mRNAsi Index: Machine Learning in Mining Lung Adenocarcinoma Stem Cell Biomarkers

Identification of hub genes and compounds controlling ovarian cancer stem cell characteristics via stemness indices analysis

Background

Ovarian cancer (OC) is the most lethal gynecological malignancy. It has been reported that cancer stem cells (CSCs) play a crucial role in disseminated metastases in abdominal cavity and chemotherapy resistance of high-grade serous OC. However, the overall gene expression features of OC stem cells have not been clarified.

Methods

Expression datasets of 379 OC samples and 88 normal tissues were downloaded from The Cancer Genome Atlas (TCGA) and the Genotype Tissue Expression (GTEx) project. Differentially expressed genes (DEGs) were screened using the “limma” package in R software. Among the DEGs, modules and hub genes that were highly related to messenger RNA expression-based stemness index (mRNAsi) and epigenetically regulated mRNAsi indices were identified via weighted gene co-expression network analysis (WGCNA). These hub genes were considered to be associated with OC stem cells. The Gene Ontology (GO) project and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis was used to identify the main biological processes that hub genes participated in. Finally, Connectivity Map (CMap) was used to predict compounds that disturb the hub genes.

Results

We identified 2,253 DEGs; of these, 31 had a significantly positive correlation to mRNAsi indices and were upregulated in OC, while 41 of them had a significantly negative correlation with mRNAsi indices and were downregulated in OC. Correlation analysis indicated that hub genes from the same module composed a dense interaction network. GO and KEGG enrichment analysis demonstrated that hub genes primarily play roles in cell division and proliferation. Moreover, the compounds that may disturb hub genes were identified. Of these, 11 compounds, including MS-275, DL-thiorphan, and GW-8510, which have never been studied in OC stem cells, were screened as underlying treatments targeting OC stem cells.

Conclusions

Altogether, 72 hub genes that were closely linked to OC stem cell characteristics were found to mainly participate in cell division and proliferation. Moreover, compounds that disturb these hub gens were identified and can be considered underlying targets for inhibiting OC stem cells.

The Integrative Analysis Identifies Three Cancer Subtypes and Stemness Features in Cutaneous Melanoma

Background: With the growing uncovering of drug resistance in melanoma treatment, personalized cancer therapy and cancer stem cells are potential therapeutic targets for this aggressive skin cancer. Methods: Multi-omics data of cutaneous melanoma were obtained from The Cancer Genome Atlas (TCGA) database. Then, these melanoma patients were classified into different subgroups by performing "CancerSubtypes" method. The differences of stemness indices (mRNAsi and mDNAsi) and tumor microenvironment indices (immune score, stromal score, and tumor purity) among subtypes were investigated. Moreover, the Least Absolute Shrinkage and Selection Operator (LASSO) and Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithms were performed to identify a cancer cell stemness feature, and the likelihood of immuno/chemotherapeutic response was further explored. Results: Totally, 3 specific subtypes of melanoma with different survival outcomes were identified from TCGA. We found subtype 2 of melanoma with the higher immune score and stromal score and lower mRNAsi and tumor purity score, which has the best survival time than the other subtypes. By performing Kaplan-Meier survival analysis, we found that mRNAsi was significantly associated with the overall survival time of melanomas in subtype 2. Correlation analysis indicated surprising associations between stemness indices and subsets of tumor-infiltrating immune cells. Besides, we developed and validated a prognostic stemness-related genes feature that can divide melanoma patients into high- and low-risk subgroups by applying risk score system. The high-risk group has a significantly shorter survival time than the low-risk subgroup, which is more sensitive to CTLA-4 immune therapy. Finally, 16 compounds were screened out in the Connectivity Map database which may be potential therapeutic drugs for melanomas. Conclusion: Thus, our finding provides a new framework for classification and finds some potential targets for the treatment of melanoma.

ITM2A as a Tumor Suppressor and Its Correlation With PD-L1 in Breast Cancer

Background

High expression of integral membrane protein 2A (ITM2A) was reported to be associated with favorable prognosis in several solid tumors including breast cancer. This study aimed to investigate the role of ITM2A in breast cancer, especially in respect to tumor microenvironment.

Methods

ITM2A expression was evaluated based on qRT-PCR results on breast cancer specimens, as well as TCGA and GEO datasets. The influence of ITM2A expression on breast cancer cell proliferation and tumor growth were evaluated by CCK-8 assay, clonogenic assay, and murine xenograft models. Transwell assay was performed to observe the changes of invasion and migration capacity in breast cancer cells. To determine the biological functions of ITM2A, differentially expressed genes (DEGs) were screened based on RNA-sequencing data of MCF-7 cells overexpressed ITM2A. Then, functional annotation on DEGs was given by Gene Ontology and KEGG analysis. The stimulation on programmed cell death ligand 1 (PD-L1) expression when ITM2A overexpressed was determined by flow cytometry. Meanwhile, the correlation on expression levels between PD-L1 and ITM2A was tested via qRT-PCR on 24 breast cancer tissues, as well as public database.

