Oral-gut axis in inflammation: periodontitis exacerbates ulcerative colitis via microbial dysbiosis and barrier disruption

Seven-week-old male C57BL/6J mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. and housed under specific pathogen-free (SPF) conditions at the Laboratory Animal Center of Nanjing Agricultural University. All animal experiments adhered to ethical guidelines and were approved by the Laboratory Animal Welfare and Ethics Committee of Nanjing Agricultural University (No. PZW2024032, China). Based on our previous work and existing literature in the field [10, 11], and in adherence to the 3Rs principle of animal welfare (Replacement, Reduction, and Refinement), we chose to use 5 mice per group to ensure both ethical responsibility and adequate statistical power.

After one week of acclimatization, 20 mice were randomly assigned to four groups: (1) group Sham(n = 5), receiving regular drinking water for 30 days; (2) group P(n = 5), subjected to ligature-induced periodontitis starting on day 7 and maintained for the following 23 days; (3) group UC(n = 5), in which UC was induced by receiving alternating 1% and 0.5% w/v DSS for 30 days; and (4) group P-UC(n = 5), receiving the same DSS regimen as the UC group, with the same ligature-induced periodontitis as the P group.

After grouping, animals in each group were not allowed to enter cages belonging to the other groups. All lids, cages, bedding, bottles, and water were sterilized before use. Animals severely injured from fighting or having lost teeth with periodontitis were excluded.

Experimental periodontitis was induced by ligature. Briefly, mice were anesthetized via intraperitoneal injection of tribromoethanol (Jitian, China) according to body weight. After achieving full anesthesia, 5 − 0 silk sutures (Jinhuan, China) was ligated around the left maxillary second molars [12]. Following the procedure, mice were placed on a heating pad for recovery. The ligatures were checked every other day. Once they fell off, the tooth was ligated again immediately.

The chronic ulcerative colitis model was induced by administering 1% w/v dextran sulfate sodium (DSS; MP Biomedical) in drinking water. Mice received 1% DSS for 16 days, followed by 0.5% DSS for 7 days, and then 1% DSS for another 7 days [13].

Mice were anesthetized via intraperitoneal injection of 1.25% tribromoethanol (0.2 mL/10 g body weight). Blood samples (∼ 0.5 mL) were collected from the orbital venous sinus. After blood sampling, mice were anesthetized via intraperitoneal injection of sodium pentobarbital (150 mg/kg). After confirming full unconsciousness, mice were humanely euthanized by cervical dislocation under anesthesia. Following euthanasia, the maxillary bone and surrounding tissues were harvested and fixed in 4% neutral paraformaldehyde. Ligatures were collected from the P and P-UC groups, while in the Sham and UC groups, sterile 5 − 0 silk sutures were briefly inserted into the gingival sulcus before removal. All samples were stored at -80 °C. The thoracic and abdominal cavities were opened to collect the cecum and proximal colon, measuring their distance and weight. Approximately 2 mm of proximal colon tissue was preserved in RNA stabilization solution, while a 5 mm colon segment was washed and fixed in 4% neutral paraformaldehyde. Cecal contents were stored at -80 °C. To minimize bias, group allocation and treatment were performed by a single unblinded investigator. All subsequent sample collection, tissue processing, and outcome assessments were conducted by personnel blinded to group identity. Histological evaluation was performed by two independent, blinded pathologists.

Maxillae were fixed in 4% paraformaldehyde for 72 h before micro–Computed Tomography (CT) scanning (SkyScan 1172, Bruker, Belgium). The scanning voxel resolution was 18 μm. Two-dimensional images were reconstructed into three-dimensional models using NReconstruction (version 1.4.4, Bruker). The distance from the cemento-enamel junction to the alveolar crest was measured using DataViewer (version 1.5.4.0, Bruker) to assess bone resorption.

The maxillae were decalcified in 10% EDTA solution (pH 7.2; Servicebio), with the solution changed every 3 days for 4 weeks. Sections were then prepared for hematoxylin and eosin (HE) staining. Images were captured using a slide scanner and analyzed with CaseViewer software.

