Hyperuricemia is associated with periodontal ligature dysbiosis in a murine model of experimental periodontitis Authors: Zhicong Wu, Tiantian Xu, Li Zhao, Yinghong Zhou, Ting Yu Published: 2026 ABSTRACT AbstractBackgroundHyperuricemia (Hu) is a systemic metabolic disorder implicated in inflammatory conditions and has been epidemiologically associated with an increased prevalence of periodontitis. However, whether Hu is associated with periodontal microbial dysbiosis in experimental periodontitis remains unclear.ObjectiveThis exploratory pilot study aimed to investigate whether Hu is associated with periodontal dysbiosis in experimental periodontitis by characterizing the diversity, taxonomic composition, and predicted functional potential of ligature-associated microbiota using 16S rRNA sequencing.MethodsA mouse model combining potassium oxonate-induced Hu with ligature-induced periodontitis was established. Male C57BL/6 mice were allocated to normouricemia with periodontitis (NuP), hyperuricemia with periodontitis (HuP), and allopurinol-treated hyperuricemia with periodontitis (HuP+Allo) groups. Ligature-associated microbiota were analyzed by 16S rRNA gene sequencing. Correlations between microbial alterations and biochemical parameters were evaluated using redundancy analysis.ResultsHu significantly increased microbial diversity and evenness, as indicated by higher Shannon index (P = 0.007) and Shannoneven index (P = 0.002), but did not affect community richness. At the phylum level, Hu increased the Firmicutes/Bacteroidota ratio and Proteobacteria abundance while decreasing Bacteroidota and Actinobacteriota. At the genus level, Hu enriched potentially pathogenic taxa, including Streptococcus and Escherichia-Shigella. Allopurinol partially reversed these changes. Redundancy analysis revealed that microbial shifts were significantly correlated with serum uric acid (R2 = 0.711, P = 0.040) and creatinine (R2 = 0.695, P = 0.035). Functional prediction suggested that Hu was associated with higher predicted abundances of NOD-like receptor signalling and lower predicted abundance of IL-17 signalling pathways.ConclusionThese exploratory findings suggest that Hu may be associated with dysbiotic remodeling of the periodontal ligature microbiota in experimental periodontitis, linking systemic metabolic disturbance with local microbial imbalance and inflammatory pathway alteration. These alterations were partially reversed by allopurinol and correlated with serum uric acid and creatinine, providing mechanistic insight into how Hu may aggravate periodontal tissue destruction.Keywords: Hyperuricemia, periodontitis, dysbiosis, microbiome, allopurinol, mouse model Introduction Periodontitis is a chronic inflammatory disease characterised by the progressive destruction of the tooth-supporting periodontal tissues, ultimately leading to alveolar bone loss and tooth loss [1]. The pathogenesis of periodontitis is fundamentally driven by disruption of the symbiotic homoeostasis within the subgingival microbial community, resulting in a dysbiotic state in which pathogenic taxa progressively outcompete health-associated commensal species [2,3]. This ecological imbalance promotes sustained activation of host immune-inflammatory responses, leading to excessive cytokine production, alveolar bone resorption and clinical attachment loss [4,5]. High-throughput sequencing studies have confirmed distinct microbial profiles in individuals with gingivitis or periodontitis compared with periodontally healthy counterparts [6–8]. Furthermore, clinical interventions that target biofilm removal and help restore microbial ecological balance, such as periodontal debridement combined with antimicrobial agents, have been shown to effectively control inflammation and arrest disease progression [9–11], further supporting the pivotal role of microbial factors in periodontitis. Although dental plaque biofilm is considered the primary initiating factor of periodontitis, accumulating evidence indicates that periodontitis is not merely a localised infectious disease, but is substantially modulated by systemic conditions that influence disease susceptibility and severity. Environmental exposures, lifestyle factors, and a range of systemic diseases can alter both host immune competence and the composition and functional capacity of the oral microbiota, thereby disrupting periodontal microbial homoeostasis and increasing the risk of disease development and progression. Several noncommunicable chronic diseases, such as diabetes and cardiovascular disease, are considered potential or established risk factors for periodontitis [12,13]. Conversely, periodontal pathogens may also affect systemic health, including cardiovascular disease and inflammatory bowel disease, through hematogenous dissemination or via the digestive tract [14–16]. In essence, systemic disorders are thought to exacerbate periodontitis through two interconnected mechanisms: amplification of host inflammatory responses