Results

We demonstrated that ITM2A was frequently downregulated in breast cancer. Patients with high expression levels of ITM2A had longer overall survival and relapse free survival. Overexpression of ITM2A inhibited proliferation and impaired cells capacity of invasion and migration in vitro and in vivo. The DEGs in breast cancer cells overexpressed ITM2A were found to be associated with immunity responses. Moreover, ITM2A was found to facilitate breast cancer cells to express PD-L1. The correlation between PD-L1 and ITM2A was verified with both qRT-PCR assay and public database. Additionally, it was found that breast cancer had higher ITM2A expression frequently had more tumor-infiltrating lymphocytes (TILs).

Conclusion

In summary, we found that high expression of ITM2A reduced the aggressivity of breast cancer cells and had a favorable effect on outcomes of patients with breast cancer. Moreover, ITM2A induced PD-L1 expression in breast cancer cells was accompanied with higher TILs numbers in tumor microenvironment.

Partial acquisition of stemness properties in tumorspheres obtained from interleukin-8-treated MCF-7 cells

The interleukin-8 is an important regulator of the tumor microenvironment, promoting the epithelial-mesenchymal transition and the acquisition of stem-like cell properties in cancer cells. The tumorsphere-formation assay has been used for the identification of cancer stem cell. Interleukin-8 induces the formation of larger tumorspheres in Michigan Cancer Foundation-7 (MCF-7) cells, suggesting cancer stem cell enrichment. In this work, we aimed to study the phenotypic and functional characteristics of the cells present within the tumorspheres of MCF-7 cells previously treated with interleukin-8. MCF-7 cells treated for 5 days or not with this cytokine were further cultivated in ultralow attachment plates for another 5 days to allow tumorspheres formation. We showed that the enhanced sphere formation by MCF-7 cells was not a consequence of higher cell proliferation by interleukin-8 stimulation. Despite maintaining an epithelial-mesenchymal transition phenotype with the presence of epithelial and mesenchymal markers, basic stemness properties were impaired in tumorspheres and in those treated with interleukin-8, while others were increased. Self-renewal capacity was increased in interleukin-8-treated cells only in the first generation of tumorspheres but was not sustained in consecutive assays. Accordingly, self-renewal and reprogramming gene expression, differentiation capacity to adipocytes, and clonogenicity were also impaired. We showed also that tumorspheres were enriched in differentiated luminal cells (EpCAM+/CD49f-). Nevertheless, cells were more quiescent and maintain a partial epithelial-mesenchymal transition, consistent with their increased resistance to Paclitaxel and Doxorubicin. They also presented higher migration and interleukin-8-directed invasion. Therefore, the breast cancer cell line MCF-7, having a low stemness index, might partially acquire some stem-like cell attributes after interleukin-8 stimulation, increasing its aggressiveness.

Identifying 8-mRNAsi Based Signature for Predicting Survival in Patients With Head and Neck Squamous Cell Carcinoma via Machine Learning

Cancer stem cells (CSCs) have been characterized by several exclusive features that include differentiation, self-renew, and homeostatic control, which allows tumor maintenance and spread. Recurrence and therapeutic resistance of head and neck squamous cell carcinomas (HNSCC) have been identified to be attributed to CSCs. However, the biomarkers led to the development of HNSCC stem cells remain less defined. In this study, we quantified cancer stemness by mRNA expression-based stemness index (mRNAsi), and found that mRNAsi indices were higher in HNSCC tissues than that in normal tissue. A significantly higher mRNAsi was observed in HPV positive patients than HPV negative patients, as well as in male patients than in female patients. The 8-mRNAsi signature was identified from the genes in two modules which were mostly related to mRNAsi screened by weighted gene co-expression network analysis. In this prognostic signatures, high expression of RGS16, LYVE1, hnRNPC, ANP32A, and AIMP1 focus in promoting cell proliferation and tumor progression. While ZNF66, PIK3R3, and MAP2K7 are associated with a low risk of death. The riskscore of eight signatures have a powerful capacity for 1-, 3-, 5-year of overall survival prediction (5-year AUC 0.77, 95% CI 0.69-0.85). These findings based on stemness indices may provide a novel understanding of target therapy for suppressing HNSCC stem cells.

