Zinc finger protein 804A (ZNF804A) has been identified by genome‑wide association studies as a robust risk gene in schizophrenia, but how ZNF804A contributes to schizophrenia and its upstream regulation remains unknown. Previous studies have indicated that microRNAs (miRs) are key factors that regulate the expression levels of their target genes. The present study revealed significantly increased expression of miR‑148b‑3p in the peripheral blood of patients with first‑onset schizophrenia compared with healthy controls, and bioinformatics analysis predicted that the ZNF804A gene is a target of miR‑148b‑3p. Therefore, the present study investigated the possible upstream regulation of ZNF804A by miR‑148b‑3p in the human neuroblastoma SH‑SY5Y cell line, and assessed the implications for schizophrenia. The results revealed significantly reversed expression levels of miR‑148b‑3p (P=0.0051) and ZNF804A (P=0.0218) in the peripheral blood of patients with first‑onset schizophrenia compared with healthy individuals. Furthermore, it was demonstrated that miR‑148b‑3p directly targeted ZNF804A via binding to conserved target sites in the 3'‑untranslated region of ZNF804A mRNA, where it inhibited the endogenous expression of ZNF804A at both the mRNA (P=0.048) and protein levels (P=0.013) in SH‑SY5Y cells. Furthermore, miR‑148b‑3p was revealed to regulate the expression levels of catechol‑O‑methyltransferase (COMT) and serine protease 16 (PRSS16) by targeting ZNF804A in SH‑SY5Y cells. Collectively, the present results indicated that there was a direct upstream regulation of the schizophrenia risk gene ZNF804A by miR‑148b‑3p, which contributed to the regulation of the downstream genes COMT and PRSS16. Thus, the miR‑148b‑3p/ZNF804A/COMT/PRSS16 pathway may play an important role in the pathophysiology of schizophrenia, and may serve as a potential target in drug discovery and gene therapy for this disorder.

Accumulated evidence has demonstrated that the gut microbiome can contribute to pain modulation through the microbiome–gut–brain axis. Various relevant microbiome metabolites in the gut are involved in the regulation of pain signaling in the central nervous system. In this review, we summarize recent advances in gut–brain interactions by which the microbiome metabolites modulate pain, with a focus on orofacial pain, and we further discuss the role of gut–brain crosstalk in the central mechanisms of orofacial pain whereby the gut microbiome modulates orofacial pain via the vagus nerve-mediated direct pathway and the gut metabolites/molecules-mediated indirect pathway. The direct and indirect pathways both contribute to the central regulation of orofacial pain through different brain structures (such as the nucleus tractus solitarius and the parabrachial nucleus) and signaling transmission across the blood-brain barrier, respectively. Understanding the gut microbiome-regulated pain mechanisms in the brain could help us to develop non-opioid novel therapies for orofacial pain.

Background/Objectives: Previous studies have demonstrated an association between migraine headache and temporomandibular joint disorders (TMDs), with a higher prevalence of TMD symptoms in patients with migraine. Methods: In this study, we conducted RNA sequencing to identify differentially expressed genes (DEGs) in the spinal trigeminal nucleus caudalis of mice with migraine-like pain and/or myogenic TMD. Results: We observed 204 upregulated and 274 downregulated genes in the comorbid migraine and TMD group compared to the control group. We identified 15 ferroptosis-related DEGs enriched in the pathways of neurodegeneration, cellular homeostasis, interleukin signaling, and pain response. Gene Ontology analysis highlighted the involvement of neuroinflammatory response and monoamine transmembrane transporter activity, while Gene Set Enrichment analysis showed enrichment in chemokine signaling, cell cycle, and calcium signaling pathways. Immune infiltration analysis identified M0 macrophages, immature dendritic cells, neutrophils, and eosinophils as key responders. Hub genes in the protein–protein interaction network included Gm7536, Rpl17, Rpl22l1, Rpl14, Rps8, Rps29, Rpl35, Gm4889, Gm11808, Rps27rt, Rps12-ps3, Rpl10-ps3, Gm9843, Oas1c, Il1b, and Serpine1, indicating their synergistic roles in such orofacial pain comorbidity. Conclusions: Our results suggest that the comorbid migraine and TMD can regulate gene expressions involving ferroptosis and immune cell responses and the identified DEGs could be targeted to develop novel therapies for this painful comorbidity.

Pain is the most common symptom of temporomandibular joint (TMJ) disorders, which present significant clinical challenges due to their complexity and limited treatment options. Our previous study demonstrates that gut microbiome-derived butyrate is critical for the modulation of TMJ pain. In this study, we investigated its underlying mechanisms, and we found that oral administration of tributyrin, a prodrug of butyrate, not only significantly alleviated TMJ pain but also reversed the reduction in histone acetylation in the spinal trigeminal nucleus caudalis (Sp5C) under the TMJ pain condition. Using single-cell multi-omics sequencing, we profiled gene expression and chromatin accessibility in the Sp5C cells at the single-cell resolution. Bioinformatics analysis revealed that TMJ pain disrupted both the expression and chromatin accessibility of Nop14, Matk, Idh3b, Ndst2, and Tomm6 across four cell types in the Sp5C, and these alterations were reversed by tributyrin treatment. Specifically, Nop14 exhibited increased chromatin accessibility at its promoter region under TMJ pain condition, and knockdown of Nop14 in the Sp5C restored histone acetylation and alleviated TMJ pain. Together, our findings reveal cell-type–specific gene regulation that underlies butyrate-mediated epigenetic regulation of TMJ pain, which suggesting that targeting gut microbiome metabolites could develop a non-opioid novel therapy for TMJ disorders.

Preclinical orofacial pain research produces unusually diverse data. Facial expression, jaw use, feeding, stimulus-evoked responses, trigeminal neural activity, tissue imaging, biomechanics and molecular profiles each capture a different part of the pain process. Artificial intelligence (AI) is often discussed in this field as if it were synonymous with automated grimace scoring. That view is too narrow. Direct applications are concentrated in video-based phenotyping, whereas many other experimental systems are better described as AI-ready. To distinguish opportunity from accomplishment, this review separates direct orofacial evidence from task-level precedents in computational ethology, non-orofacial pain, systems neuroscience, digital pathology, joint imaging, single-cell modeling, physics-guided biomechanics and organoid screening. Each transfer claim is framed as an evidence bridge: an existing orofacial experiment, an external AI precedent, a defined input-output task and a validation requirement in the target system. We examine facial and functional phenotyping, tissue imaging, trigeminal calcium imaging, single-cell and spatial omics, multimodal state estimation, therapeutic discovery and closed-loop experimentation. We also define what should not be labeled AI, including routine clustering, differential-expression testing and conventional finite-element analysis. Two forward-looking concepts receive particular attention. A virtual pain cell would predict cell-state responses to perturbations in trigeminal neurons, satellite glia, fibroblasts or immune cells. A virtual temporomandibular joint would couple geometry and mechanics to tissue remodeling, nociceptor activity and pain-related behavior. Neither system yet exists in a validated form. Their value will depend on prospective perturbation tests, animal-level validation, uncertainty reporting and performance against simple baselines. AI can move the field from isolated measurements toward integrated, testable models, but only when external precedents are treated as hypotheses for transfer rather than evidence that the transfer has already succeeded.