and direct or indirect perturbation of the periodontal microbial ecosystem toward a more pathogenic, dysbiotic state [17]. Hyperuricemia (Hu), defined by elevated serum uric acid (UA) levels, is a prevalent metabolic disorder closely linked to dietary habits and systemic comorbidities, such as hypertension, chronic kidney disease, and metabolic syndrome [18]. Accumulating epidemiological and clinical evidence suggests that Hu and gout are associated with increased prevalence and severity of periodontitis [19,20]. Despite these associations, the biological mechanisms linking Hu to periodontal disease progression remain incompletely understood. Our recent study established a mouse model of concurrent potassium oxonate-induced Hu and ligature-induced periodontitis, and demonstrated that Hu exacerbates periodontitis by aggravating local periodontal inflammation and oxidative stress, whereas pharmacological urate lowering with allopurinol effectively mitigates these pathological changes [21]. However, given that microbial dysbiosis is a fundamental driver of periodontitis pathogenesis, it remains unclear whether Hu also aggravates disease progression through modulation of the ligature-associated periodontal microbiota. Therefore, the present exploratory pilot study aimed to determine whether Hu is associated with ligature-associated periodontal dysbiosis in experimental periodontitis. To address this question, ligature-associated microbial communities under Hu conditions were characterised by 16S ribosomal RNA gene sequencing in an established mouse model of concurrent potassium oxonate-induced Hu and ligature-induced periodontitis [21], with particular attention to microbial diversity, taxonomic composition, and predicted functional potential. It was hypothesised that Hu induces a dysbiotic shift in the ligature-associated periodontal microbiota and that these microbial alterations are associated with systemic UA levels. This study may provide preliminary insight into the potential microbiological links between Hu and periodontitis from a host–microbe interaction perspective. Materials and methods Animals The animals used in this study were supplied by the Guangdong Medical Experimental Animal Centre and housed at the same facility. All experimental procedures were approved by the Animal Ethics Committee of the Guangdong Medical Experimental Animal Centre (Approval No. C202112-03) and were conducted in accordance with the Guide for the Care and Use of Laboratory Animals [22]. The mice were group-housed under specific-pathogen-free conditions at a controlled ambient temperature (20 to 26 °C) with a 12-hour light/dark cycle, and provided ad libitum access to standard chow and water. Mouse models of potassium oxonate-induced Hu and experimental periodontitis After a three-day quarantine and acclimatisation, fifty healthy mice were ear-tagged and randomly assigned using a randomisation function to five groups (n = 10 per group): hyperuricemia (Hu) mice with (HuP) or without (HuC) periodontitis, normouricemia (Nu) mice with (NuP) or without (NuC) periodontitis, and HuP mice treated with allopurinol (HuP+Allo). The combined mouse model of potassium oxonate-induced Hu and ligature-induced experimental periodontitis was established [21]. Briefly, Hu was induced by daily intraperitoneal injection of potassium oxonate (600 mg/kg body weight) for 28 consecutive days (from day 0 to day 28), while control mice received an equivalent volume of 0.5% sodium carboxymethylcellulose. Experimental periodontitis was induced by placing sterile 5-0 silk ligatures subgingival around the bilateral maxillary second molars on day 7, which remained in situ for 21 days (from day 7 to day 28). For allopurinol intervention, mice in the HuP+Allo group received daily intraperitoneal injection of allopurinol (5 mg/kg body weight) starting 7 days after periodontal ligation (from day 14 to day 28). At the experiment endpoint, all mice were fasted overnight and euthanized in an alternating sequence for tissue retrieval. Blood samples were obtained for serum biochemical analyses. Thoracic and abdominal organs, including the kidneys, liver, spleen, heart, lungs and epididymal visceral adipose tissue, were collected and weighed. Silk ligatures placed around the second maxillary molars in the ligated groups were carefully removed under a stereomicroscope. Ligatures were divided into two portions: one portion was immediately frozen in liquid nitrogen and stored at −80 °C for subsequent 16S ribosomal RNA gene sequencing, and the remaining portion was immersed in 100 μL phosphate buffered saline, vortexed for 30 seconds, and used for UA quantification. Validation of the Hu–periodontitis model, as well as data on visceral adipose tissue weight, ligature UA content, serum lipid, glucose, UA, creatinine (Cr), and urea nitrogen (UN), have been previously published [21]. 