DNA was extracted from cecal contents using a fecal DNA extraction kit (OMEGA Soil DNA Kit (M5635-02), Omega Bio-Tek, Norcross, GA, USA). The integrity and quantity of extracted DNA were initially verified by 2% agarose gel electrophoresis and purified using an AxyPrep DNA gel extraction kit. PCR amplification of the 16S rRNA gene (V3-V4 region) was performed using barcoded universal primers 338 F (5′-ACTCCTACGGGAGGCAGCA-3′) and 806R (5′-GGACTACHVGGGTWTCTAAT-3′). Amplification products were confirmed via 2% agarose gel electrophoresis and subsequently purified using a silica column-based gel extraction kit. Sequencing libraries were constructed without additional PCR enrichment to minimize amplification bias, and library quality was assessed using a Bioanalyzer system. Paired-end sequencing (2 × 250 bp) was conducted on an Illumina MiSeq platform, yielding high-throughput data for downstream analysis of gut microbial community structure.

Samples were homogenated for 1 min with 500 µL of water and 100 mg of glass beads, and then centrifuged at 4℃ for 10 min at 12,000 rpm. 200 µL supernatant was extracted with 100 µL of 15% phosphoric acid and 20 µL of 375 µg/mL 4-methylvaleric acid solution as IS and 280 µL ether. Subsequently, the samples were centrifuged at 4℃ for 10 min at 12,000 rpm after vortexing for 1 min and the supernatant was transferred into the vial prior to GC-MS analysis. The GC analysis was performed on trace 1310 gas chromatograph (Thermo Fisher Scientific, USA). The GC was fitted with a capillary column Agilent HP-INNOWAX (30 m × 0.25 mm ID × 0.25 μm) and helium was used as the carrier gas at 1 mL/min. Injection was made in split mode at 10:1 with an injection volume of 1 µL and an injector temperature of 250℃. Mass spectrometric detection of metabolites was performed on ISQ LT (Thermo Fisher Scientific, USA) with electron impact ionization mode. Single ion monitoring (SIM) mode was used with the electron energy of 70 eV. Extracted-ion chromatograms for each ion were abstracted based on the information of diagnostic ions and quantification ions. Peak areas would be detected by retention times for each compound. The calibration curves were obtained as plots of the peak area ratio of the target compounds to an internal standard versus the target compound concentration and calculate the concentration in samples.

Raw sequencing data were processed using QIIME (version 1.9.1, University of Colorado, USA). Paired-end reads were assembled, quality-controlled, and filtered. Sequences were assigned to operational taxonomic units (OTUs) at 97% similarity using UPARSE (version 7.1). Chimeric sequences were removed using UCHIME. Taxonomic classification was performed using RDP Classifier with the Greengenes2 database (70% confidence threshold). Bray-Curtis distances were calculated for community structure analysis. Statistical comparisons were conducted using the Wilcoxon rank-sum test. Linear discriminant analysis (LDA) was performed to identify differential taxa, with results visualized using R (version 3.2.5, University of Auckland, New Zealand).

Data analysis was performed using SPSS 28 (IBM, NY, USA) and GraphPad Prism 9 (GraphPad Software, CA, USA). Results are presented as mean ± standard deviation. Group comparisons were conducted using independent t-tests and Wilcoxon rank-sum tests. For multi-group comparisons, one-way ANOVA followed by post-hoc multiple comparisons or Kruskal-Wallis tests were applied. Pearson correlation analysis was used to assess associations between variables. A p-value < 0.05 was considered statistically significant. All graphical representations were generated using GraphPad Prism 9.

The ligatures remained intact throughout the experiment in both the P and P-UC groups. Micro-CT analysis revealed significant alveolar bone resorption around the maxillary second molars in both P and P-UC groups, with greater distances from the alveolar crest to the cementoenamel junction (CEJ) at the mesial, distal, buccal, and palatal sites compared to the Sham group (P < 0.0001), confirming the successful establishment of the periodontitis model (Fig. 1A, B). H&E staining further validated the model, showing gingival recession below the CEJ and inflammatory cell infiltration in the gingiva of both the P and P-UC groups (Fig. 1C).

