In-silico drug screening and potential target identification for hepatocellular carcinoma using Support Vector Machines based on drug screening result

Hepatocellular carcinoma (HCC) is a severe liver malignancy with few drug treatment options. In finding an effective treatment for HCC, screening drugs that are already FDA-approved will fast track the clinical trial and drug approval process. Connectivity Map (CMap), a large repository of chemical-induced gene expression profiles, provides the opportunity to analyze drug properties on the basis of gene expression. Support Vector Machines (SVM) were utilized to classify the effectiveness of drugs against HCC using gene expression profiles in CMap. The results of this classification will help us (1) identify genes that are chemically sensitive, and (2) predict the effectiveness of remaining chemicals in CMap in the treatment of HCC and provide a prioritized list of possible HCC drugs for biological verification.     

Functional Module Connectivity Map (FMCM): A Framework for Searching Repurposed Drug Compounds for Systems Treatment of Cancer and an Application to Colorectal Adenocarcinoma

Drug repurposing has become an increasingly attractive approach to drug development owing to the ever-growing cost of new drug discovery and frequent withdrawal of successful drugs caused by side effect issues. Here, we devised Functional Module Connectivity Map (FMCM) for the discovery of repurposed drug compounds for systems treatment of complex diseases, and applied it to colorectal adenocarcinoma. FMCM used multiple functional gene modules to query the Connectivity Map (CMap). The functional modules were built around hub genes identified, through a gene selection by trend-of-disease-progression (GSToP) procedure, from condition-specific gene-gene interaction networks constructed from sets of cohort gene expression microarrays. The candidate drug compounds were restricted to drugs exhibiting predicted minimal intracellular harmful side effects. We tested FMCM against the common practice of selecting drugs using a genomic signature represented by a single set of individual genes to query CMap (IGCM), and found FMCM to have higher robustness, accuracy, specificity, and reproducibility in identifying known anti-cancer agents. Among the 46 drug candidates selected by FMCM for colorectal adenocarcinoma treatment, 65% had literature support for association with anti-cancer activities, and 60% of the drugs predicted to have harmful effects on cancer had been reported to be associated with carcinogens/immune suppressors. Compounds were formed from the selected drug candidates where in each compound the component drugs collectively were beneficial to all the functional modules while no single component drug was harmful to any of the modules. In cell viability tests, we identified four candidate drugs: GW-8510, etacrynic acid, ginkgolide A, and 6-azathymine, as having high inhibitory activities against cancer cells. Through microarray experiments we confirmed the novel functional links predicted for three candidate drugs: phenoxybenzamine (broad effects), GW-8510 (cell cycle), and imipenem (immune system). We believe FMCM can be usefully applied to repurposed drug discovery for systems treatment of other types of cancer and other complex diseases.     

Predicting hepatitis B virus-positive metastatic hepatocellular carcinomas using gene expression profiling and supervised machine learning

Hepatocellular carcinoma (HCC) is one of the most common and aggressive human malignancies. Its high mortality rate is mainly a result of intra-hepatic metastases. We analyzed the expression profiles of HCC samples without or with intra-hepatic metastases. Using a supervised machine-learning algorithm, we generated for the first time a molecular signature that can classify metastatic HCC patients and identified genes that were relevant to metastasis and patient survival. We found that the gene expression signature of primary HCCs with accompanying metastasis was very similar to that of their corresponding metastases, implying that genes favoring metastasis progression were initiated in the primary tumors. Osteopontin, which was identified as a lead gene in the signature, was over-expressed in metastatic HCC; an osteopontin-specific antibody effectively blocked HCC cell invasion in vitro and inhibited pulmonary metastasis of HCC cells in nude mice. Thus, osteopontin acts as both a diagnostic marker and a potential therapeutic target for metastatic HCC.     

Survival prediction and response to immune checkpoint inhibitors: A prognostic immune signature for hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the most common cancers all over the world. Several studies have explored if immune-related genes and tumor immune microenvironment could play roles in HCC prognoses. This study is aimed at developing a prognostic signature of HCC based on immune-related genes or tumor immune microenvironment to predict survival and response to immune checkpoint inhibitors (ICIs). We constructed a prognostic signature using bioinformatics method and validated its predictive capability. The mechanisms of the signature prediction were explored with The Cancer Immunome Atlas (TCIA) and mutation analysis. We also explored the association between the signature and immunophenoscore (IPS), which is the marker of ICIs response. A 6 immune-related-gene (6-IRG) signature was developed. It was revealed in a multivariate analysis that the 6-IRG signature was an independent prognostic factor of overall survival and progression-free interval among HCC patients. In the high-risk group of 6-IRG signature score, macrophage M0 cells and regulatory T cells, which are observed associated with poor overall survival in our study, were higher. The low-risk group had a higher IPS, which meant a better response to ICIs. Taken together, we constructed a reliable 6-IRG signature for prediction of survival and response to ICIs. The signature needs further testing for clinical application.     