16S ribosomal RNA gene sequencing The ligature samples analysed in this study were derived from a larger set of animals in our previously published study (n = 10 per group) [21]. From this study, three mice were randomly selected from each of the three experimental groups (NuP, HuP, and HuP+Allo) for 16S rRNA gene sequencing. The inclusion criteria were as follows: the silk ligatures remained intact throughout the 21-day experimental period without detachment; serum UA levels were successfully elevated in HuP mice compared with NuP mice and reduced in HuP+Allo mice compared with HuP mice; and the samples passed standard quality control procedures for DNA extraction and sequencing. Samples that did not meet these criteria were excluded from the sequencing analysis. For the three ligated groups (i.e. NuP, HuP and HuP+Allo), total microbial genomic DNA was extracted from silk ligature samples using a commercial DNA extraction kit (E.Z.N.A.® Omega Biotek, Norcross, GA, U.S.A.). The V3 to V4 hypervariable regions of the bacterial 16S ribosomal RNA gene were amplified using the primer pair 338F (5'-ACTCCTACGGGAGGCAGCAG-3') and 806 R (5'-GGACTACHVGGGTWTCTAAT-3'). PCR amplicons were purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, CA, U.S.A.) and quantified with a Quantus® Fluorometer (Promega, Madison, WI, U.S.A.). Equimolar amounts of purified amplicons were pooled and subjected to paired-end sequencing on a NovaSeq PE250 platform (Illumina, San Diego, CA, U.S.A.). Raw sequencing reads were generated using Illumina MiSeq (v1.9.1). Raw FASTQ files were de-multiplexed using an in-house Perl script, quality-filtered using Fastp (v0.19.6), and merged using FLASh (v1.2.7). High-quality sequences were clustered into operational taxonomic units (OTUs) at a 97% sequence similarity threshold using UPARSE (v7.1). To minimise bias introduced by unequal sequencing depth, all samples were rarefied to 72,706 reads prior to downstream diversity analyses. Taxonomic classification of OTUs was performed using the RDP classifier (http://rdp.cme.msu.edu/, v2.11, accessed Feb 2024) against the SILVA 16S ribosomal RNA gene reference database (v138), with a confidence threshold of 70%. Microbial community composition was summarised at multiple taxonomic levels. Alpha (α) diversity indices, including Ace, Chao, Sobs, Shannon, Shannon evenness, and coverage, were calculated to assess within-sample diversity. Welch’s t-test was used for two-group comparisons of α-diversity, with multiple testing corrections performed using the Benjamini-Hochberg false discovery rate method. Beta (β) diversity was evaluated using principal component analysis (PCA) based on Bray-Curtis dissimilarity matrices. Linear discriminant analysis effect size (LEfSe) was used to identify differentially abundant taxa between experimental groups. Redundancy analysis (RDA) and correlation heatmap analysis were conducted to evaluate associations between microbial community composition at the genus level and 22 biological indicators, including ligature UA content, body weight gain, organ weights, serum glucose and lipid levels, and renal function parameters. Multicollinearity among explanatory variables was assessed using variance inflation factor analysis, resulting in the exclusion of eight variables with variance inflation factor values > 10. The significance and robustness of the RDA model were assessed by permutation test. To descriptively assess whether the selected samples captured the major features of the microbial community, Pan/Core species analysis was performed to examine the relationship between sample number and observed taxonomic richness. Core-species accumulation curves and pan-species curves were generated for each group. In addition, rarefaction analysis of α-diversity indices, including Sobs and Shannon, was conducted to evaluate sequencing coverage and the stability of diversity estimates. Building on prior evidence that Hu modulates host inflammatory signalling pathways in experimental periodontitis, including NLRP3 inflammasome activation, osteoclast-related gene expression, and IL-17- and HIF-1α-associated pathways [21], we further hypothesised that Hu induces corresponding alterations in the functional potential of the ligature-associated microbiota. To test this hypothesis, PICRUSt2 (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States, version 2.5.0) against the Kyoto Encyclopaedia of Genes and Genomes (KEGG) database was used to predict metagenomic function based on OTU representative sequences [23,24]. The workflow involved phylogenetic placement of amplicon sequence variants, hidden state prediction of gene family copy numbers, and normalisation by predicted 16S rRNA copy numbers to generate per-sample functional profiles [24]. Notably, PICRUSt2 predicts functional potential based on reference genomes rather than directly measuring gene expression or metabolic activity. Statistical analyses A single mouse was considered the experimental unit for all statistical analyses. Statistical analyses were conducted using commercial software (SPSS v20.0; IBM Inc., NY, U.S.A.). Data