that influences odontoclastic activity, remains speculative but is plausible and emerges further investigation [26]. Whether oral microbial communities are able to directly alter bone mineralisation and demineralisation, beyond the established acidogenic-cariogenic paradigm, has further implications beyond feline dentistry.

A study by Anderson et al. (2023) indicated that microbiome profile of teeth derived from cats with TR could be distinguished from those of cats with FCGS or periodontitis, supporting the view that each oral disease produces a distinct pattern of microbial alteration [8]. Beta-diversity analyses display a clear separation between disease groups in Principal Coordinates Analysis (PCoA) [8]. If confirmed in larger cohorts, disease-specific microbiome signatures would have practical implications for microbiome-based diagnostics: the subsequent profiling might distinguish between major oral disease entities.

6. Feline immunodeficiency virus and the oral microbiome

FIV is a lentivirus (family Retroviridae) genetically related to HIV-1; it establishes lifelong infection in domestic cats and produces a progressive immunodeficiency syndrome with depletion of CD4⁺ T-lymphocytes, inversion of the CD4/CD8 ratio, B-cell dysregulation, and impaired innate immunity [27]. Oral manifestations (gingivitis, periodontitis, FCGS, oral candidiasis) are among the most common clinical signs of FIV infection and affect up to 50% of FIV-positive cats, often before systemic disease becomes evident [18, 27]. FIV is therefore a naturally occurring large-animal model of retroviral immunosuppression with relevance to human HIV.

Bashor et al. (2025) examined the effect of combined antiretroviral therapy (cART) on the oral microbiome and periodontal health of FIV-positive cats by 16S rRNA sequencing at four longitudinal time points [27]. At the endpoint of the study, FIV-infected cats receiving placebo had significantly more gingival inflammation than cART-treated cats and uninfected controls [27]. cART ameliorated FIV-associated oral disease, which the authors attributed at least partly to preserved oral mucosal microbiota homeostasis [27]. Oral microbiome alpha-diversity increased in all groups across the study period (likely reflecting a maturation effect), but the rate and trajectory of community change differed between treatment groups, with cART-treated cats more closely resembling uninfected controls than untreated FIV-positive cats [27].

These findings parallel human HIV research, in which antiretroviral therapy partially restores oral microbiome composition and reduces (but does not eliminate) the increased prevalence of oral candidiasis, periodontitis, and oral hairy leukoplakia [27]. The mechanism is probably multifactorial: recovery of CD4⁺ T-cell numbers and function restores mucosal IL-17 production by Th17 cells, which is critical for control of Candida and the maintenance of mucosal barrier function; secretory IgA production by mucosal plasma cells is partially restored; and innate antimicrobial defense from oral epithelial cells (β-defensins, LL-37, calprotectin) are recovered. The FIV-cat system is therefore a valuable comparative model, not only for retrovirology, but also for studying how systemic immunity shapes mucosal microbial homeostasis.

7. The feline oral mycobiome

Although bacteria have received the bulk of research attention in the feline oral microbiome, fungi are an increasingly recognised (and probably under-appreciated) component of the oral microbial ecosystem. Krumbeck et al. (2021) reported the first characterisation of the feline oral mycobiome, analysing both bacterial and fungal communities in cats with FCGS and in healthy controls by parallel 16S rRNA and ITS (internal transcribed spacer) amplicon sequencing [15].

A high number of fungal species (n = 186) was detected across all samples, far exceeding what conventional culture would have recovered [15]. In healthy cats, Saccharomyces cerevisiae was the most relevant fungal species, whereas FCGS samples were enriched in Malassezia restricta, Malassezia arunalokei, Cladosporium penidielloides/salinae, and with few members of the family Aspergillaceae [15]. The compositional shift from Saccharomyces-dominated mycobiomes in healthy cats to Malassezia-dominated mycobiomes in disease was striking, and suggests that fungi may play active roles in FCGS pathogenesis rather than acting only as bystanders or opportunistic colonisers [15].

The pathogenic potential of Malassezia species is an important mechanistic issue. Malassezia are lipophilic, lipid-dependent basidiomycetous yeasts that are well-recognised agents in feline and canine inflammatory dermatoses and in human seborrhoeic dermatitis, atopic dermatitis, dandruff, and pityriasis versicolor. They are obligate lipid scavengers (having lost the genes for de novo fatty acid synthesis during evolution), and they secrete a spectrum of lipases and phospholipases that liberate free fatty acids, indoles, and other inflammatory mediators from host lipids. In cutaneous disease, Malassezia-derived metabolites have been reported to modulate keratinocyte differentiation, activate the aryl hydrocarbon receptor, and induce IL-6, IL-17, and Thymic Stromal Lymphopoietin (TSLP) secretion [15]. Whether analogous mechanisms operate in the oral mucosa is unknown but plausible.