Current acute UC models which are widely used in mice employ 3% DSS for 7–10 days. While they replicate some acute pathological features but fail to reflect the alternating active and remission phases characteristic of the disease [14, 15] making them unsuitable for investigating the long-term impact of periodontitis on intestinal inflammation.

Chronic colitis models involve either prolonged administration of low-dose DSS or repeated cyclic exposure to DSS, which typically consists of 7 days of 2–5% DSS treatment followed by 14 days of water, repeated for 4–8 cycles [14]. Unlike acute models, chronic inflammation in these models is characterized by mononuclear leukocyte infiltration, crypt architectural disarray, increased lymphocytic infiltration in the submucosa, and an increased crypt base-to-muscularis mucosae distance [16, 17]. However, during alternating administration of DSS solution and water, higher DSS concentrations may results in increased mortality, and mice tend to recover rapidly during water consumption periods, limiting long-term intestinal inflammation.

To address these limitations, we developed a novel chronic UC model using alternating 1% and 0.5% DSS solutions. This approach more accurately simulated clinical UC cycles while improving the animal survival. HE staining of colonic tissues in both UC and P-UC groups confirmed that the newly established model closely resembles the histopathological features of chronic UC models [18–20]. Importantly, the 0.5% DSS solution maintained a chronic inflammatory state during remission phases, enabling the investigation of periodontitis-driven effects under sustained UC burden. Additionally, the lower DSS concentrations reduced colonic damage, potentially amplifying the effects of periodontitis on UC while also serving as a valuable reference for future model selection in studies examining the systemic impact of periodontitis.

Body weight loss, a key UC model indicator, was greater in the P-UC group than in the UC group, along with higher DAI scores, suggesting more severe colonic inflammation. Furthermore, the colonic inflammation is associated with immune cell infiltration, enhanced inflammatory response, and gut microbiota dysbiosis [21]. Histological analysis and elevated levels of pro-inflammatory cytokines, along with increased intestinal permeability, further support the notion that periodontitis aggravates UC. Correlation analysis also revealed a significant positive association between the two conditions. Further analysis indicated that disruptions in the intestinal barrier and gut microbiota may be critical mediators of periodontitis-induced UC exacerbation, as demonstrated by the following findings:

Although this study established a clinically relevant chronic UC model and revealed that periodontitis may exacerbate UC by disrupting the intestinal barrier and altering gut microbiota, several limitations should be acknowledged. First, microbial analyses were limited to cecal contents, which may not fully represent microbial diversity across different intestinal regions or timepoints. Longitudinal and spatially resolved sampling in future studies would offer deeper insight into the dynamic progression of gut dysbiosis associated with periodontitis. Second, while previous evidence and our findings suggest a link between periodontitis and UC, the therapeutic implications of periodontal treatment for UC remain unclear. Well-designed randomized controlled trials are needed to determine whether managing periodontitis can improve clinical outcomes in UC patients. Third, the mechanisms underlying the oral-gut axis remain poorly defined. It is unclear whether periodontitis influences UC primarily through the entry of periodontal pathogens or inflammatory mediators into systemic circulation via the ulcerated pocket epithelium [11], or through alterations in the salivary microbiota that subsequently reshape the gut microbial ecosystem [32]. Further studies are needed to clarify whether microbial or immunological pathways dominate this process. Specifically, future research should identify the key bacterial species or consortia involved, or, in the case of immune-driven mechanisms, delineate the critical immune phases, responsible cell subsets, and signaling cascades mediating this interplay.

The study received approval from the Ethics Committee of Nanjing Agricultural University (No. PZW2024032).

Not applicable.

The authors declare no competing interests.

Sequence data that support the findings of this study have been deposited in the National Library of Medicine Archive with the primary accession code PRJNA1244589 and PRJNA1244561.