are presented as mean ± standard deviation (SD) unless stated otherwise. For comparisons among multiple groups, one-way analysis of variance (ANOVA) was applied following assessment of variance homogeneity using Levene’s test (significance level set at 0.10). When the assumption of equal variances was not met, Welch’s ANOVA was used. Post hoc multiple comparisons were conducted using the least significant difference (LSD) test or Dunnett’s test, as appropriate (for comparisons of multiple treatment groups against a single control group). Comparisons of bacterial community composition were assessed using the non-parametric Mann-Whitney U test, with data presented as medians. Statistical significance was considered at P < 0.05 unless stated otherwise. Details of the bioinformatic and microbiome-specific statistical analyses are provided in the 16S ribosomal RNA gene sequencing section above. Results Hu alters ligature-associated microbial diversity in experimental periodontitis The sequencing coverage and the descriptive sampling characteristics of the dataset were evaluated using rarefaction analysis and Pan/Core species curves. As shown in Figure 1a–d, the Pan/Core species curves (Figure 1a, b) tended to plateau, and the rarefaction curves (Figure 1c, d) also approached saturation, suggesting that the major shared taxa and overall diversity patterns were captured in this dataset. The results from 16S ribosomal RNA sequencing revealed that the ligature-associated periodontal microbiota in the three experimental groups (i.e. NuP, HuP and HuP+Allo) comprised 51 OTUs, spanning 7 phyla and 37 genera (Figure 1e). Approximately 70% of detected OTUs were shared among all three groups, indicating a substantial core microbial community. α-diversity analysis revealed that Hu significantly increased microbial community diversity, as reflected by a higher Shannon index, and enhanced community evenness, as indicated by the Shannoneven index, in ligated mice. However, Hu did not affect community richness (Sobs, Ace and Chao indices) or coverage (coverage index) (Figure 1f–k). Conversely, allopurinol intervention tended to reduce both diversity (P = 0.082) and evenness (P = 0.057) in HuP mice (Figure 1i, j). (a, b) Pan/Core species curve analysis showing that the curves tended to plateau as sample number increased, suggesting that the major shared taxa were represented in the sampled dataset. (c, d) Rarefaction curves based on the Sobs and Shannon indices showing that the curves approached saturation, indicating adequate sequencing coverage and stable diversity estimates. (e) Venn diagrams showing the number and proportions of endemic and common species in the three ligated groups at the OTU, phylum and genus levels. (f–h) Sobs, Ace and Chao indices at the OTU level showing microbiota richness among the three ligated groups. (i) Shannon index of the OTU level showing microbiota diversity among the three ligated groups. (j) Shannoneven index at the OTU level showing microbiota evenness among the three ligated groups. (k) Coverage index at the OTU level showing microbiota coverage among the three ligated groups. All analyses were conducted with n = 3 per group. Allo, allopurinol; P, periodontitis; Nu/Hu, normouricemia/hyperuricemia. Hu alters ligature microbial composition in periodontitis To assess differences in microbial community structure, PCA of β-diversity was performed. PCA showed a distinct separation in microbiome composition at the phylum level across the three groups (Figure 2a). A stacked bar chart further illustrated notable differences in the dominant taxa at the phylum level (Figure 2b). Detailed analyses of community composition within each group revealed that Hu treatment increased the abundances of Firmicutes and Proteobacteria and decreased those of Bacteroidota and Actinobacteriota, resulting in an elevated Firmicutes-to-Bacteroidota ratio in the ligature of periodontitis mice (Figure 2c–f). Conversely, allopurinol treatment decreased the abundances of Proteobacteria but increased that of Firmicutes in HuP mice (Figure 3a; Figure 2d–f). (a) Principal component analysis of periodontal microbiota profiles at the phylum level in silk ligatures, based on Bray-Curtis distance. (b) Comparison of the relative abundances of periodontal microbiota at the phylum level across the three ligated groups. (c) Linear discriminant analysis effect size was used to compare microbial composition between the HuP and NuP groups. Taxonomic classifications are as follows: p_, phylum; c_, class; o_, order; f_, family; g_, genus. (d–f) Relative abundances of Firmicutes and Bacteroidota and the ratio of Firmicutes to Bacteroidota, across the three ligated groups. Data are presented as medians. All analyses were conducted with n = 3 per group. Allo, allopurinol; P, periodontitis; Nu/Hu, normouricemia/hyperuricemia. (a) Linear discriminant analysis effect size was calculated to compare microbial composition between HuP+Allo and HuP groups. Taxonomic