Integrated transcriptomic profiling combined with in vitro validation reveals the involvement of TMEM140 in the link...

Periodontitis (PD) is a prevalent chronic inflammatory disorder in adults, and moderate-to-severe PD (Stage II-III/IV) may accelerate brain aging and neurodegenerative changes via the peripheral-central immune-neural axis, although the molecular connections and mechanisms of interaction have yet to be fully elucidated. This study sought to identify senescence-associated molecules potentially shared by PD and Alzheimer's disease (AD) using integrated transcriptomic analysis, machine learning, and in vitro RNA interference assays, and to further assess the role of TMEM140 in linking PD to brain aging.

Transcriptomic datasets related to PD and AD were retrieved from the GEO database, and differential gene expression analysis was performed following batch effect correction; shared aging-associated genes were subsequently identified by combining weighted gene co-expression network analysis (WGCNA) with aging gene databases (HAGR and aging Atlas). Four machine learning algorithms, namely random forest (RF), support vector machine (SVM), generalized linear model (GLM), and extreme gradient boosting (XGB), were further applied to identify key genes, and their diagnostic value was assessed using receiver operating characteristic (ROC) analysis and nomogram models. DSigDB was used to predict candidate small-molecule compounds. In the in vitro experiments, a Porphyromonas gingivalis lipopolysaccharide (PG-LPS)-induced inflammatory model in human gingival fibroblasts (HGFs) and an Aβ1-42 and D-galactose-induced senescence model in SH-SY5Y neuron-like cells were established; TMEM140 in SH-SY5Y cells was then silenced using small interfering RNA (siRNA), and the neuron-like cells were treated with the same batch of standardized conditioned medium (CM; prepared from the supernatant of PG-LPS-treated HGFs) to observe changes in cellular responses to inflammatory stimulation after TMEM140 downregulation.

Seven aging-related genes common to PD and AD were identified, and comprehensive analysis using multiple algorithms selected TMEM140, TIMP1, and ALDH2 as key genes. Notably, TMEM140 was upregulated in PD and downregulated in AD, showed significant correlations with plasma cell and γδ T-cell infiltration, and single-cell analysis further revealed its cell type-specific expression in distinct brain cell subsets. In vitro experiments demonstrated that PG-LPS treatment markedly increased TMEM140 expression in HGFs, whereas treatment with Aβ1-42 and D-galactose reduced TMEM140 expression in neuron-like cells. When exposed to the same batch of conditioned medium, neuron-like cells with TMEM140 knockdown displayed more evident injury and senescence-related phenotypes, including reduced cell viability, increased reactive oxygen species (ROS) production, a higher percentage of senescence-associated β-galactosidase (SA-β-Gal)-positive cells, and marked upregulation of IL-1β, IL-6, TNF-α, p16, p21, RELA, NFKBIA, and TP53, indicating that reduced TMEM140 expression may contribute to enhanced susceptibility of neuron-like cells to inflammatory stress.

Through integrated transcriptomic analysis together with in vitro experimental validation, this study indicates that TMEM140 may be a candidate bridge molecule connecting PD and AD comorbidity. TMEM140 may participate in shaping the peripheral-central immunosenescence network and contribute to the cross-system transmission of inflammatory signaling.