In recent years, magnesium hydroxide has been widely studied due to its bioactivity and biocompatibility. The bactericidal effects of magnesium hydroxide nanoparticles on oral bacteria have also been reported. Therefore, in this study, we investigated the biological effects of magnesium hydroxide nanoparticles on inflammatory responses induced by periodontopathic bacteria. Macrophage-like cells, namely J774.1 cells, were treated with LPS derived from Aggregatibacter actinomycetemcomitans and two different sizes of magnesium hydroxide nanoparticles (NM80/NM300) to evaluate their effects on the inflammatory response. Statistical analysis was performed using an unresponsive Student’s t-test or one-way ANOVA followed by Tukey’s post hoc test. NM80 and NM300 inhibited the expression and secretion of IL-1β induced by LPS. Furthermore, IL-1β inhibition by NM80 was dependent on the downregulation of PI3K/Akt-mediated NF-κB activation and the phosphorylation of MAPK molecules such as JNK, ERK1/2, and p38 MAPK. By contrast, only the deactivation of the ERK1/2-mediated signaling cascade is involved in IL-1β suppression by NM300. Although the molecular mechanism involved varied with size, these results suggest that magnesium hydroxide nanoparticles have an anti-inflammatory effect against the etiologic factors of periodontopathic bacteria. These properties of magnesium hydroxide nanoparticles can be applied to dental materials.

Large language models (LLMs) show varying accuracy on professional examinations, particularly for questions involving visual materials. This study compared GPT-4o, o3, GPT-5 Thinking, and Gemini 3 on 400 Japanese National Dental Technician Examination questions from 2018–2022 and investigated whether between-model differences varied by question format. The dataset comprised 252 text-only questions and 148 questions accompanied by visual materials. Accuracy was compared using Cochran’s Q test and Holm-adjusted exact McNemar tests. A question-level mixed-effects logistic regression assessed model-by-item-type interactions while accounting for examination year. Overall accuracy was 63.8% for GPT-4o, 70.8% for o3, 74.8% for GPT-5 Thinking, and 80.5% for Gemini 3 (Cochran’s Q, p < 0.001). Gemini 3 significantly outperformed the other models, whereas o3 and GPT-5 Thinking did not differ significantly after adjust-ment. All models showed lower accuracy on visual-material questions than on text-only questions. The overall model-by-item-type interaction was significant (likelihood-ratio χ² = 11.756, p = 0.008), indicating that between-model differences varied by question format. The relative advantages of GPT-5 Thinking and Gemini 3 over GPT-4o were at-tenuated for visual-material questions. These findings highlight format-dependent dif-ferences in LLM performance and the limitations of aggregate accuracy as a bench-marking metric.