[This corrects the article on p. 350 in vol. 18, PMID: 39822678.].
Head and neck cancers represent a critical global health issue, contributing to substantial morbidity and mortality. Recent research has explored the role of microRNAs (miRNAs) in these cancers by constructing miRNA-associated disease networks using bipartite graphs. Graph attention networks (GATs) have emerged as a powerful tool for predicting disease associations within such biological networks, offering enhanced accuracy in identifying potential miRNA-disease relationships. This study employs GATs to uncover and predict potential miRNA contributors to head and neck cancers. Data on miRNA-disease associations were sourced from the HMDD v4.0 database, a platform based on SQLite and Django. The head and neck neoplasms dataset included miRNA, disease, causality, category, and PubMed ID (PMID). GATs were applied to analyze the network, leveraging their ability to capture the significance and interdependencies of nodes and edges. The model used a learnable weight matrix to compute attention coefficients, normalize them, and aggregate information from neighboring nodes for edge prediction. The GAT model, integrating graph neural networks with attention mechanisms, achieved an accuracy of 83% in predicting miRNA-disease associations for head and neck neoplasms. This study highlights the potential of graph-based deep learning models, particularly GATs, in accurately predicting miRNA-disease associations. A functional enrichment analysis revealed significant involvement of miRNAs in oral cancer pathways, notably highlighting the critical roles of the TGF-beta and PI3K-Akt signaling pathways in tumor progression and cell survival. These findings offer a pathway to better understanding the molecular mechanisms underlying head and neck cancers. Future improvements in dataset size, model evaluation, and interpretability could further enhance prediction accuracy, potentially advancing diagnostic and therapeutic strategies for these cancers.
AIMS/HYPOTHESIS: Residual pancreatic beta cells in type 1 diabetes show reduced insulin production but the mechanisms remain unclear. Beta cells undergo stress responses during type 1 diabetes, including endoplasmic reticulum (ER) stress and DNA damage-associated senescence, which may affect insulin production. ER stress reduces insulin production but whether senescence disrupts insulin production in human beta cells has not been investigated.
METHODS: DNA damage-mediated senescence was induced using bleomycin in human donor islets. Relative levels of prohormone convertase 1/3 (PC1/3), prohormone convertase 2 (PC2), carboxypeptidase E (CPE) and the endogenous PC1/3 inhibitor, proprotein convertase subtilisin/kexin type 1 inhibitor (proSAAS), were quantified by western blot. Levels of proinsulin and insulin were measured by ELISA. Flow cytometry was used to measure insulin expression in islet cells. FACS was used to sort endogenous senescent beta cells from islets for analysis of insulin content. Proinsulin immunofluorescence staining was quantified in endogenous senescent vs non-senescent beta cells in pancreas tissue from control donors and donors with type 1 diabetes. Publicly available datasets were used to interrogate relationships between senescence effectors, proinsulin-processing genes and insulin content. DNA damage was induced with bleomycin in the non-proliferative female-fetus-derived EndoC-βH5 human beta cell model to study the impact of the DNA damage response on insulin production in clonal cells growth-arrested due to p16INK4A expression.
RESULTS: DNA damage-mediated senescence led to increased PC1/3 without changes in levels of PC2, CPE or proSAAS in human islets. Consistent with these changes, no significant differences in proinsulin or insulin content were observed, compared with control islets. Flow cytometry confirmed maintenance of insulin content in DNA damage-mediated senescent beta cells vs control cells and sorted endogenous senescent beta cells had similar insulin content to non-senescent beta cells. Proinsulin staining was similar in endogenous senescent vs non-senescent beta cells from a control donor and donor with type 1 diabetes. Analysis of proteomics datasets from Humanislets.com and single-cell RNA-seq datasets from the Human Pancreas Analysis Program corroborated these findings. In EndoC-βH5 beta cells, which are growth-arrested, DNA damage led to decreased levels of CPE and proSAAS, and reduced levels of insulin.
