Oral microbiome collapse marks diabetic complications beyond glycemic control

Background The metabolic memory phenomenon demonstrates that diabetic complications can progress independently of contemporaneous glycemic control, yet the upstream drivers sustaining this progression remain poorly defined. Whether the microbiome-metabolome axis participates in such post-glycemic mechanisms is unknown. Methods We enrolled 154 participants across three disease stages: prediabetes (Pre. DM, n = 51), type 2 diabetes without complications (T2DM, n = 53), and T2DM with complications (T2DM+Comp., n = 50). Multi-omics profiling included gut metagenomics (132 samples, 338 genera), oral metagenomics (108 samples, 254 genera), serum metabolomics (126 samples, 9, 004 compounds), and 64 clinical parameters. A four-way paired subset of 91 participants (Pre. DM = 41, T2DM = 29, T2DM+Comp. = 21) underwent integrated analyses spanning alpha/beta diversity, differential abundance testing with Benjamini-Hochberg correction, Spearman-based mediation analysis with 5, 000 bootstrap iterations, and nested leave-one-out cross-validated random forest classification with 1, 000-iteration permutation testing, in which every feature-selection step was recomputed inside each cross-validation fold. Results At the T2DM to T2DM+Comp. transition, glycemic parameters showed no significant differences (HbA1c P = 0.156, fasting glucose P = 0.059, 2h postprandial glucose P = 0.134), yet oral microbiome beta diversity separated significantly (PERMANOVA P = 0.015) while gut beta diversity did not ( P = 0.196). Oral Bacteroides exhibited a precipitous decline across disease stages (log 2 fold change = −2.60 at T2DM to T2DM+Comp.), an effect independent of metformin use. The serum metabolome underwent large-scale remodeling (1, 536 FDR significant compounds among 2, 549 tested in the full cohort). Mediation analysis associated oral Bacteroides with disease progression through multiple metabolites, with all 30 triads achieving bootstrap significance and mediation proportions ranging from 42.5% to 118.5% (median = 72.9%) against total effects of |r| = 0.49–0.51 ( P < 1 × 10 -5 ). Diagnostic models discriminated disease stages with the serum metabolome carrying the dominant signal (metabolite-only model: accuracy = 96.7%, AUC = 0.998; combined model: 95.6%, AUC = 0.998; oral microbiome: 70.3%, AUC = 0.877; gut microbiome: 51.6%, AUC = 0.638), and permutation testing confirmed statistical significance ( P < 0.001, 1, 000 iterations). A leakage audit showed that full-dataset feature pre-filtering inflated apparent accuracy by only 1.1–2.2 percentage points for the metabolite-based models and by at most 6.6 percentage points overall; the fully nested estimates are reported throughout. Conclusions This study provides multi-omics evidence that oral microbiome collapse and serum metabolome remodeling persist and amplify beyond the point where glycemic parameters plateau, consistent with a post-glycemic maintenance mechanism. The oral Bacteroides -to-metabolite-to-disease progression mediation pathway, combined with the diagnostic capacity of multi-omics signatures (AUC ≈ 1.0), identifies the microbiome-metabolome axis as a candidate non-glycemic amplifier of complication progression and a potential target for early detection.

Authors

Institutions

Publication Details

Journal
Frontiers in Endocrinology
Published
2026-11-13
DOI
https://doi.org/10.3389/fendo.2026.1988886
Primary Topic
Oral microbiology and periodontitis research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Oral microbiome collapse marks diabetic complications beyond glycemic control

Kunxin Wang, Jiawei Jiang, Ying Li, Ruifang Liu et al.
5 links
Frontiers in Endocrinology
Oral microbiology and periodontitis research
article

Oral microbiome collapse marks diabetic complications beyond glycemic control

Kunxin Wang, Jiawei Jiang, Ying Li, Ruifang Liu, Yingrong Liu, Zhaoyang Yang
article en

Abstract

Background The metabolic memory phenomenon demonstrates that diabetic complications can progress independently of contemporaneous glycemic control, yet the upstream drivers sustaining this progression remain poorly defined. Whether the microbiome-metabolome axis participates in such post-glycemic mechanisms is unknown. Methods We enrolled 154 participants across three disease stages: prediabetes (Pre. DM, n = 51), type 2 diabetes without complications (T2DM, n = 53), and T2DM with complications (T2DM+Comp., n = 50). Multi-omics profiling included gut metagenomics (132 samples, 338 genera), oral metagenomics (108 samples, 254 genera), serum metabolomics (126 samples, 9, 004 compounds), and 64 clinical parameters. A four-way paired subset of 91 participants (Pre. DM = 41, T2DM = 29, T2DM+Comp. = 21) underwent integrated analyses spanning alpha/beta diversity, differential abundance testing with Benjamini-Hochberg correction, Spearman-based mediation analysis with 5, 000 bootstrap iterations, and nested leave-one-out cross-validated random forest classification with 1, 000-iteration permutation testing, in which every feature-selection step was recomputed inside each cross-validation fold. Results At the T2DM to T2DM+Comp. transition, glycemic parameters showed no significant differences (HbA1c P = 0.156, fasting glucose P = 0.059, 2h postprandial glucose P = 0.134), yet oral microbiome beta diversity separated significantly (PERMANOVA P = 0.015) while gut beta diversity did not ( P = 0.196). Oral Bacteroides exhibited a precipitous decline across disease stages (log 2 fold change = −2.60 at T2DM to T2DM+Comp.), an effect independent of metformin use. The serum metabolome underwent large-scale remodeling (1, 536 FDR significant compounds among 2, 549 tested in the full cohort). Mediation analysis associated oral Bacteroides with disease progression through multiple metabolites, with all 30 triads achieving bootstrap significance and mediation proportions ranging from 42.5% to 118.5% (median = 72.9%) against total effects of |r| = 0.49–0.51 ( P < 1 × 10 -5 ). Diagnostic models discriminated disease stages with the serum metabolome carrying the dominant signal (metabolite-only model: accuracy = 96.7%, AUC = 0.998; combined model: 95.6%, AUC = 0.998; oral microbiome: 70.3%, AUC = 0.877; gut microbiome: 51.6%, AUC = 0.638), and permutation testing confirmed statistical significance ( P < 0.001, 1, 000 iterations). A leakage audit showed that full-dataset feature pre-filtering inflated apparent accuracy by only 1.1–2.2 percentage points for the metabolite-based models and by at most 6.6 percentage points overall; the fully nested estimates are reported throughout. Conclusions This study provides multi-omics evidence that oral microbiome collapse and serum metabolome remodeling persist and amplify beyond the point where glycemic parameters plateau, consistent with a post-glycemic maintenance mechanism. The oral Bacteroides -to-metabolite-to-disease progression mediation pathway, combined with the diagnostic capacity of multi-omics signatures (AUC ≈ 1.0), identifies the microbiome-metabolome axis as a candidate non-glycemic amplifier of complication progression and a potential target for early detection.

Frontiers in EndocrinologyVol. 17
Fujian University of Traditional Chinese Medicine (CN), Gannan Medical University (CN)
Good health and well-being
Oral microbiology and periodontitis research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon