Development and Validation of a Novel PPAR Signaling Pathway-Related Predictive Model to Predict Prognosis in Breast Cancer

Development of a promising PPAR signaling pathway-related prognostic prediction model for hepatocellular carcinoma

The peroxisome proliferator-activated receptor (PPAR) signaling pathway plays a crucial role in systemic cell metabolism, energy homeostasis and immune response inhibition. However, its significance in hepatocellular carcinoma (HCC) has not been well documented. In our study, based on the RNA sequencing data of HCC, consensus clustering analyses were performed to identify PPAR signaling pathway-related molecular subtypes, each of which displaying varying survival probabilities and immune infiltration status. Following, a prognostic prediction model of HCC was developed by using the random survival forest method and Cox regression analysis. Significant difference in survival outcome, immune landscape, drug sensitivity and pathological features were observed between patients with different prognosis. Additionally, decision tree and nomogram models were adopted to optimize the prognostic prediction model. Furthermore, the robustness of the model was verified through single-cell RNA-sequencing data. Collectively, this study systematically elucidated that the PPAR signaling pathway-related prognostic model has good predictive efficacy for patients with HCC. These findings provide valuable insights for further research on personalized treatment approaches for HCC.

TREM1 is involved in the mechanism between asthma and lung cancer by regulating the Toll‑like receptor signaling pathway

Lung cancer and asthma are both global health problems with significant economic consequences. Recent studies have demonstrated that asthma may be a risk factor for lung cancer. The present study aimed to explore the pathogenesis between these two diseases through a comprehensive analysis. Differentially expressed genes (DEGs) screened in the asthma-related GSE165934 dataset were analyzed to find relevant inflammatory pathways. Overlapping genes regulated by inflammatory pathways and lung cancer-DEGs from The Cancer Genome Atlas (TCGA) were obtained and subjected to survival and gene-wide mutation analyses, and nomogram construction to determine the hub gene. The hub gene was further analyzed through expression validation, immunoassays and functional experiments to investigate its role and mechanism in lung cancer. Functional enrichment analysis showed that 1,275 DEGs from GSE165934 were closely associated with the Toll-like receptor signaling pathway, and 8 overlapping genes were identified from 12 genes regulated by the Toll-like receptor signaling pathway and 3,134 TCGA-DEGs. After a series of bioinformatics analyses, it was found that triggering receptor expressed on myeloid cells 1 (TREM1) was the hub gene involved in the mechanism of asthma and lung cancer. TREM1 was also found to be a suppressor gene in lung cancer correlated with immune cells, immune checkpoint-related genes and tumor mutational burden score. Additionally, the results of Cell Counting Kit-8 and Transwell experiments demonstrated that overexpression of TREM1 could significantly inhibit the invasion, proliferation and migration of lung cancer cells. Reverse transcription-quantitative PCR and western blotting demonstrated that the overexpression of TREM1 could also significantly reduce the level of Toll-like receptor signaling pathway proteins. The present findings suggest that TREM1 is associated with the mechanism of asthma and lung cancer through its regulation of the Toll-like receptor signaling pathway. Furthermore, TREM1 may serve as a potential treatment target and prognostic indicator for patients with lung cancer.

Mdwgan-gp: data augmentation for gene expression data based on multiple discriminator WGAN-GP

BACKGROUND

Although gene expression data play significant roles in biological and medical studies, their applications are hampered due to the difficulty and high expenses of gathering them through biological experiments. It is an urgent problem to generate high quality gene expression data with computational methods. WGAN-GP, a generative adversarial network-based method, has been successfully applied in augmenting gene expression data. However, mode collapse or over-fitting may take place for small training samples due to just one discriminator is adopted in the method.

RESULTS

In this study, an improved data augmentation approach MDWGAN-GP, a generative adversarial network model with multiple discriminators, is proposed. In addition, a novel method is devised for enriching training samples based on linear graph convolutional network. Extensive experiments were implemented on real biological data.

CONCLUSIONS

The experimental results have demonstrated that compared with other state-of-the-art methods, the MDWGAN-GP method can produce higher quality generated gene expression data in most cases.

An untargeted metabolomics investigation in liver of flaviviruses-infected mice

Dengue virus (DENV), Japanese encephalitis virus (JEV) and Zika virus (ZIKV) are the three most important flaviviruses, which can cause health problems worldwide. All these flaviviruses can cause liver damage, however, the mechanism of liver injury is still unclear. Metabolomics can give insight into the full complexity of a disease. In our study, we used an LC-MS method to analysis the metabolites in liver samples of the three flaviviruses-infected mice and the non-infected mice. Compared with the control mice, the liver of the DENV-infected, JEV-infected, and ZIKV-infected mice had 32, 34, and 55 differential metabolites. We also found that there were obvious differences in some metabolic pathways among the four groups. Metabonomic analysis of liver is very important for understanding the pathogenesis of flaviviruses.