CONCLUSIONS/INTERPRETATION: Our findings suggest that the expression of proinsulin-processing enzymes and the production of insulin are sustained in both chemically induced DNA damage-related senescence and in endogenous senescent adult human beta cells. Collectively, these findings suggest that senescent beta cells may be a source of insulin production among residual beta cells in type 1 diabetes.
This corrects the article on p. 39 in vol. 13, PMID: 40347125.
Educational Engagement Modules (EEMs) are teaching materials for educators and students that facilitate a deeper understanding of key epidemiological methods and concepts. Each EEM poses a series of questions using a recently published paper in Annals to further understanding of a specific study design and to encourage critical thinking and careful evaluation. This EEM focuses on the use of propensity score matching and multinomial models in a study exploring the association between experiencing a stroke and psychological distress and references the following article: Dagli C, Patel PG, Gonzalez K, Nair M, Al-Antary N, Lin C, Adjei Boakye E. Psychological distress among stroke survivors in the US: An analysis of the National Health Interview Survey. Ann Epidemiol. 2025 Jun 30;109:8-13. doi: 10.1016/j.annepidem.2025.06.019. Epub ahead of print. PMID: 40602697.
[This corrects the article on p. 1 in vol. 50, PMID: 40124828.].
Although oligodendrocytes (OLs) synthesize laminin-γ1, the most widely used γ subunit, its functional significance in the CNS remains unknown. To answer this important question, we generated a conditional knockout mouse line with laminin-γ1 deficiency in OL lineage cells (γ1-OKO). γ1-OKO mice exhibit weakness/paralysis and die by post-natal day 33. Additionally, they develop blood-brain barrier (BBB) disruption in the cortex and striatum. Subsequent studies reveal decreased major facilitator superfamily domain containing 2a expression and increased endothelial caveolae vesicles, but unaltered tight junction protein expression and tight junction ultrastructure, indicating a transcellular, rather than a paracellular, mechanism of BBB breakdown. Furthermore, significantly reduced OL lineage cells, OL precursor cells (OPCs), proliferating OPCs, and mature OLs are observed in γ1-OKO brains in a region-specific manner. Consistent with this finding, various defects in myelination are detected in γ1-OKO brains at biochemical and ultrastructural levels. Overall, these results highlight important roles of OL-derived laminin-γ1 in BBB maintenance and OL biology (proliferation, differentiation, and myelination).
Mural cells synthesize and deposit laminin to the basement membrane. To investigate the function of mural cell-derived laminin, we generated a mutant mouse line lacking mural cell-derived laminin (termed PKO). In a previous study, we showed that the PKO mice were grossly normal under homeostatic condition, but developed blood-brain barrier (BBB) breakdown with advanced age (> 8 months), suggesting that these mutants are intrinsically weak. Based on these findings, we hypothesized that PKO mice have exacerbated injuries in pathological conditions.
Using collagenase-induced intracerebral hemorrhage (ICH) as an injury model, we examined various stroke outcomes, including hematoma volume, neurological function, neuronal death, BBB integrity, paracellular/transcellular transport, inflammatory cell infiltration, and brain water content, in PKO mice and their wildtype littermates at young age (6–8 weeks). In addition, transmission electron microscopy (TEM) analysis and an in vitro ICH model were used to investigate the underlying molecular mechanisms.
Compared to age-matched wildtype littermates, PKO mice display aggravated stroke outcomes, including larger hematoma size, worse neurological function, increased neuronal cell death, enhanced BBB permeability, increased transcytosis, and elevated inflammatory cell infiltration. These mutants also exhibit high baseline brain water content independent of aquaporin-4 (AQP4). In addition, mural cell-derived laminin significantly reduced caveolin-1 without affecting tight junction proteins in the in vitro ICH model.
These results suggest that mural cell-derived laminin attenuates BBB damage in ICH via decreasing caveolin-1 and thus transcytosis, regulates brain water homeostasis, and plays a beneficial role in ICH.