classifications are as follows: p_, phylum; c_, class; o_, order; f_, family; g_, genus. (b, c) Comparison of relative abundances of differential periodontal microbiota at the genus level between HuP and NuP. (d, e) Comparison of relative abundances of differential periodontal microbiota at the genus level between HuP+Allo and HuP. (f) Comparison of relative abundances of differential periodontal microbiota at the genus level among the three ligated groups. *, P < 0.05; All analyses were conducted with n = 3 per group. Allo, allopurinol; P, periodontitis; Nu/Hu, normouricemia/hyperuricemia. At the genus level, Hu significantly altered the microbial composition in ligated mice. Specifically, Hu increased the relative abundances of several genera, including Rodentibacter, Escherichia-Shigella, Proteus, and Delftia (all within the phylum Proteobacteria), as well as Veillonella, Streptococcus, Enterococcus, Lactobacillus, and Coprobacillus (all within the phylum Firmicutes). Conversely, Hu decreased the abundances of Bifidobacterium, Microbacterium, and Enterorhabdus (phylum Actinobacteriota), and Bacteroides, Bergeyella, and unclassified_f_Chitinophagaceae (phylum Bacteroidota) (Figure 3b, c). Allopurinol intervention, on the other hand, reduced the abundance of many Hu-related genera, such as Rodentibacter, Escherichia-Shigella, Enterococcus, and Coprobacillus (Figure 3d, e). Notably, in the ligature, eight genera were co-regulated by Hu and allopurinol intervention, with most genera showing increased relative abundance under Hu and decreased relative abundance following allopurinol treatment (Figure 3f). Ligature microbial composition is associated with UA and Cr in serum To verify whether Hu is a clinical driver that affects sample distribution, and to identify key species related to the disease, RDA and correlation heatmap analyses were conducted. Correlation analysis demonstrated that, of the 14 host-related factors examined, only serum levels of UA and Cr were correlated with the microbial composition in the ligature (Figure 4a–c). Correlation heatmaps further revealed significant associations between serum UA levels and the abundances of nine genera, and between serum Cr levels and those of eleven genera (Figure 4d). Notably, among the top five most abundant genera, Streptococcus showed significant positive correlations with both metabolites. (a–c) Redundancy analysis showed correlations between periodontal microbial composition and serum levels of uric acid, creatinine, urea nitrogen, glucose, lipids, as well as ligature UA content and body weight gain and organ weights. (d) Correlation heatmaps show the correlations between dominant microorganisms and serum levels of uric acid and creatinine in the HuP and NuP groups. *, P < 0.05; **, P < 0.01;***; P < 0.005. The proportion of all bacteria at the genus level is highlighted on the left side. All analyses were conducted with n = 3 per group. Cr, creatinine; eWAT, epididymal white adipose tissue; FSG, fasting serum glucose; HDL/LDL/VLDL-C, high/low/very low-density lipoprotein cholesterol; UA, uric acid; UN, urea nitrogen. Allo, allopurinol; P, periodontitis; Nu/Hu, normouricemia/hyperuricemia. Predicted functional alterations in ligature microbiota associated with Hu Based on pathway enrichment analysis of differentially expressed genes from our previous gingival mRNA sequencing data, Hu was found to significantly modulate several inflammatory pathways, including IL-17 and HIF-1α signalling. This prompted us to explore whether these host transcriptional alterations might be accompanied by changes in the predicted functional potential of the ligature-associated microbiota. To preliminarily investigate this possibility, we performed PICRUSt2-based functional prediction analysis of the 16S rRNA sequencing data using the KEGG database. The results suggested that Hu was associated with higher predicted abundances of KEGG pathways annotated as NOD-like receptor, MAPK, and FoxO signalling, together with a lower predicted abundance of the IL-17 signalling pathway (Figure 5a–f). Conversely, allopurinol treatment was associated with lower predicted abundances of KEGG pathways annotated as NOD-like receptor, IL-17, and PI3K-AKT signalling, and a higher predicted abundance of the HIF-1α signalling pathway in the ligature microbiota (Figure 5a–f). (a–f) Comparative analysis of six pro-inflammatory and pro-osteoclast PICRUSt2-predicted pathways, including NOD-like receptor (a), IL-17 (b), PI3K-AKT (c), HIF-1α (d), MAPK (e) and FoxO (f) signalling pathways, displayed as violin plots. All analyses were conducted with n = 3 per group. Allo, allopurinol; P, periodontitis; Nu/Hu, normouricemia/hyperuricemia. Discussion This exploratory pilot study provides the first systematic characterisation of the ligature microbiome in an experimental periodontitis animal model under asymptomatic Hu, aiming to elucidate whether and how asymptomatic Hu alters this niche-specific microbial community. Based on 16S rRNA gene