Prognostic and immune microenvironment analysis of cuproptosis-related LncRNAs in breast cancer

Breast cancer is the most common tumor and the leading cause of cancer death in women. Cuproptosis is a new type of cell death, which can induce proteotoxic stress and eventually lead to cell death. Therefore, regulating copper metabolism in tumor cells is a new therapeutic approach. Long non-coding RNAs play an important regulatory role in immune response. At present, cuproptosis-related lncRNAs in breast cancer have not been reported. Breast cancer RNA sequencing, genomic mutations, and clinical data were downloaded from The Cancer Genome Atlas (TCGA). Patients with breast cancer were randomly assigned to the train group or the test group. Co-expression network analysis, Cox regression method, and least absolute shrinkage and selection operator (LASSO) method were used to identify cuproptosis-related lncRNAs and to construct a risk prognostic model. The prediction performance of the model is verified and recognized. In addition, the nomogram was used to predict the prognosis of breast cancer patients. Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and immunoassay were used to detect the differences in biological function. Tumor mutation burden (TMB) was used to measure immunotherapy response. A total of 19 cuproptosis genes were obtained and a prognostic model based on 10 cuproptosis-related lncRNAs was constructed. Kaplan-Meier survival curves showed statistically significant overall survival (OS) between the high-risk and low-risk groups. Receiver operating characteristic curve (ROC) and principal component analysis (PCA) show that the model has accurate prediction ability. Compared with other clinical features, cuproptosis-related lncRNAs model has higher diagnostic efficiency. Univariate and multivariate Cox regression analysis showed that risk score was an independent prognostic factor for breast cancer patients. In addition, the nomogram model analysis showed that the tumor mutation burden was significantly different between the high-risk and low-risk groups. Of note, the additive effect of patients in the high-risk group and patients with high TMB resulted in reduced survival in breast cancer patients. Our study identified 10 cuproptosis-related lncRNAs, which may be promising biomarkers for predicting the survival prognosis of breast cancer.

The potential effect of BMSCs with miR-27a in improving steroid-induced osteonecrosis of the femoral head

Steroid induced osteonecrosis of the femoral head (ONFH) frequently leads to femoral head collapse and subsequent hip arthritis. This study aimed to investigate the potential therapeutic mechanism of miR-27a on steroid-induced ONFH. Levels of IL-6, TNF-α, miR-27a, Runx2, PPAR-γ and ApoA5 were first examined in bone marrow tissues from steroid-induced ONFH and controls. Subsequently, we overexpressed or knocked down miR-27a in bone marrow mesenchymal stem cells (BMSCs) and detected cell proliferation, osteogenic differentiation, adipogenic differentiation. In addition, miR-27a mimics and BMSCs were injected into the established steroid-induced ONFH rats, and the osteoprotective effects of both were evaluated. Dual luciferase reporter was used to test the targeting effect of miR-27a-3p and PPARG. miR-27a and Runx2 were lowly expressed in steroid-induced ONFH, PPAR-γ and ApoA5 were highly expressed. Overexpression of miR-27a in BMSCs promoted cell proliferation and osteogenic differentiation, inhibited adipogenic differentiation. Furthermore, increasing miR-27a and BMSCs obviously reduced bone loss in steroid induced ONFH rats. The expressions of Runx2 in BMSCs and steroid-induced ONFH rats was significantly up-regulated, while IL-6, TNF-α, PPAR-γ and ApoA5 were down-regulated with miR-27a overexpression. Additionally, PPARG was the target of miR-27a-3p. The results of the present study reveal a role for miR-27a in promoting osteogenesis and may have a synergistic effect with BMSCs.

Cell Division Cycle-Associated Protein 3 (CDCA3) Is a Potential Biomarker for Clinical Prognosis and Immunotherapy in Pan-Cancer

CDCA3 is an essential regulator in cell mitosis and can regulate many physiological and pathological processes in the human body by stimulating certain proteins such as cell cycle regulatory proteins, transcription factors, and signal transduction molecules. Although several studies have shown that dysregulation of CDCA3 is a common phenomenon in human cancers, no systematic pan-cancer analysis has been performed. In this study, we comprehensively investigated the role of CDCA3 in 33 human cancer types by utilizing multiple cancer-related databases and bioinformatics analysis tools, including TCGA, GTEx, GEPIA, TIMER, STRING, Metascape, and Cytoscape. Evidence from bioinformatics databases shows that CDCA3 is overexpressed in almost all human cancer types, and its overexpression is significantly associated with survival in patients with more than ten cancer types. CDCA3 expression positively correlates with immune cell infiltration levels in multiple human cancer types. Furthermore, the results of the GSEA analysis revealed that overexpression of CDCA3 may promote the malignant progression of cancer by activating various oncogenic signaling pathways in human cancers. In conclusion, our pan-cancer analysis provides a comprehensive overview of the oncogenic role of CDCA3 in multiple human cancer types, suggesting that CDCA3 may serve as a potential therapeutic target and prognostic biomarker in multiple human cancer types.

Prognostic Signature and Therapeutic Value Based on Membrane Lipid Biosynthesis-Related Genes in Breast Cancer

There is a need to improve diagnostic and therapeutic approaches to enhance the prognosis of breast cancer, the most common malignancy worldwide. Membrane lipid biosynthesis is a hot biological pathway in current cancer research. It is unclear whether membrane lipid biosynthesis is involved in the prognosis of BRCA. With LASSO regression, a 14-gene prediction model was constructed using data from the TCGA-BRCA cohort. The prediction model includes GPAA1, PIGF, ST3GAL1, ST6GALNAC4, PLPP2, ELOVL1, HACD1, SGPP1, PRKD2, VAPB, CERS2, SGMS2, ALDH3B2, and HACD3. BRCA patients from the TCGA-BRCA cohort were divided into two risk subgroups based on the model. Kaplan–Meier survival curves showed that patients with lower risk scores had significantly improved overall survival (P=2.49e − 09). In addition, risk score, age, stage, and TNM classification were used to predict mortality in BRCA patients. In addition, the 14 genes in the risk model were analyzed for gene variation, methylation level, drug sensitivity, and immune cell infiltration, and the miRNA-mRNA network was constructed. Afterward, the THPA website then analyzed the protein expression of 14 of these risk model genes in normal and pathological BRCA tissues. In conclusion, the membrane lipid biosynthesis-related risk model and nomogram can be used to predict BRCA clinical prognosis.