Mycobiome research lags behind bacterial microbiome research for several reasons: (I) ITS reference databases are less well curated for animal-associated fungi than for plant pathogens; (II) ITS amplicons have variable length, complicating bioinformatic processing; (III) and DNA extraction from fungal cells (with their robust, β-glucan- and chitin-rich cell walls) frequently requires more aggressive bead-beating protocols than those optimised for bacteria, leading to under-recovery of fungal DNA from samples processed with bacterial-focused protocols. These technical issues partly explain why the mycobiome has been under-sampled historically, and argue for routine optimised ITS sequencing (and ideally shotgun metagenomics, which captures both domains simultaneously) in future feline oral microbiome studies.

8. Host immune responses and microbiome interactions

8.1. Innate and adaptive immunity in feline oral disease

Oral microbiome and host immune interactions are bidirectional and self-reinforcing: microbial dysbiosis triggers inflammation, and immune dysregulation in turn promotes further microbial imbalance through changes in the oral microenvironment [6, 7, 19]. Innate immune sensing in the oral mucosa relies on pattern recognition receptors expressed on epithelial cells, neutrophils, macrophages, and dendritic cells, including the TLR family (notably TLR2 for lipoteichoic acid and lipopeptides; TLR4 for LPS; TLR9 for unmethylated Cytosine-phosphate-Guanine (CpG) DNA), the Nucleotide-binding Oligomerization Domain (NOD)-like receptors (recognising peptidoglycan fragments), and the C-type lectin receptors Dectin-1 and Dectin-2 (recognising fungal β-glucans and α-mannans, respectively). Antimicrobial peptides, feline β-defensins, LL-37 cathelicidin orthologues, calprotectin (S100A8/A9), are presumed, by analogy with other species, to be constitutively expressed and inducibly upregulated in the oral epithelium, providing an important first line of mucosal defence whose dysregulation in disease has been incompletely characterised in cats.

In FCGS, immunohistochemistry consistently shows dense T-lymphocyte infiltrates in affected tissues, with increased CD3⁺, CD4⁺, and CD8⁺ T cells and a predominance of CD8⁺ cytotoxic over CD4⁺ helper cells [18, 21]. The CD8⁺ predominance is informative: it suggests that the FCGS inflammatory response is a sustained cytotoxic, MHC class I-restricted response to persistent antigenic stimulation, most plausibly from intracellular pathogens: viruses (notably FCV), or less likely intracellular bacteria [18, 21, 22]. This CD8 dominant pattern is qualitatively different from the predominantly CD4⁺-driven response of chronic periodontitis, in which Th1 and Th17 positive cells are important.

Proinflammatory cytokines, such as IL-6, tumour necrosis factor (TNF), and interferon-γ (IFN-γ) are elevated in FCGS, indicating local immune response [21, 22]. Peralta et al. (2023) published the transcriptomic signature of FCGS tissue, that was associated with upregulation of IL-6 and downstream Janus Kinase (JAK)/Signal Transducer and Activator of Transcription 3 (STAT3) signalling, with upregulation of Nuclear Factor κB (NF-κB) pathway and Matrix Metalloproteinases (MMPs) [22]. The IL-6/STAT3 axis is interesting given its roles in Th17 differentiation, plasma-cell maturation, and the chronic, self-sustaining character of mucosal inflammation. It is also a potential therapeutic target given the availability of IL-6 receptor antagonists in human medicine.

8.2. Immunoglobulin responses

Altered immunoglobulin expression has been described in cats with oral disease [21]. In FCGS, increased tissue infiltration by IgG- and IgM-producing plasma cells together with Mott cells (plasma cells distended by aggregated, electron-dense Russell bodies of misfolded immunoglobulin) reflects chronic, intense antigenic stimulation and B-cell activation [18, 21]. The role of mucosal IgA, normally produced in dimeric form by submucosal plasma cells, transcytosed across epithelial cells via the polymeric immunoglobulin receptor, and secreted as secretory IgA, which neutralises pathogens, blocks adhesion, and shapes the microbial community, is poorly understood in feline oral disease. Quantitative or qualitative defects in mucosal IgA production, class-switch recombination abnormalities, IgA glycosylation defects, or increased IgA degradation by bacterial proteases (a well-described virulence strategy in P. gingivalis, P. gulae, and Streptococcus spp.) are all plausible mechanisms warranting investigation in cats [21].