profiling at the ligature site, we observed that Hu consistently reshaped the local microbiota, leading to increased α-diversity and evenness, along with a compositional shift toward Firmicutes and Proteobacteria. At the genus level, eight bacterial genera exhibited inverse regulation by Hu and allopurinol: most genera were upregulated under Hu and were reversed following allopurinol treatment. Moreover, Hu-associated alterations in ligature community structure were significantly correlated with higher serum UA and Cr levels in the Hu group. In summary, asymptomatic Hu remodels the ligature microbiome by increasing microbial diversity and evenness, driving compositional shifts toward Firmicutes and Proteobacteria, and enriching specific bacterial genera that are partially reversed by urate-lowering therapy. These findings indicate that Hu is associated with ligature-associated microbial dysbiosis in experimental periodontitis. A recent study identified an altered composition of the salivary microbiome, with an increased abundance of Prevotella intermedia, in patients with gout/Hu [25], suggesting a potential association between Hu and oral dysbiosis. However, this observation was derived from a non-periodontitis population. It remains unclear whether the influence of Hu on the oral microbial composition persists or changes in the context of established periodontal infection. The present study provides the first description of the periodontal microbiome in the context of Hu in periodontitis mice. Previous studies have shown that, compared with the control group, patients with periodontitis exhibited significantly increased diversity in their subgingival microbiota [6]. Furthermore, increased microbial diversity has been observed in ligature mice during the transition from gingival inflammation to periodontal bone destruction [26], suggesting that the destructive process of periodontitis may be associated with this increased diversity. In our study, Hu significantly increased both the diversity and evenness of communities in ligated mice. In addition, Hu was found to alter the β-diversity (i.e. community composition) of the periodontal microbiome in ligated mice, characterised by increased abundances of Firmicutes and Proteobacteria and decreased abundances of Bacteroidota and Actinobacteriota. Specifically, Hu led to a higher Firmicutes-to-Bacteroidota ratio in the periodontal microbiome, mirroring observations in the gut microbiome [27,28]. A higher Firmicutes-to-Bacteroidota ratio in the gut has been associated with intestinal dysbiosis, which is linked to various inflammatory diseases such as obesity and diabetes [29]. Similarly, a higher Firmicutes-to-Bacteroidota ratio in the salivary/periodontal microbiome has been associated with obesity and increased age [30,31]. Moreover, we found that Hu increased the abundance of Proteobacteria and decreased the abundance of Actinobacteriota in the periodontal microbiome. Similar shifts in the oral microbiome, specifically, an increase in Proteobacteria and a decrease in Actinobacteriota, have been observed in periodontitis mice [32]. Proteobacteria have been proposed as a potential microbial signature of disease, with increased abundance linked to metabolic disorders and inflammatory bowel diseases [33,34]. Bifidobacterium, a genus within the phylum Actinobacteriota, is the most abundant genus in this phylum and plays a crucial role in maintaining periodontal health and the success of treatment in patients with periodontitis [35]. In our study, Hu significantly decreased the proportion of the phylum Actinobacteriota (8.9% vs. 83.7%) and its genus Bifidobacterium (8.4% vs. 73.8%) in the context of periodontitis (HuP vs. NuP), suggesting that Hu drives the periodontal microbiome toward a pathogenic state. Regarding the effect of allopurinol on the periodontal microbiome, allopurinol treatment decreased the abundance of Proteobacteria in the periodontal microbiome of ligated mice with Hu. However, allopurinol did not reverse the increased Firmicutes-to-Bacteroidota ratio or the decreased abundance of Actinobacteriota induced by Hu, implying that allopurinol therapy may only partially mitigate the periodontal dysbiosis exacerbated by Hu. This observation could also explain the finding that allopurinol only partially reversed the exacerbated alveolar bone loss induced by Hu [21]. In the present study, among the 14 host-related factors examined, only changes in two metabolites (i.e. UA and Cr) in serum were found to correlate with the periodontal microbial composition. Redundancy analysis also showed a trend of positive association between serum UA levels and periodontal microbial compositions, suggesting that UA might directly contribute to periodontal dysbiosis. Previous reports have shown that UA and its precursor, purine, can serve as substrates incorporated into bacterial cell components [36]. Moreover, many Gram-positive (G+) and Gram-negative (G-) bacteria can produce purine-degrading