Introduction

Quorum sensing (QS) is a cell-density-dependent bacterial communication system mediated by small diffusible signaling molecules called autoinducers (AI)1,2. There are three major types of AIs:(i) AI-1 (N-acyl homoserine lactones or AHLs)3,4,5; (ii) AI-26,7,8,9 consisting of 4,5-dihydroxy-2,3-pentandedione (DPD) derivatives10; and (iii) autoinducing peptides (AIPs), specific to Gram-positive bacteria11. AIs are typically secreted into the environment and can re-enter the cells either by passive diffusion in the case of AI-13 or via active transport for AI-2 and AIPs6,12,13. During re-entry, AIs bind to their target receptors on the membrane (AI-2; AIPs) or in the cytoplasm (AI-1) regulating the expression of numerous genes2. Indeed, QS can regulate up to 26% of the bacterial genome14. By controlling gene expression, QS regulates several bacterial behaviors such as biofilm formation, virulence, and antimicrobial resistance15,16. These behaviors are critical for bacterial adaptation to varying ecological niches, often characterized by hostile environmental conditions and inconsistent nutrient availability17.

Dental plaque is a complex oral biofilm adhering to the tooth surface, and is composed of commensal, symbiotic, and pathogenic members of oral microbiota18. The core microbiome of the human oral cavity is normally comprised of microorganisms, including bacteria, fungi, archaea, protozoa, and viruses19,20 that inhabit distinct ecological niches in unique oral environments such as gingival sulcus, tongue, cheek, soft and hard palates, throat, saliva, and teeth21,22. The bacterial taxonomic diversity in the oral cavity includes about 700 species belonging to 185 genera and 12 phyla23,24. The healthy oral microbiome is dominated by Gram-positive commensal bacteria such as Streptococcus and Actinomyces ssp. with high propensity for fermentation.25,26,27. Abrupt ecological changes and alterations of host-associated innate and adaptive immunity in the oral cavity can cause dysbiosis. Dysbiosis is characterized by an increased prevalence of anaerobic Gram-negative pathogenic bacteria with high proteolytic activity, such as Porphyromonas28, often leading to the development of periodontal diseases25,26,27.

Dental plaque formation is characterized by asynchronous spatial and temporal colonization of the tooth surface requiring interactions between several groups of bacteria29,30,31. Interbacterial interactions are hypothesized to depend in part on QS-dependent regulation of cellular metabolism, physiology, secretion, virulence, motility, attachment, and cooperativity/competition32,33. AI-2 and AIPs are known to be important in the development of oral biofilms. Many dental plaque colonizers were shown to produce and/or respond to AI-2,34,35,36,37,38,39,40,41. Gram-positive pioneer colonizers like oral Streptococcus species can utilize AIPs42,43,44,45. Hence, AI-2/AIP-mediated QS appears to be the primary mode of signaling and regulation that contributes to the formation and development of oral biofilms including dental plaque33. The potential role of AI-1/AHL-based QS in the development of dental plaque and other oral biofilms remains elusive. In fact, AHLs were long undetected in cultures of pathogenic oral microbiota36,46,47. Given the lack of identified AHL synthases and receptor homologs in plaque pathogens, AHLs have been considered to make an insignificant contribution to oral biofilm formation48,49.

This paradigm may be ripe for change. Various types of AHLs are found in human saliva samples50,51,52. Sources include the AHL-producing Gram-negative bacteria Pseudomonas putida53, Enterobacter54, Klebsiella pneumoniae55, Citrobacter amalonaticus56 and Burkholderia57, found on tongue surfaces and within dentine caries. Certain strains of the oral pathogen Porphyromonas gingivalis also possess AHL synthase/receptor homologs58,59 and produce small quantities of AHLs in both axenic and multispecies cultures52. Indeed, putative AHL biosynthetic genes exist in oral microbial genomes60. Metagenomic analysis of human dental plaque revealed a high abundance of AHL-synthase (HdtS), receptor homologs and Quorum Quenching (QQ) enzymes61. Muras and colleagues have recently shown50,52,61 that while exogenously added AHLs promote the growth of pathogens in oral communities, AHL-degrading QQ enzymes inhibit oral biofilm formation and alter their microbial composition.

