The functional decline of bone marrow mesenchymal stem cells (BMMSCs) critically contributes to skeletal aging. However, rapid and label-free methods to assess this decline remain limited. Here, we establish a framework linking Raman spectroscopic fingerprints to the natural aging process in mice (young, middle-aged, and old), demonstrating age-related changes in molecular composition that corresponded with the loss of stem cell potency. Based on principal component analysis of spectral data, we developed an optimal convolutional neural network model that accurately predicts the age of BMMSCs. We identified a typical set of 20 spectral components that illustrate changes in proteins, lipids, and nucleic acids during aging. These spectral signatures correlate strongly with the cellular senescence marker and bone loss. Parallel single-cell RNA sequencing reveals transcriptomic reprogramming associated with inflammation, metabolism, and extracellular matrix remodeling, which covaries with spectral features. This spectral-transcriptomic association was successfully validated in independent cohorts. The typical spectral profiles accurately assessed the impact of inflammatory stimuli on stem cell and human tissues, and predicted bone loss in the ovariectomized mouse model. This integrative spectral fingerprinting approach offers a novel, label-free tool for diagnosing skeletal senescence and evaluating interventions targeting mesenchymal stem cells (MSCs) functionality.

Periodontitis is associated with microbial dysbiosis and an imbalanced host response. We investigated whether salivary large extracellular vesicles (EVs) associated bacterial DNA can reflect periodontal status and monitor changes after non-surgical periodontal therapy (NSPT). In our longitudinal cohort (18 healthy and 18 Stage Ⅲ/Ⅳ periodontitis), saliva was collected at baseline for all participants and at 3 and 6 months after NSPT for the periodontitis group. Clinical parameters, including plaque index (PI), bleeding on probing (BOP), probing pocket depth (PPD), and clinical attachment level (CAL), were recorded. Salivary large EVs were isolated, characterized and analyzed by RT-qPCR for bacterial genomic DNA from 11 oral pathogens. At baseline, the periodontitis group showed higher PI, BOP, and mean PPD than the healthy group (P<0.0001). After NSPT, PI, BOP, mean PPD, mean CAL, and percentage of deep pockets were lower at 3 and 6 months versus baseline, with no difference between follow-ups. Large EVs-associated DNA forPorphyromonas gingivalis, Fusobacterium nucleatum, Aggregatibacter actinomycetemcomitans, Tannerella forsythia, Prevotella intermedia, Eikenella corrodens, and Streptococcus mutans was higher in periodontitis than in health (P<0.01). Following NSPT, large EV-DNA levels for P. gingivalis, F. nucleatum, and A. actinomycetemcomitans declined at 3 months (P<0.05 to <0.0001), and P. gingivalis, F. nucleatum, and P. intermedia were reduced at 6 months relative to baseline (P<0.05 to <0.001). These findings indicate that bacterial DNA in salivary large EVs, particularly from P. gingivalis, F. nucleatum, and A. actinomycetemcomitans, can differentiate health from periodontitis and be useful for monitoring the response to NSPT. This suggests that bacterial gDNA in salivary large EVs may be useful as a non-invasive biomarker for monitoring periodontal activity and treatment outcomes, though validation in larger multicenter cohorts with longer follow-up is warranted to enable clinical translation.

The overuse of antibiotics has led to the emergence of drug-resistant bacteria, posing a serious threat to human health and increasing the medical burden. Metal-organic frameworks (MOFs) have shown great potential as enzyme mimics for antibacterial applications; however, their performance is often limited by the low density and poor accessibility of active sites. Herein, we report an innovative surface-active agent intercalation-assisted confinement growth strategy for constructing two-dimensional (2D) NH₂-BDC-Cu, Fe nanosheets (NSs) with highly exposed active site to enhance reactive oxygen species (ROS) generation, thereby improving the antibacterial efficiency against drug-resistant bacteria. We demonstrate that the synthesized 2D NH₂-BDC-Cu, Fe NSs exhibit 6.6-fold higher ROS production compared to NH₂-BDC-Cu, Fe nanopowders (NPs). Consequently, NH₂-BDC-Cu, Fe NSs achieve 100% antibacterial activity against both Staphylococcus aureus and Escherichia coli within 1 h, markedly outperforming NH₂-BDC-Cu, Fe NPs and previously reported biomaterials. Moreover,in vivo experiments confirm that NH₂-BDC-Cu, Fe NSs accelerate wound healing, achieving a 100% healing rate within 10 days, which is substantially higher than that of NH₂-BDC-Cu, Fe NPs. Overall, this work introduces a novel and efficient strategy for fabricating 2D MOF nanoenzymes with high active-site exposure, offering great promise for the development of advanced biomaterial-based antibacterial therapies.

This study evaluated the influence of print angle and layer thickness on the accuracy and fit of resin-based ceramic crowns fabricated using digital light processing (DLP). A mandibular first molar crown was virtually designed and then printed at two layer thicknesses (50 μm and 100 μm) and nine print angles (90°-270°). Accuracy was assessed by trueness, precision, and surface-resolved deviation metrics using 3D deviation analysis, whereas fit was evaluated with the triple-scan method. Both print angle and layer thickness significantly affected accuracy and fit (P<0.001). Trueness and fit values showed a nonlinear trend across print angles, with optimal results consistently observed at moderate inclinations of 150°-180°. Crowns printed at 50 μm showed better trueness and fit, whereas those printed at 100 μm showed more concentrated deviation distributions. Surface-resolved metrics revealed negative deviations at the marginal surfaces and positive deviations at the occlusal surfaces. Although all crowns showed clinically acceptable fit (<120 μm), the results indicate that a moderate print angle of 150°-180° combined with a 50 μm layer thickness provides the best balance between accuracy and clinical fit, while a 100 μm layer thickness may be preferable when printing consistency is prioritized.