enzymes to accelerate purine catabolism and increase UA production [36,37]. This implies that the production or presence of UA could favour bacterial growth or modify bacterial metabolism. Cr, detectable in the gingival crevicular fluid of both periodontally healthy and compromised individuals [38], was not detected in the silk ligature in this study. It remains uncertain whether Cr, if present in the ligature, exhibits similar changes to those observed in blood in the context of Hu. Nonetheless, it cannot be excluded that Cr may also play a role in periodontal dysbiosis. Many bacteria express creatinase and Cr deiminase, which can degrade Cr [39,40]. The breakdown products of Cr could provide a source of carbon and nitrogen, potentially promoting bacterial growth [41]. Regardless, both UA and Cr might contribute to the exacerbation of periodontal dysbiosis associated with Hu. The findings of the present study, together with accumulating epidemiological evidence, support the potential utility of UA and Cr as accessible biomarkers for periodontal risk assessment. UA is the major antioxidant in saliva, accounting for approximately 70% of its total antioxidant capacity [42]. However, under pathological conditions such as Hu, UA may also act as a pro-inflammatory mediator by activating the NLRP3 inflammasome, promoting M1 macrophage polarisation, and stimulating pro-inflammatory cytokine release [21]. This dual role highlights UA as a dynamic biomarker reflecting the transition from periodontal health to disease. In periodontitis, UA levels vary across biological matrices. Serum UA levels tend to be elevated, although usually within the normal range [20], whereas UA levels in gingival crevicular fluid and saliva are generally decreased [43]. This contrasting pattern may reflect active UA consumption or metabolism by periodontal pathogens, as the subgingival microbiota in periodontitis exhibits accelerated purine catabolism and enhanced purine-degrading enzyme activity [43]. In addition, systemic conditions such as smoking, kidney transplantation, and haemodialysis may influence UA and Cr levels in oral samples [44–46]. These observations suggest that UA and Cr levels in oral samples are not merely passive reflections of systemic concentrations, but are dynamically shaped by local inflammation, microbial activity, and host-related factors. Collectively, UA, Cr, or their ratio may serve as promising non-invasive biomarkers for periodontal disease, with potential for integration into point-of-care diagnostic platforms for early detection and personalised management. In our study, Streptococcus, a group of indigenous bacteria associated with the severity of periodontitis [47], was the only genus among the five most abundant genera that showed a positive correlation with changes in both UA and Cr levels in serum. These findings suggest that UA and Cr may impact the periodontal microbiota, possibly by regulating certain indigenous genera such as Streptococcus. However, the exact mechanisms by which UA and other potential candidate molecules promote periodontal dysbiosis require further exploration in future studies. In the present study, PICRUSt2-based functional prediction analysis suggested that Hu-associated alterations in the ligature microbiota were accompanied by changes in predicted microbial functional potential, including higher predicted abundance of KEGG pathways annotated as NOD-like receptor signalling and lower predicted abundance of the IL-17 signalling pathway. These predicted microbial functional shifts were consistent with our previously reported host transcriptomic findings in the same animal model, in which Hu was associated with activation of the NLRP3 inflammasome, a key component of NOD-like receptor signalling, and reduced transcript levels of several IL-17-related antimicrobial effectors, including β-defensins-3/-4/-6 and S100A9, in periodontal tissues of mice with periodontitis [21]. Together, these findings raise the possibility that Hu may be associated with coordinated changes in host inflammatory responses and the predicted functional potential of the ligature-associated microbiota. Reduced expression of antimicrobial effectors may contribute to impaired mucosal antimicrobial defence under Hu and may help explain the Hu-associated dysbiosis observed in this study, including increased diversity and higher relative abundances of Firmicutes and Proteobacteria. For example, β-defensin and S100A9 have been reported to inhibit Streptococcus growth [48,49], a major genus within Firmicutes that is enriched in gingivitis-associated communities [50]. Several limitations warrant acknowledgement in the present study. First, this study used a murine model, and known interspecies differences in the composition and ecological structure of the oral and systemic microbiota may limit the generalisability of the results to human periodontitis. Future work should therefore incorporate human-derived clinical samples to further validate and