Identification of a Novel Cancer Stemness-Associated ceRNA Axis in Lung Adenocarcinoma via Stemness Indices Analysis

The aim of this study was to identify a novel cancer stemness-related ceRNA regulatory axis in lung adenocarcinoma (LUAD) via weighted gene coexpression network analysis of a stemness index. The RNA sequencing expression profiles of 513 cancer samples and 60 normal samples were obtained from the TCGA database. Differentially expressed mRNAs (DEmRNAs), lncRNAs (DElncRNAs), and miRNAs (DEmiRNAs) were identified with R software. Functional enrichment analysis was conducted using DAVID 6.8. The ceRNA network was constructed via multiple bioinformatics analyses, and the correlations between possible ceRNAs and prognosis were analyzed using Kaplan–Meier plots. WGCNA was then applied to distinguish key genes related to the mRNA expression-based stemness index (mRNAsi) in LUAD. After combining the weighted gene coexpression and ceRNA networks, a novel ceRNA regulatory axis was identified, and its biological functions were explored in vitro and vivo. In total, 1,825 DElncRNAs, 291 DEmiRNAs, and 3,742 DEmRNAs were identified. Functional enrichment analysis revealed that the DEmRNAs might be associated with LUAD onset and progression. The ceRNA network was constructed with 14 lncRNAs, 10 miRNAs, and 52 mRNAs. Kaplan–Meier analysis identified 2 DEmiRNAs, 5 DElncRNAs, and 41 DEmRNAs with remarkable prognostic power. One gene (MFAP4) in the ceRNA network was found to be closely related to mRNAsi by using WGCNA. Functional investigation further confirmed that the C8orf34-as1/miR-671-5p/MFAP4 regulatory axis has important functions in LUAD cell migration and stemness. This study provides a deeper understanding of the lncRNA–miRNA–mRNA ceRNA network and, more importantly, reveals a novel ceRNA regulatory axis, which may provide new insights into novel molecular therapeutic targets for inhibiting LUAD stem characteristics.

Comprehensive analysis of transcriptome data stemness indices identifies key genes for controlling cancer stem cell characteristics in gastric cancer

BACKGROUND

Cancer stem cells (CSCs) are the tumor cell of origin with self-renewing ability and multi-differentiation potency. CSCs can play vital roles in gastric cancer (GC) metastasis and relapse. However, the genes that regulate the stemness maintenance of CSCs in GC patients remain largely unknown. In the present study, we sought to determine the key genes associated with stemness in GC patients.

METHODS

mRNA expression-based stemness index (mRNA SI) was analyzed with regard to the differential expression levels between normal and GC tissues, as well as clinical features and survival outcomes. Weighted gene co-expression network analysis (WGCNA) was performed to identify modules of interest and key genes. The differences in mRNA expression of key genes between normal and GC tissues were calculated by "ggpubr" package in R. Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) analysis were carried out to annotate the function of key genes. Protein-protein interaction (PPI) and gene co-expression analyses were conducted using STRING and "corrplot" package in R, respectively.

RESULTS

mRNA SI score was markedly increased in GC tumor compared to normal tissues. High mRNA SI score was remarkably associated with more advanced tumor stage and higher pathologic grade, but longer survival times. Based on the results of WGCNA, 19 key genes (i.e., BUB1, BUB1B, KIF14, NCAPH, RACGAP1, KIF15, CENPF, TPX2, RAD54L, KIF18B, TTX, KIF4A, SGO2, PLK4, ARHGAP11A, XRCC2, C1orf112, NCAPG, ORC6) were identified. GO and KEGG functional analyses revealed that these 19 key genes were mainly related to cell proliferation. From PPI and gene co-expression analyses, these 19 key genes were discovered to be intensively associated with each other at both protein and transcription levels.

CONCLUSIONS

our study identified 19 key genes that play vital roles in the stemness maintenance of CSCs in GC patients. Targeting these key genes may help to control CSC characteristics in GC.

Stemness Related Genes Revealed by Network Analysis Associated With Tumor Immune Microenvironment and the Clinical Outcome in Lung Adenocarcinoma