Periodontitis is a highly prevalent chronic inflammatory disease, primarily driven by microbial dysbiosis and subsequent dysregulation of host immune responses. Therefore, targeting the pathogenesis, modulating the balance of periodontal microbiota, and subsequently controlling local inflammatory responses may offer an effective therapeutic strategy for periodontitis. Here, we propose a "bacteria-versus-bacteria" temporal therapeutic strategy and developed a genetically engineered strain HbEcN-PEI-MTZ@MnO₂ (HEpM) for periodontitis. This two-stage temporal therapy begins with pH-responsive release of metronidazole (MTZ) anchored on HEpM, eliminating periodontal pathogens and rebalancing the microbial dysbiosis. Concurrently, the engineered HEpM was endowed with a local oxygen supply system via the expression of Vitreoscilla hemoglobin (VHb), which enables HEpM to release oxygen as an oxygen-protective barrier forming a protective "anaerobic invisible cloak" that safeguards HEpM from the self-inflicted cytotoxicity of metronidazole. In the second therapeutic phase, remote thermal activation triggers the expression of anti-inflammatory enzymes to mitigate oxidative stress induced by microbial dysbiosis. Consequently, HEpM temporal therapy effectively reshaped the oral microbial community, reducing periodontal pathogens and biofilm formation. Single-cell RNA sequencing analysis further revealed a reduction in inflammatory CXCL1+ fibroblasts and enhanced fibroblast-centered intercellular communication, which facilitated immune microenvironment remodeling. This study establishes a smart, bacteria-based platform for temporal and targeted therapy of periodontitis, offering a novel paradigm for precision treatment of inflammatory diseases.

Anterior disc displacement (ADD) in the temporomandibular joint (TMJ) is highly prevalent and difficult to treat. It significantly affects craniofacial structures and functions, as well as orthodontic outcomes. It is a common and potentially detrimental comorbid condition in orthodontic treatment. There is a complex bidirectional relationship between orthodontic treatment and ADD, and this has led to clinical interest in the interplay between these entities and the corresponding diagnostic and therapeutic strategies. Hence, we performed this review to provide clinicians with an evidence-based framework for identifying potential risks and for formulating individualized orthodontic plans for patients with TMJ disorders. To achieve this aim, we synthesized current evidence on the development of the TMJ and mandible, their biomechanical coupling, and the orthodontic treatments used to improve the disc-condyle relationship, as well as their attendant risks. We also systematically analyzed factors that influence treatment efficacy and available interventions.

Nanotechnology holds great promise for disease therapy; however, the therapeutic efficacy of nanoparticles (NPs) is often limited by rapid recognition and clearance by the immune system. Recent studies have demonstrated that coating NPs with natural cell membranes imparts physicochemical properties closely resembling those of the source cells. This biomimetic coating also transfers key biological components, such as surface proteins and antigens, onto the NPs, endowing them with unique functionalities, including immune evasion, targeted delivery, and modulation of the immune microenvironment. These attributes underscore the significant potential of cell membrane-coated nanoparticles (CNPs) for a wide range of biomedical applications, including those in oral healthcare. Oral diseases are among the most prevalent non-communicable conditions globally, significantly affecting the quality of daily life and posing major public health challenges. As an emerging class of biological carriers, CNPs have been effectively utilized in the diagnosis and treatment of various oral diseases, including head and neck tumors and periodontitis. Owing to their high biocompatibility, targeted delivery capabilities, and extended circulation time, CNPs exhibit excellent therapeutic potential and offer a novel approach that complements conventional oral treatments. This review provides a comprehensive overview of the definition, classification, and fabrication methodologies of CNPs. It highlights recent advances in their diverse applications, with a particular focus on oral healthcare, critically examines current challenges, and concludes with perspectives to inform and guide future research.

Artificial intelligence (AI) is rapidly transforming dentistry through innovations in clinical practice, education, and research. Machine learning, and deep learning in particular, support disease diagnosis, treatment planning, prognosis prediction, and patient care. They do so by leveraging large-scale datasets from disciplines such as periodontology, endodontics, oral and maxillofacial surgery, orthodontics, and prosthodontics. This review first examines the technical and ethical foundations of dental AI, emphasizing data quality, representativeness, privacy protection, and standardized evaluation metrics. It then synthesizes current clinical applications, including AI-assisted diagnosis, image interpretation, surgical planning, robotic interventions, and predictive modeling of treatment outcomes. Educational uses of AI are examined in the context of simulation-based training, virtual reality integration, automated assessment, and patient education. The role of AI in scientific discovery is discussed with respect to data mining, biomarker identification, drug design, literature synthesis, and peer review support. Despite its potential, dental AI adoption faces challenges including limited and biased datasets, interoperability barriers, model interpretability, ethical concerns, and insufficient user training. Future integration will require rigorous external validation, regulatory compliance, seamless workflow incorporation, and interdisciplinary collaboration to ensure safety, transparency, and clinical utility. By addressing these limitations, AI can serve as a robust, multidimensional decision-support tool to improve diagnostic accuracy, personalizing care, and advancing dental education and research.