Ferrous ascorbate as a potential biomarker for diabetic retinopathy: a vitreous humour metabolomics study.

BACKGROUND: This study aimed to explore differences in vitreous humour metabolites and metabolic pathways between patients with and without diabetic retinopathy (DR) and identify potential metabolite biomarkers. METHODS: Clinical data and vitreous fluid samples were collected from 125 patients (40 without diabetes, 85 with DR). The metabolite profiles of the vitreous fluid samples were analysed using ultra-high performance liquid chromatography, Q-Exactive mass spectrometry, and multivariate statistical analysis. A machine learning model based on Least Absolute Shrinkage and Selection Operator Regularized logistic regression was used to build a risk scoring model based on selected metabolite levels. Candidate metabolites were regressed to glycated haemoglobin levels by a logistic regression model. RESULTS: Twenty differential metabolites were identified between the DR and control groups and were significantly enriched in five Kyoto Encyclopedia of Genes and Genomes pathways (arginine biosynthesis; tricarboxylic acid cycle; alanine, aspartate, and glutamate metabolism; tyrosine metabolism; and D-glutamate metabolism). Ferrous ascorbate significantly contributes to poorer glycaemic control outcomes, offering insights into potential new pathogenic pathways in DR. CONCLUSIONS: Disorders in the metabolic pathways of arginine biosynthesis, tricarboxylic acid cycle, alanine, aspartate, glutamate metabolism, tyrosine metabolism, and D-glutamate metabolism were associated with DR. Risk scores based on vitreous fluid metabolites can be used for the diagnosis and management of DR. Ferrous ascorbate can provide insights into potential new pathogenic pathways for DR.

MediDRNet: Tackling category imbalance in diabetic retinopathy classification with dual-branch learning and prototypical contrastive learning.

BACKGROUND AND OBJECTIVE: The classification of diabetic retinopathy (DR) aims to utilize the implicit information in images for early diagnosis, to prevent and mitigate the further worsening of the condition. However, existing methods are often limited by the need to operate within large, annotated datasets to show significant advantages. Additionally, the number of samples for different categories within the dataset needs to be evenly distributed, because the characteristic of sample imbalance distribution can lead to an excessive focus on high-frequency disease categories, while neglecting the less common but equally important disease categories. Therefore, there is an urgent need to develop a new classification method that can effectively alleviate the issue of sample distribution imbalance, thereby enhancing the accuracy of diabetic retinopathy classification. METHODS: In this work, we propose MediDRNet, a dual-branch network model based on prototypical contrastive learning. This model adopts prototype contrastive learning, creating prototypes for different levels of lesions, ensuring they represent the core features of each lesion level. It classifies by comparing the similarity between data points and their category prototypes. Our dual-branch network structure effectively resolves the issue of category imbalance and improves classification accuracy by emphasizing subtle differences in retinal lesions. Moreover, our approach combines a dual-branch network with specific lesion-level prototypes for core feature representation and incorporates the convolutional block attention module for enhanced lesion feature identification. RESULTS: Our experiments using both the Kaggle and UWF classification datasets have demonstrated that MediDRNet exhibits exceptional performance compared to other advanced models in the industry, especially on the UWF DR classification dataset where it achieved state-of-the-art performance across all metrics. On the Kaggle DR classification dataset, it achieved the highest average classification accuracy (0.6327) and Macro-F1 score (0.6361). Particularly in the classification tasks for minority categories of diabetic retinopathy on the Kaggle dataset (Grades 1, 2, 3, and 4), the model reached high classification accuracies of 58.08%, 55.32%, 69.73%, and 90.21%, respectively. In the ablation study, the MediDRNet model proved to be more effective in feature extraction from diabetic retinal fundus images compared to other feature extraction methods. CONCLUSIONS: This study employed prototype contrastive learning and bidirectional branch learning strategies, successfully constructing a grading system for diabetic retinopathy lesions within imbalanced diabetic retinopathy datasets. Through a dual-branch network, the feature learning branch effectively facilitated a smooth transition of features from the grading network to the classification learning branch, accurately identifying minority sample categories. This method not only effectively resolved the issue of sample imbalance but also provided strong support for the precise grading and early diagnosis of diabetic retinopathy in clinical applications, showcasing exceptional performance in handling complex diabetic retinopathy datasets. Moreover, this research significantly improved the efficiency of prevention and management of disease progression in diabetic retinopathy patients within medical practice. We encourage the use and modification of our code, which is publicly accessible on GitHub: https://github.com/ReinforceLove/MediDRNet.

Applications of nanomaterials in anti-VEGF treatment for ophthalmic diseases.

Angiogenesis has been determined to be essential in the occurrence and metastasis of diabetic retinopathy (DR), age-related macular degeneration (AMD), retinal vein occlusion (RVO), choroidal neovascularization (CNV), retinopathy of prematurity (ROP), tumor, etc. However, the clinical use of anti-vascular endothelial growth factors (VEGF) drugs is currently limited due to its high cost, potential side effects, and need for repeated injections. In recent years, nanotechnology has shown promising results in inhibiting neovascularization and reducing reactive oxygen species (ROS) or inflammatory factors. Some nanomaterials can also act as vehicles for drug delivery, such as lipid nanoparticles and PLGA. The process of angiogenesis and its molecular mechanism are discussed in this article. At the same time, this study aims to systematically review the research progress of nanotechnology and offer more treatment options for neovascularization-related diseases in clinical ophthalmology.