Yet, the specific role(s) of AHL-mediated QS in oral communities remain unclear. To increase our understanding of the importance of AI-1, one strategy is to apply different AHL signals and AHL-degrading lactonases onto oral microbial communities grown under different oxygen conditions and evaluate their effects. Lactonases are QQ enzymes that hydrolyze the lactone ring of AHLs and disrupt AHL-mediated QS62,63,64,65. We have previously characterized a variety of lactonases, including SsoPox from the Phosphotriesterase-like Lactonase (PLL) family66,67,68 and GcL from the Metallo-β-Lactamase (MLL) family69. Interestingly, these two lactonases exhibit distinct, yet overlapping substrate specificities: SsoPox preferentially degrades long-chain AHLs (C8 or longer)68 and GcL is a generalist lactonase with broad substrate specificity69. This distinction is important because AHLs mainly vary by the length and decoration of their acyl chain, and these changes in chemical structure modulate signal specificity65,70,71,72,73. These chemically diverse AHLs are produced and sensed by a variety of bacteria. Therefore, the use of these enzymes may capture signal-specific changes to the oral community. Lactonases reduce the pathogenicity of Gram-negative bacterial species both in vitro74,75 and in vivo76,77,78,79. Lactonases hydrolyze AHLs and therefore quench AHL-dependent QS signaling pathways. QQ downregulates networks of genes, responsible for the production of virulence factors and enzymes that contribute to phenotypes commonly associated with pathogenicity such as antibiotic resistance, biofilm formation, host colonization and tissue damage. The substrate preference of lactonases correlates with distinct proteome profiles, virulence factor expression, antibiotic resistance profiles, and biofilm formation16,80,81.

In this study, we investigated the importance of AHL-signaling in biofilms and planktonic cultures of a model human supragingival-origin plaque community82,83 under both aerobic (5% CO2 atmosphere) and anaerobic conditions. AHLs were detected in plaque community cultures using a plasmid-based AHL biosensor and High-Performance Liquid Chromatography Mass Spectrometry (HPLC-MS). We also evaluated the effects of exogenous AHLs of varying acyl chain lengths (C6 and C12) and QQ lactonases (SsoPox and GcL) with distinct substrate preference on the microbial composition of the plaque community. In response to these treatments, changes in microbial composition and population structure of the community were observed using 16S rRNA sequencing. Additionally, lactonase treatment under 5% CO2 conditions altered biofilm formation, increased the propensity of the biofilms to ferment sucrose to lactate and altered the ability of the community to utilize various carbon sources. Our observations from this study suggest that AI-1/ AHL-based QS is crucially important in determining the pathogenic profile of the oral microbiome.

Results

AHLs are present in supragingival plaque community cultures under CO2 atmosphere but not under anaerobic conditions

Oxygen availability critically affects oral biofilm development and the progression of cariogenic and periodontal diseases84,85. Indeed, oral biofilms experience rapid and continuous changes in oxygen levels, among other physio-chemical changes, which impact the microbial composition of the oral microbiota86,87,88. The oral cavity is normally highly aerobic due to continuous delivery of oxygen through air exposure and the circulatory system. Early (thin) biofilms ( < 100 µm thick) allow rapid oxygen diffusion85, but as plaque matures and thickens, oxygen diffusion becomes limited, creating microaerophilic and hypoxic conditions that favor cariogenic facultative anaerobes such as Streptococcus mutans89,90. Given that AHLs were previously reported in anaerobic co-cultures of Porphyromonas gingivalis with Streptococci52 - two bacteria commonly found in dental plaque – we hypothesized that supragingival plaque communities could produce AHLs in an oxygen-dependent manner.

We investigated the presence of AHLs in a previously characterized oral microbial community derived from the pooled supragingival plaque of five healthy human volunteers82,83 using HPLC-MS analysis. C6-HSL was detected in the supragingival plaque biofilm community cultured under 5% CO2 (Fig. S1A, G) but not in anaerobic cultures (Fig. S1B, G). C6-HSL was present at lower levels in the culture media (Fig. S1C, G), confirming community production rather than medium contamination. C4-HSL was detected in cultures grown under 5% CO2 (Fig. S2A, G) and anerobic conditions (Fig. S2B, G) but was also present in the culture media (Fig. S2C, G), preventing reliable assignment of this compound to community production. Neither C6-HSL nor C4-HSL was detected in saliva samples (Fig. S1D, G; S2D, G, respectively). Despite optimization efforts, our extraction procedure (see methods) was ineffective for longer-chain AHLs, suggesting that additional AHLs beyond C6-HSL may have been present but undetectable due to poor extraction yields.