Lung adenocarcinoma (LUAD) is one of the leading fatal malignancy with high morbidity and mortality worldwide. However, due to its complicated mechanism and lack of effective clinical therapeutics, early diagnosis and prognosis are still unsatisfactory. Most of the previous studies focused on cancer stem cells (CSCs), the relationship between cancer stemness (stem-like characteristics) and anti-tumor immunity has not been clearly revealed. Therefore, this study aimed to comprehensively analyze the role of cancer stemness and tumor microenvironment (TME) in LUAD using weighted gene co-expression network analysis (WGCNA). We constructed a gene co-expression network, identified key modules, and hub genes, and further explored the relationship between hub gene expression and cancer immunological characteristics through a variety of algorithms, including Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE) and Gene Set Enrichment Analysis (GSEA). The hub genes were renamed stemness related genes (SRGs), whose functions were examined at the transcription and protein levels through survival analysis with additional samples, Oncomine database, immunohistochemistry, single cell RNA sequencing (scRNA-seq) and single-sample Gene Set Enrichment Analysis (ssGSEA). Subsequently, Tumor Immune Dysfunction and Exclusion (TIDE) and Connectivity Map (CMap) were implemented for treatment and prognosis analyses. As a result, 15 co-expressed SRGs (CCNA2, CCNB1, CDC20, CDCA5, CDCA8, FEN1, KIF2C, KPNA2, MCM6, NUSAP1, RACGAP1, RRM2, SPAG5, TOP2A, and TPX2) were identified. The overexpression of which was discovered to be associated with reduced immune infiltration in LUAD. It was discovered that there was a general negative correlation between cancer stemness and immunity. The expression of SRGs could probably affect our tumor occurrence, progression, the efficacy of chemotherapy and immunotherapy, and clinical outcomes. In conclusion, the 15 SRGs reported in our study may be used as potential candidate biomarkers for prognostic indicators and therapeutic targets after further validation.

Prognostic Prediction Using a Stemness Index-Related Signature in a Cohort of Gastric Cancer

Background

With characteristic self-renewal and multipotent differentiation, cancer stem cells (CSCs) have a crucial influence on the metastasis, relapse and drug resistance of gastric cancer (GC). However, the genes that participates in the stemness of GC stem cells have not been identified.

Methods

The mRNA expression-based stemness index (mRNAsi) was analyzed with differential expressions in GC. The weighted gene co-expression network analysis (WGCNA) was utilized to build a co-expression network targeting differentially expressed genes (DEG) and discover mRNAsi-related modules and genes. We assessed the association between the key genes at both the transcription and protein level. Gene Expression Omnibus (GEO) database was used to validate the expression levels of the key genes. The risk model was established according to the least absolute shrinkage and selection operator (LASSO) Cox regression analysis. Furthermore, we determined the prognostic value of the model by employing Kaplan-Meier (KM) plus multivariate Cox analysis.

Results

GC tissues exhibited a substantially higher mRNAsi relative to the healthy non-tumor tissues. Based on WGCNA, 17 key genes (ARHGAP11A, BUB1, BUB1B, C1orf112, CENPF, KIF14, KIF15, KIF18B, KIF4A, NCAPH, PLK4, RACGAP1, RAD54L, SGO2, TPX2, TTK, and XRCC2) were identified. These key genes were clearly overexpressed in GC and validated in the GEO database. The protein-protein interaction (PPI) network as assessed by STRING indicated that the key genes were tightly connected. After LASSO analysis, a nine-gene risk model (BUB1B, NCAPH, KIF15, RAD54L, KIF18B, KIF4A, TTK, SGO2, C1orf112) was constructed. The overall survival in the high-risk group was relatively poor. The area under curve (AUC) of risk score was higher compared to that of clinicopathological characteristics. According to the multivariate Cox analysis, the nine-gene risk model was a predictor of disease outcomes in GC patients (HR, 7.606; 95% CI, 3.037-19.051; P < 0.001). We constructed a prognostic nomogram with well-fitted calibration curve based on risk score and clinical data.

Conclusion

The 17 mRNAsi-related key genes identified in this study could be potential treatment targets in GC treatment, considering that they can inhibit the stemness properties. The nine-gene risk model can be employed to predict the disease outcomes of the patients.

Establishment and Analysis of a Combined Diagnostic Model of Polycystic Ovary Syndrome with Random Forest and Artificial Neural Network

Polycystic ovary syndrome (PCOS) is one of the most common metabolic and reproductive endocrinopathies. However, few studies have tried to develop a diagnostic model based on gene biomarkers. In this study, we applied a computational method by combining two machine learning algorithms, including random forest (RF) and artificial neural network (ANN), to identify gene biomarkers and construct diagnostic model. We collected gene expression data from Gene Expression Omnibus (GEO) database containing 76 PCOS samples and 57 normal samples; five datasets were utilized, including one dataset for screening differentially expressed genes (DEGs), two training datasets, and two validation datasets. Firstly, based on RF, 12 key genes in 264 DEGs were identified to be vital for classification of PCOS and normal samples. Moreover, the weights of these key genes were calculated using ANN with microarray and RNA-seq training dataset, respectively. Furthermore, the diagnostic models for two types of datasets were developed and named neuralPCOS. Finally, two validation datasets were used to test and compare the performance of neuralPCOS with other two set of marker genes by area under curve (AUC). Our model achieved an AUC of 0.7273 in microarray dataset, and 0.6488 in RNA-seq dataset. To conclude, we uncovered gene biomarkers and developed a novel diagnostic model of PCOS, which would be helpful for diagnosis.