8.3. Cytokine-mediated microbiome-host crosstalk

Interactions between oral microbial composition and host cytokine expression form a self-reinforcing pathological cycle in chronic oral disease [21, 22]. Dysbiotic communities produce pathogen-associated molecular patterns (PAMPs), namely LPS, lipoteichoic acid, peptidoglycan fragments, flagellin, and microbial nucleic acids, that engage host pattern recognition receptors and trigger cytokine release. Crevicular inflammation is accompanied by metabolic changes, including an increase in gingival crevicular fluid volume. Higher fluid volume delivers certain serum proteins to the subgingival space. These nutrients, especially iron from haem, give a selective advantage to proteolytic, asaccharolytic anaerobes such as Porphyromonas and Treponema species, over their saccharolytic competitors. This nutritional shift (together with altered oxygen tension and pH) selects a more pathogenic community, completing a positive feedback loop between inflammation and dysbiosis [6, 7, 13]. Interrupting this cycle (for example by biofilm removal, immunomodulation, or by microbiome restoration) is a potential therapeutic approach.

9. Viral contributions to oral microbiome dysbiosis

Certain viruses are associated with feline oral diseases, particularly FCGS, and any model of oral disease pathogenesis includes them [8, 19, 20]. Feline calicivirus (FCV) is a small, non-enveloped, positive-sense single-stranded RNA virus (family Caliciviridae) with tropism for oral and conjunctival epithelial cells. Acute FCV infection produces a characteristic vesicular-ulcerative oropharyngitis, and a subset of cats persistently carries them up to several years. FCV detection is more frequent in cats with FCGS than in age- and environment-matched non-FCGS controls, and viral load correlates with lesion severity [18–20]. Isolation of “virulent systemic” FCV strains, with altered tropism and pathogenicity, indicated the plasticity of caliciviruses, complicating simple correlations between FCV detection and FCGS phenotype. A study by Druet and Hennet (2017) describes a positive association between FCV viral load and lesion outcome in feline caudal stomatitis, indicating a contributing role [24]. Feline herpesvirus-1 (FHV-1), an alphaherpesvirus, establishes latency in the trigeminal ganglia and reactivates under stress. This virus also contributes to acute ulceration in the mouth and conjunctiva, but its specific role in chronic FCGS pathogenesis is not clear yet [18, 19].

Viral infections might alter the oral microbiome possibly through several ways. First, virus-induced epithelial damage disrupts the oral mucosal barrier, allowing bacterial invasion of subepithelial compartments, providing more surface for biofilms [18]. Second, viral infection elicits innate immune responses (recruitment of myeloid and Natural Killer (NK) cells), altering the local microenvironment (redox state, growth-limiting nutrient availability), selectively favouring or certain bacterial taxa [19]. Third, the immunosuppressive effects of FIV and feline leukaemia virus (FeLV) compound these changes by impairing host control of opportunistic pathogens whose expansion is normally restrained by the hosts immune response [19, 27]. Lee et al. (2020) proposed an integrative model in which the interplay among viral infection, microbial dysbiosis, and immune dysregulation generates a self-perpetuating cycle of chronic oral inflammation in FCGS [18]. In this model, initial viral infection triggers mucosal damage and immune activation, precipitating microbial dysbiosis; the dysbiotic community then provides continuing antigenic and PAMP-mediated stimulation that sustains inflammation even after the initiating viral insult has resolved or become quiescent. This model accounts for the chronic, self-sustaining, often FCV-status-independent course of FCGS, and yields testable predictions; for example, interventions targeting any single arm of the triangle (virus, bacterium, or immune response) in isolation should produce only partial or transient responses, while combined approaches should be more effective [18]. This prediction matches the empirical observation that combined surgical extraction (microbial reservoir reduction), immunomodulation, and (where applicable) antiviral therapy generally outperform single-modality treatment.

The bacteriophage component of the oral virome (probably substantial given the high density of bacterial hosts) is essentially uncharacterised in cats and warrants further attention. In humans, oral phages influence bacterial community structure through lytic and lysogenic interactions, mediate horizontal transfer of virulence and antibiotic-resistance determinants, and modulate host immune responses [28]. Comparable studies in cats are, to our knowledge, lacking.