To further validate AHL presence, we used a PluxI-GFP based plasmid biosensor, which detects AHLs in the 10 nM to 1 μM range, to evaluate the AHL content of these cultures. The biosensor successfully detected and quantified pure AHL standards (C6-HSL; Fig. S3), confirming its functionality. Consistent with HPLC-MS results, the biosensor detected AHLs in supragingival plaque communities grown under 5% CO2, with signals equivalent to approximately 100 nM of C6-HSL (Fig. 1). No AHLs were detected in anaerobic cultures (Fig. S4), confirming the oxygen-dependent nature of AHL production in these communities.

Fig. 1: AHLs are detected in the spent supernatants of a supragingival plaque community cultured in vitro under 5% CO2 atmosphere.

Cell-free culture supernatants of 6 biological replicates of plaque community cultures treated with lactonases 5A8 (inactive lactonase mutant), SsoPox, or GcL were incubated with E. coli AHL biosensor strain JM109 pJBA132. The resulting GFP fluorescence is indicated as relative fluorescence units (RFU) per unit OD600nm of biosensor cultures on the left Y-axis. The equivalent C6-HSL concentration for the corresponding RFU/OD600 values as shown on the right Y-axis was extrapolated from the standard curve of the biosensor response against C6-HSL in Fig. S3 and is only represented as an indication, since the used biosensor is broad spectrum and the reading may be a reflection of many different produced AHLs. Results represent the mean and standard deviation of 6 biological replicate cultures for each lactonase treatment. As each of the cell-free supernatants of 6 biological replicate cultures of the plaque community for all lactonase treatments was further incubated with 6 biological replicates of biosensor cultures, each data point on this graph represents the mean of the 6 biosensor replicates. Statistical significance of all treatments compared to the control (5A8) was calculated using unpaired two-tailed t-tests with Welch’s correction and significance values are indicated as - ***p < 0.0005, **p < 0.005 and *p < 0.05.

To further validate these results, we treated samples with lactonases SsoPox and GcL. Both lactonases significantly decreased biosensor signals: SsoPox reduced signals by 86% (~13.7 nM of C6-HSL equivalent) and GcL by 49% (~50.8 nM of C6-HSL equivalent) (Fig. 1). These reductions confirm that detected signals represent genuine AHL activity rather than biosensor artifacts.

It should be noted that the C6-HSL equivalent concentration values serve as indicative estimates only. Since the biosensor can detect C6-C12 HSLs91 and may recognize multiple AHLs produced by the community beyond C6-HSL, these values may not reflect precise concentrations. The absence of AHLs under anaerobic conditions aligns with previous observations in other anaerobic biofilms92,93, likely indicating that AHL production ceases under these conditions, although alternative explanations, such as extracellular AHL lactonases or acylases secretion cannot be excluded.

CO2 atmosphere conditions affect the population structure of supragingival plaque community

Although supragingival plaque microbial population structure has been previously well documented94,95,96, the effects of varying oxygen conditions on the community remain unclear. This question is particularly relevant because oxygen levels vary dramatically within dental plaque and decrease sharply with increasing thickness97. To address this gap, we analyzed the microbial population structures of a supragingival-origin plaque communities using V4 amplicon sequencing under either 5% CO2 or anaerobic conditions.

The 5% CO2 and anaerobic communities exhibited distinct beta diversity patterns, as demonstrated by clear clustering separation in PCoA using Bray-Curtis distances (Fig. 2A). Statistical analysis confirmed significantly greater differences between oxygen treatment groups compared to within-group variations (ANOSIM: R = 1, p = 0.003 for biofilm samples; R = 1, p = 0.006 for planktonic samples), indicating that oxygen availability is a major driver of community structure.

Fig. 2: Microbial composition of a supragingival plaque community cultured in vitro under 5% CO2 and anaerobic culture conditions.

A Principal-coordinate analysis (PCoA) of Bray-Curtis distances of both biofilm and planktonic plaque communities. B Linear discriminant analysis effect size (LEfSe) of biofilm dental plaque communities. The bar graph of LDA scores shows the taxa (OTU: Operational Taxonomic Unit) with a statistical difference (Wilcoxon test: p < 0.004) between 5% CO2 and anaerobic communities (Only taxa meeting a LDA significant threshold > 4 are shown).

To identify the taxa responsible for discriminating between oxygen conditions, we applied Linear Discriminant Analysis Effect Size (LEfSe) analysis98 to biofilm samples (Fig. 2B). This analysis revealed that four nested taxa characterized the 5% CO2 community, while nine nested taxa distinguished the anaerobic community. Under 5% CO2 conditions, the discriminating taxa are mainly Gram-positive bacteria, including Abiotrophia, Schaalia (formerly known as Actinomyces), a member of Lactobacillales order, and Streptococcus. In contrast, the anaerobic community was mainly characterized by Gram-negative bacteria, including Fusobacterium, Prevotella, Porphyromonas, Haemophilus, and Veillonella.