Identification of 5 microRNA biomarkers associated with the prognosis of uveal melanoma.

To uncover the role of microRNAs in the occurrence and development of uveal melanoma (UM), we used R language packages in this study to analyze the correlations between the expression of microRNA isoforms, their target genes, and the clinical data for UM patients retrieved from The Cancer Genome Atlas (TCGA). We used Weighted Correlation Network Analysis (WGCNA) to divide the expression profiles of different microRNAs into 10 modules, among which blue and yellow modules were associated with UM survival. Hsa-miR-513a-5p, miR-506-3p, miR-508-3p, miR-140-3p, and miR-103a-2-5p were further identified as the top 5 node microRNAs based on the risk scores in both modules using least absolute shrinkage and selection operator (LASSO) Cox regression analysis. After combining these 5 microRNAs into an integrated risk signature, the prognostic performance of the risk signature was evaluated by area under the receiver operating characteristic (AUROC) curve, and their association with UM clinical characteristics was further analyzed using multiple Cox regression. Our results showed that this risk signature was sensitivity and specificity, and could serve as an independent prognostic factor. In addition, Spearman correlation analysis showed that expression of almost all target mRNAs were significantly positively or negatively correlated with the associated microRNAs. The gene ontology (GO), pathways, and disease enrichment analyses also showed that these 5 microRNAs were closely related to the incidence and progression of tumor, indicating their potential for predicting the outcome of UM.

Transcriptome Analysis of the Anti-Proliferative Effects of Ginsenoside Rh3 on HCT116 Colorectal Cancer Cells.

The mechanism of ginsenoside Rh3 activity against cancer remains unclear. This study aimed to investigate the underlying mechanism. The effects of Rh3 on the cell proliferation, migration and invasion, and cycle and apoptosis were analyzed using CCK-8 assay, transwell migration assay and flow cytometry, respectively. The RNA transcriptome was sequenced and data were analyzed by R software. Protein expression and protein-protein interactions were determined by Western blotting and co-immunoprecipitation, respectively. The results showed Rh3 inhibited HCT116 cell proliferation, invasion, and migration, arrested cells at G1 phase; and increased apoptosis. Rh3 downregulated 314 genes and upregulated 371 genes. Gene Set Enrichment Analysis (GSEA) using The Kyoto Encyclopedia of Genes Genomics ranked DNA replication first, while GSEA using Gene Ontology ranked the initiation of DNA replication first. Compared with tumor data from The Cancer Genome Atlas (TCGA), most of genes related to DNA replication were oppositely regulated by Rh3. Furthermore, Rh3 down-regulated key protein expression related to DNA replication (Orc6, Cdt1, and Mcm2), but did not affect the loading of Mcm complexes onto ORC complexes nor the phosphorylation at ser139 of Mcm2. Therefore, Rh3 may inhibit colorectal cancer HCT116 cells by downregulation of genes related to DNA replication.

FGF-induced Pea3 transcription factors program the genetic landscape for cell fate determination.

FGF signaling is a potent inducer of lacrimal gland development in the eye, capable of transforming the corneal epithelium into glandular tissues. Here, we show that genetic ablation of the Pea3 family of transcription factors not only disrupted the ductal elongation and branching of the lacrimal gland, but also biased the lacrimal gland epithelium toward an epidermal cell fate. Analysis of high-throughput gene expression and chromatin immunoprecipitation data revealed that the Pea3 genes directly control both the positive and negative feedback loops of FGF signaling. Importantly, Pea3 genes are also required to suppress aberrant Notch signaling which, if gone unchecked, can compromise lacrimal gland development by preventing the expression of both Sox and Six family genes. These results demonstrate that Pea3 genes are key FGF early response transcriptional factors, programing the genetic landscape for cell fate determination.

Effects of R type and S type ginsenoside Rg3 on DNA methylation in human hepatocarcinoma cells.

Ginsenoside Rg3, a bioactive constituent isolated from Panax ginseng, exhibits antitumorigenic, antioxidative, antiangiogenic, neuroprotective and other biological activities are associated with the regulation of multiple genes. DNA methylation patterns, particularly those in the promoter region, affect gene expression, and DNA methylation is catalyzed by DNA methylases. However, whether ginsenoside Rg3 affects DNA methylation is unknown. High performance liquid chromatography assay, MspI/HpaII polymerase chain reaction (PCR) and reverse transcription­quantitative PCR were performed to assess DNA methylation. It was demonstrated that 20(S)­ginsenoside Rg3 treatment resulted in increased inhibition of cell growth, compared with treatment with 20(R)­ginsenoside Rg3 in the human HepG2 hepatocarcinoma cell line. It was additionally revealed that treatment with 20(S)­ginsenoside Rg3 reduced global genomic DNA methylation, altered cystosine methylation of the promoter regions of P53, B cell lymphoma 2 and vascular endothelial growth factor, and downregulated the expression of DNA methyltransferase (DNMT) 3a and DNMT3b more than treatment with 20(R)­ginsenoside Rg3 in HepG2 cells. These results revealed that the modulation of DNA methylation may be important in the pharmaceutical activities of ginsenoside Rg3.