10. Diagnostic and therapeutic implications

10.1. Microbiome-based diagnostics

Growing evidence linking specific microbiome signatures to particular feline oral diseases has opened the way for microbiome-based diagnostics. One study used shotgun metagenomic sequencing in cohorts of more than 38,000 cats to identify 8,344 microbial species and to derive predictive algorithms for periodontal disease, tooth resorption, and halitosis from oral microbial profiles [29]. The scale of such datasets is impressive, but their clinical utility (sensitivity, specificity, positive predictive value at realistic prevalence, and clinical actionability) requires rigorous validation in independent, prospectively enrolled cohorts with blinded clinical assessment. Several methodological caveats apply: algorithms built from convenience samples (cats whose owners chose to test them) may suffer selection bias; the lack of standardised dental phenotyping in such cohorts limits the precision of disease labels; and the dynamic nature of the oral microbiome means that single-time-point sampling can misclassify cats whose microbial state is in transition. Integrating microbiome profiling with conventional dental assessment (rather than replacing the latter) is the most realistic near-term application.

10.2. Antimicrobial therapy

Davis and Weese (2022) reviewed antimicrobial therapy in companion-animal periodontal disease and concluded that, although systemic antibiotics are frequently prescribed, evidence for their efficacy as monotherapy is limited [6]. Antimicrobials can produce transient clinical improvement by reducing pathogen burden, but they do not address the underlying dysbiosis or the immune dysregulation that perpetuates disease, and they carry the recognised risks of resistance development, microbiome disruption, and adverse drug effects [6]. Consensus guidelines therefore recommend antimicrobials as an adjunct to mechanical debridement (professional dental cleaning and root planing under general anaesthesia) and, in refractory FCGS, full-mouth or premolar/molar extraction [4, 6]. Antimicrobial choice should ideally be guided by susceptibility data, but routine culture-based susceptibility testing of oral biofilms is rarely performed in practice, and biofilm heterogeneity complicates the interpretation of any single susceptibility result.

10.3. Emerging therapeutic approaches

Several current preclinical and clinical studies focus on the oral-microbiome therapies. Probiotic approaches (with Lactobacillus and Bifidobacterium strains) aim to restore microbial homeostasis through competitive exclusion, by mucosal immunomodulation, and direct interference with periodontal pathogen virulence [16, 19]. Their efficacy in feline oral disease has not been convincingly demonstrated yet, although preliminary studies suggest possible benefits. Bacteriophage therapy would be also an attractive approach, but its practical use is hindered by phage host-range limitations and bacterial resistance mechanisms.

Immunomodulatory approaches range from broad-spectrum agents (cyclosporin, glucocorticoids) to increasingly targeted interventions. Arzi et al. (2016) reported that fresh autologous mesenchymal stem cell therapy can produce substantial clinical improvement in cats with severe refractory FCGS, presumably via immunomodulation of T-cell responses, induction of regulatory T cells (FOXP3⁺), and modulation of macrophage polarisation [21]. The transcriptomic data of Peralta et al. (2023) suggest that targeting the IL-6/STAT3 axis (for example, by IL-6 receptor antagonism) could be a rational FCGS therapy, although feline-specific clinical evidence is lacking [22]. Disease-specific metabolic signatures (such as enrichment of polyamine biosynthesis and nitrogen metabolism in periodontitis) suggest potential metabolic targets for intervention [13]. Mesenchymal stem cell, immunobiological, and microbiome-modulating therapies will require extensive controlled clinical validation before they can be considered standard care.

A promising (rather conceptual) direction is “precision microbiome therapy”, in which patient-specific microbiome and immune profiles guide intervention. The feasibility of this in feline practice will depend on cost-effective rapid microbiome diagnostics, validated patient-stratification algorithms, and a wider therapeutic armamentarium than is currently available.

11. Future directions and knowledge gaps

Despite substantial progress over the past decade, several gaps remain in our understanding of the feline oral microbiome and its role in oral disease.

First, most studies to date have been cross-sectional, providing snapshots of microbial composition at single time-points but little insight into the temporal dynamics of disease initiation and progression. Longitudinal cohort studies track individually and longitudinally healthy cats until they develop oral disease. Ideally, they pair sampling with clinical phenotyping, and host transcriptomic profiling, to establish causal relationships and to identify early microbial markers of impending disease [6, 9].