Broader analysis of community composition revealed striking differences in Gram-positive and Gram-negative bacterial distribution between oxygen conditions. Under anaerobic conditions, Gram-positive taxa (including Streptococcus, Peptostreptococcus, Parvimonas and Gemella) comprised 32% of total read abundance in both anaerobic biofilm and planktonic plaque communities, while Gram-negative taxa (Fusobacterium, Prevotella, Porphyromonas, Haemophilus, and Veillonella) represented 42% (Fig. S5A).

Conversely, under 5% CO2 conditions, Gram-negative taxa (Fusobacterium, Porphyromonas, Prevotella, and Veilonella) represented less than 5% of the communities (Fig. S5B). Planktonic cells from both culture conditions exhibited similar oxygen-dependent patterns (Fig. S6). Additional analysis of biological variability (Fig. S7) and taxon-specific abundance patterns across samples (Fig. S8) further support these LEfSe-derived trends.

Notably, neither the 5% CO2 nor anaerobic microbial populations contained many taxa reported to produce AHLs, with P. gingivalis being the primary exception52. Most identified microbes have been reported to produce alternative signaling molecules such as AI-2 and/or AIPs36,46,47,49,99,100,101. This observation suggests that the observed differences in AHL levels between oxygen conditions may result from oxygen gradient effect102,103,104,105 on AHL production rather than changes in AHL-producing species abundance. This interpretation aligns with a previous report showing that oxygen availability can reduce AHL production in species such as Pseudomonas aeruginosa, thereby affecting signal concentration and effectiveness106.

AHL signal disruption by lactonases increased the abundance of commensals and early colonizers in supragingival plaque community

To examine the importance of AHL signaling in microbial population structure, we used quorum quenching (QQ) to disrupt AHL signaling and analyzed community changes through 16S rRNA gene sequencing. We employed two lactonases with different substrate specificities: GcL (broad spectrum AHL degrading enzyme69) and SsoPox (preferentially degrades long-chain AHLs68).

Biofilm and planktonic communities under 5% CO2 showed markedly different microbial compositions regardless of lactonase treatment, as demonstrated by distinct clustering patterns in PCoA analysis (Fig. S7A). Certain core taxa were consistently observed across all samples and treatments, including Streptococcus, member of the order Lactobacillales, and Actinomyces – all of which are Gram-positive commensals and early colonizers of oral biofilms107,108. Notably, Abiotrophia was exclusively detected in the planktonic communities and was absent from biofilm communities across all treatments (Fig. 3A and S8A).

Lactonase treatments produced subtle but measurable alterations in the composition of the planktonic microbial communities under 5% CO2 conditions. Key taxa showed varying abundance levels between treatments: Streptococcus increased from 51.5% (control 5A8) to 53.7% (GcL) and 63.3% (SsoPox). Other taxa also showed changes in abundance levels, such as Lactobacillales (23.0% in the control 5A8, 24.0% in GcL and 21.8% in SsoPox); Actinomyces (7.2% in the control 5A8, 9.7% in GcL and 6.5% in SsoPox); Abiotrophia (11.5% in the control 5A8, 8.5% in GcL and 9.3% in SsoPox).(Fig. 3A and S8A). Individual lactonase treatments of planktonic communities did not produce statistically robust differences compared to control (GcL vs. 5A8: AMOVA: p = 0.371; ANOSIM: R = 0.255; p = 0.007*; SsoPox vs. 5A8: AMOVA: p = 0.143; ANOSIM: R = 0.191, p = 0.081). However, when analyzed collectively, lactonase-induced population changes were statistically significant (AMOVA: p = 0.047*; ANOSIM: R = 0.305; p = 0.003*; Tables S1 and S2). Alpha diversity analysis revealed that SsoPox lactonase treatment significantly reduced community diversity compared to control (t-test: p = 0.013) and GcL lactonase (t-test: p = 0.031) (Table S3, Fig. 3B, C).

Fig. 3: Lactonases altered the biofilm and planktonic mirobiota of a supragingival plaque community cultured in vitro under 5% CO2 atmosphere conditions.

A Average relative abundance of the bacteria taxa at the genus level. “Others” represent taxa comprising less than 5% of the total relative abundance per sample. B Alpha diversity (Shannon index) of the biofilm community colored by treatment (gray: SsoPox 5A8 (control); blue: Ss