extend the microbiological links between Hu and periodontitis. Second, the sample size was relatively small (n = 3 per group), which limits the statistical power of between-group comparisons. However, rarefaction analysis suggested adequate sequencing coverage, and pan/core species analysis indicated that the major shared taxa were represented in this dataset. Accordingly, the microbiome findings should be regarded as exploratory and hypothesis-generating, with particular caution warranted for findings with borderline P-values. Future studies with larger sample sizes are warranted to further validate these microbial changes. Third, 16S sequencing analysis was performed only in male mice with experimental periodontitis, whereas periodontally healthy control groups (NuC and HuC) were not included as microbial baseline controls. Thus, the present design does not allow us to distinguish whether Hu directly alters the periodontal microbiome or indirectly reshapes it through aggravation of periodontitis. The absence of healthy controls also limits characterisation of microbial shifts from health to disease, while the use of male mice only precludes assessment of possible sex-related differences. Because the sequencing material in this study was ligature-associated microbiota, the silk ligature itself may substantially influence the local microbial ecosystem by providing a retentive surface for biofilm accumulation and creating a protected niche that favours microbial succession and dysbiosis. Accordingly, healthy control samples without matched long-term ligature exposure would not provide directly comparable ligature-associated microbial communities. Under non-ligated conditions, Hu might theoretically influence natural plaque accumulation through changes in the local inflammatory and immune environment, salivary or gingival crevicular fluid composition, or microbial purine metabolism; however, these possibilities remain speculative. Future studies including appropriate healthy controls and both sexes, together with non-ligature-based sampling approaches such as paper-point sampling of the gingival sulcus or analysis of gingival tissue-associated microbiota, are warranted to establish a healthy baseline ecological context and determine whether Hu affects the natural periodontal microbiome before the induction of periodontitis. In addition, microbiome analyses were restricted to samples from the silk-ligature site. Although the ligature model and this sampling strategy are among the most widely used approaches for characterising the murine periodontal microbiome [26,51,52], they do not fully recapitulate the ecological complexity of naturally occurring periodontitis, and more standardised, optimised sampling methods for the murine periodontal microbiome are still lacking [51]. Finally, PICRUSt2-based functional predictions only indicate putative alterations in specific metabolic and signalling pathways, and these in silico inferences require experimental validation using complementary approaches such as metabolomics and transcriptomics. Conclusion In conclusion, Hu was associated with dysbiotic remodelling of the ligature-associated microbiota in experimental periodontitis, characterised by increased α-diversity, an elevated Firmicutes/Bacteroidota ratio, and enrichment of potentially pathogenic taxa such as Streptococcus and Escherichia-Shigella. These microbial alterations were partially reversed by allopurinol treatment. In addition, serum UA and creatinine were associated with microbial compositional changes in this dataset, suggesting that they may be related to Hu-associated periodontal dysbiosis. PICRUSt2-based functional prediction further suggested accompanying changes in predicted microbial functional potential. Collectively, these findings support an association between systemic metabolic disturbance and local microbial imbalance in experimental periodontitis, and provide exploratory evidence of a potential microbiota-related link between Hu and periodontal tissue destruction. Acknowledgements All authors approved the final version for publication. Funding Statement This work was supported by Guangdong Basic and Applied Basic Research Foundation (no. 2023A1515030037) and Guangdong Medical Research Foundation (B2026206). Disclosure statement Ting Yu and Zhicong Wu have a patent PCT/CN2024/079709 pending. Ting Yu and Zhicong Wu have a patent ZL202310190961.2 granted. Other authors in this study declared no conflict of interest. Data availability statement The raw data that support the findings of this study are available from the corresponding author upon request. Ethical approval The animal study protocol was approved by the Animal Ethics Committee of the Guangdong Medical Experimental Animal Centre (Approval No. C202112-03). Informed consent statement Not applicable. References Associated Data Data Availability Statement The raw data that support the findings of this study are available from the corresponding author upon request.