Second, the functional aspects of the feline oral microbiome are only partially characterised. 16S rRNA amplicon sequencing, while informative for taxonomic profiling, gives limited insight into the metabolic activities and pathogenic potential of oral microbiota [13]. Wider use of shotgun metagenomic, metatranscriptomic, metaproteomic, and even advanced metabolomic approaches will be essential to move the field toward mechanistic understanding [13]. Integration with host transcriptomic and proteomic data, ideally from the same anatomical site at the same time, will be particularly useful for characterising microbiome-host interactions.

Third, the contributions of non-bacterial domains, namely the mycobiome, viral components, and the archaeome, remain largely unexplored in cats. Broader multi-kingdom analyses combining 16S, ITS, and viral metagenomic profiling are needed to understand the full complexity of oral microbial ecosystems and their disease associations [15].

Fourth, host-microbiome interactions in feline oral disease require more sophisticated investigation. Specific priorities include: (I) identification of host-genetic determinants of FCGS susceptibility (candidate-gene and, ideally, genome-wide association approaches in well-phenotyped cohorts); (II) characterisation of mucosal innate and adaptive immune responses in health and disease using single-cell RNA sequencing, mass cytometry, and spatial transcriptomic platforms; (III) elucidation of the gut-oral axis, including the possible contribution of gastrointestinal dysbiosis and microbial translocation to oral disease; and (IV) the development of feline-specific gnotobiotic or selectively colonised models to permit causal testing of candidate microbial mechanisms [19, 21]. The technical and ethical complexities of such models in cats are non-trivial, but selected components could be addressed using ex vivo organoid systems or in vitro biofilm mucosal coculture platforms.

In addition, translational research is needed to turn basic-science findings into clinically applicable diagnostic and therapeutic tools. Validated microbiome-based diagnostic tests, targeted antimicrobial strategies, and microbiome-modulating therapies (probiotics, prebiotics, synbiotics, postbiotics, phage cocktails, and engineered live biotherapeutic products) are promising but still largely unrealised opportunities [6, 29]. Rigorous randomised controlled trials with biologically meaningful endpoints (not just surrogate measures such as transient plaque reduction) will be essential.

Finally, greater methodological standardisation would be a benefit to the field: agreed sampling protocols, hypervariable-region or whole-genome shotgun choice, bioinformatic pipelines, reference databases (Greengenes, SILVA, RDP, GTDB, the Human Oral Microbiome Database), and clinical case definitions. Minimum-information standards (analogous to the Minimum Information about any (x) Sequence (MIxS) standards used in human microbiome research) would substantially improve inter-study comparability and meta-analytic potential.

12. Limitations

Several limitations of this review should be noted. First, this is a narrative rather than a systematic review: no pre-registered protocol was filed, no formal inclusion/exclusion criteria were applied, and no quantitative risk-of-bias assessment of cited studies was carried out. Selection was guided by topical relevance and methodological quality, but inevitably reflects the authors' judgement, and selection bias cannot be excluded. Without meta-analytic synthesis, disease-associated dysbiotic shifts are reported descriptively rather than with formal effect-size estimates and confidence intervals.

The literature search was limited to peer-reviewed English-language sources, which may introduce a language bias and exclude relevant work reported in other languages, a relevant consideration given that several active feline microbiome research groups publish in regional non-English journals. Publication bias toward positive or novel findings, common to all narrative syntheses, may further skew the apparent strength and consistency of reported microbiome-disease associations.

The recent and ongoing taxonomic restructuring of bacterial phyla (most notably the renaming of Bacteroidetes to Bacteroidota and Firmicutes to Bacillota) means that taxa cited in older versus newer studies under different names are not always directly comparable. We have tried to harmonise nomenclature, but residual inconsistencies cannot be excluded.

Finally, this is a quickly evolving field: new feline-specific datasets, modern sequencing technologies, and analytical frameworks appear at an accelerating rate. Readers are therefore advised to interpret the findings of this review as a snapshot of the current understanding of the topic rather than as a definitive record.

13. Conclusions

The feline oral microbiome is a complex, dynamic ecosystem central to oral health as well as disease. In health, the oral microbiome is dominated by Proteobacteria, with prominent genera including Porphyromonas, Moraxella, Capnocytophaga, and Bergeyella. Of the common feline oral diseases, periodontal disease (PD) and feline chronic gingivostomatitis (FCGS) show recognisable dysbiotic patterns: expansion of Bacteroidetes/Bacteroidota and Spirochaetes/Spirochaetota, enrichment of taxa including Treponema, Peptostreptococcus, Filifactor, and Fusobacterium nucleatum, and depletion of health-associated commensals. Each disease also has its own characteristic dysbiotic signature, reflecting the distinct pathogenic mechanisms underlying each condition.

FCGS shows the most profound and most widely distributed dysbiotic state, driven by bacterial, fungal, and viral factors against a background of dysregulated CD8⁺ skewed mucosal immunity. Tooth resorption (TR) is associated with characteristic (and possibly subtype-specific) microbial signatures, including taxa whose metabolism may influence local mineral homeostasis. FIV-associated oral disease is a useful natural model of how systemic immune compromise disrupts oral microbial homeostasis, and of how antiretroviral restoration can partially rebalance both immunity and microbiota.

Combining advanced sequencing with immunological, functional, and metabolic analyses is yielding increasingly detailed insights into the molecular mechanisms linking microbial dysbiosis to oral disease. These advances justify cautious optimism for diagnostic and therapeutic approaches grounded in microbial ecology rather than single-pathogen models. Substantial gaps remain, particularly in causal inference, functional and multi-kingdom microbiome characterisation, host-microbe genetic interactions, and translational application. Continued investment in rigorously designed, well-phenotyped, longitudinal, and methodologically standardised studies will be needed for microbiome-informed feline oral medicine to deliver on its promise. The goal of improving oral health, reducing morbidity, and enhancing well-being for feline patients is justified by the high prevalence and clinical impact of conditions that too often receive inadequate attention.

List of Abbreviations

16S rRNA

16S ribosomal Ribonucleic Acid

ANCOM-BC

Analysis of Compositions of Microbiomes with Bias Correction

AP

Aggressive Periodontitis

cART

Combined Antiretroviral Therapy

CD

Cluster of Differentiation

CP

Chronic Periodontitis

CpG

Cytosine-phosphate-Guanine (motif)

DNA

Deoxyribonucleic Acid

FCGS

Feline Chronic Gingivostomatitis

FCV

Feline Calicivirus

FeLV

Feline Leukaemia Virus

FHV-1

Feline Herpesvirus 1

FIV

Feline Immunodeficiency Virus

FORL

Feline Odontoclastic Resorptive Lesion

FOX

Forkhead Box

FOXP3

Forkhead box P3

GTDB

Genome Taxonomy Database

HIV

Human Immunodeficiency Virus

ICNP

International Code of Nomenclature of Prokaryotes

IFN

Interferon

Ig

Immunoglobulin (IgA, IgG, IgM = isotypes A, G, M)

IL

Interleukin

ITS

Internal Transcribed Spacer

JAK

Janus Kinase

Kgp

Lysine-specific Gingipain

LL-37

Cathelicidin Antimicrobial Peptide LL-37

LPS

Lipopolysaccharide

MHC

Major Histocompatibility Complex

MIxS

Minimum Information about any (x) Sequence

MMP

Matrix Metalloproteinase

NF-κB

Nuclear Factor kappa-light-chain-enhancer of activated B-cells

NK

Natural Killer Cell

NOD

Nucleotide-binding Oligomerization Domain

OPG

Osteoprotegerin

OTU

Operational Taxonomic Unit

PAMP

Pathogen-associated Molecular Pattern

PCoA

Principal Coordinates Analysis

PCR

Polymerase Chain Reaction

PD

Periodontal Disease

pH

Pondus Hidrogenii

PICRUSt2

Phylogenetic Investigation of Communities by Reconstruction of Unobserved States 2

PMT

Pasteurella multocida toxin

PSD

Polymicrobial Synergy and Dysbiosis

RANK

Receptor Activator of Nuclear factor Kappa-B

RANKL

Receptor Activator of Nuclear factor Kappa-B Ligand

RDP

Ribosomal Database Project

Rgp

Arginine-specific Gingipain

RNA

Ribonucleic Acid

rRNA

Ribosomal Ribonucleic Acid

STAT

Signal Transducer and Activator of Transcription

TLR

Toll-like Receptor

TNF

Tumor Necrosis Factor

Th

T helper (cell, with subset numbers e.g. Th1, Th17)

TR

Tooth Resorption

TSLP

Thymic Stromal Lymphopoietin

Funding Statement

Funding sources: No financial support was received for this study.