Prognostic genes of hepatocellular carcinoma based on gene coexpression network analysis

Hepatocellular carcinoma (HCC) is the most common subtype in liver cancer whose prognosis is affected by malignant progression associated with complex gene interactions. However, there is currently no available biomarkers associated with HCC progression in clinical application. In our study, RNA sequencing expression data of 50 normal samples and 374 tumor samples was analyzed and 9225 differentially expressed genes were screened. Weighted gene coexpression network analysis was then conducted and the blue module we were interested was identified by calculating the correlations between 17 gene modules and clinical features. In the blue module, the calculation of topological overlap was applied to select the top 30 genes and these 30 genes were divided into the green group (11 genes) and the yellow group (19 genes) through searching whether these genes were validated by in vitro or in vivo experiments. The genes in the green group which had never been validated by any experiments were recognized as hub genes. These hub genes were subsequently validated by a new data set GSE76427 and KM Plotter Online Tool, and the results indicated that 10 genes (FBXO43, ARHGEF39, MXD3, VIPR1, DNASE1L3, PHLDA1, CSRNP1, ADR2B, C1RL, and CDC37L1) could act as prognosis and progression biomarkers of HCC. In summary, 10 genes who have never been mentioned in HCC were identified to be associated with malignant progression and prognosis of patients. These findings may contribute to the improvement of the therapeutic decision, risk stratification, and prognosis prediction for HCC patients.     

A simple and robust method for connecting small-molecule drugs using gene-expression signatures

Background  

Interaction of a drug or chemical with a biological system can result in a gene-expression profile or signature characteristic of the event. Using a suitably robust algorithm these signatures can potentially be used to connect molecules with similar pharmacological or toxicological properties by gene expression profile. Lamb et al first proposed the Connectivity Map [Lamb et al (2006), Science 313, 1929–1935] to make successful connections among small molecules, genes, and diseases using genomic signatures.     

Gene expression signature analysis identifies vorinostat as a candidate therapy for gastric cancer

Background

Gastric cancer continues to be one of the deadliest cancers in the world and therefore identification of new drugs targeting this type of cancer is thus of significant importance. The purpose of this study was to identify and validate a therapeutic agent which might improve the outcomes for gastric cancer patients in the future.

Methodology/Principal Findings

Using microarray technology, we generated a gene expression profile of human gastric cancer–specific genes from human gastric cancer tissue samples. We used this profile in the Broad Institute''s Connectivity Map analysis to identify candidate therapeutic compounds for gastric cancer. We found the histone deacetylase inhibitor vorinostat as the lead compound and thus a potential therapeutic drug for gastric cancer. Vorinostat induced both apoptosis and autophagy in gastric cancer cell lines. Pharmacological and genetic inhibition of autophagy however, increased the therapeutic efficacy of vorinostat, indicating that a combination of vorinostat with autophagy inhibitors may therapeutically be more beneficial. Moreover, gene expression analysis of gastric cancer identified a collection of genes (ITGB5, TYMS, MYB, APOC1, CBX5, PLA2G2A, and KIF20A) whose expression was elevated in gastric tumor tissue and downregulated more than 2-fold by vorinostat treatment in gastric cancer cell lines. In contrast, SCGB2A1, TCN1, CFD, APLP1, and NQO1 manifested a reversed pattern.

Conclusions/Significance

We showed that analysis of gene expression signature may represent an emerging approach to discover therapeutic agents for gastric cancer, such as vorinostat. The observation of altered gene expression after vorinostat treatment may provide the clue to identify the molecular mechanism of vorinostat and those patients likely to benefit from vorinostat treatment.     

Identifying Prognostic Features by Bottom-Up Approach and Correlating to Drug Repositioning

Background

Traditionally top-down method was used to identify prognostic features in cancer research. That is to say, differentially expressed genes usually in cancer versus normal were identified to see if they possess survival prediction power. The problem is that prognostic features identified from one set of patient samples can rarely be transferred to other datasets. We apply bottom-up approach in this study: survival correlated or clinical stage correlated genes were selected first and prioritized by their network topology additionally, then a small set of features can be used as a prognostic signature.

Methods

Gene expression profiles of a cohort of 221 hepatocellular carcinoma (HCC) patients were used as a training set, ‘bottom-up’ approach was applied to discover gene-expression signatures associated with survival in both tumor and adjacent non-tumor tissues, and compared with ‘top-down’ approach. The results were validated in a second cohort of 82 patients which was used as a testing set.

Results

Two sets of gene signatures separately identified in tumor and adjacent non-tumor tissues by bottom-up approach were developed in the training cohort. These two signatures were associated with overall survival times of HCC patients and the robustness of each was validated in the testing set, and each predictive performance was better than gene expression signatures reported previously. Moreover, genes in these two prognosis signature gave some indications for drug-repositioning on HCC. Some approved drugs targeting these markers have the alternative indications on hepatocellular carcinoma.

Conclusion

Using the bottom-up approach, we have developed two prognostic gene signatures with a limited number of genes that associated with overall survival times of patients with HCC. Furthermore, prognostic markers in these two signatures have the potential to be therapeutic targets.     

Drug-Induced Regulation of Target Expression

Drug perturbations of human cells lead to complex responses upon target binding. One of the known mechanisms is a (positive or negative) feedback loop that adjusts the expression level of the respective target protein. To quantify this mechanism systems-wide in an unbiased way, drug-induced differential expression of drug target mRNA was examined in three cell lines using the Connectivity Map. To overcome various biases in this valuable resource, we have developed a computational normalization and scoring procedure that is applicable to gene expression recording upon heterogeneous drug treatments. In 1290 drug-target relations, corresponding to 466 drugs acting on 167 drug targets studied, 8% of the targets are subject to regulation at the mRNA level. We confirmed systematically that in particular G-protein coupled receptors, when serving as known targets, are regulated upon drug treatment. We further newly identified drug-induced differential regulation of Lanosterol 14-alpha demethylase, Endoplasmin, DNA topoisomerase 2-alpha and Calmodulin 1. The feedback regulation in these and other targets is likely to be relevant for the success or failure of the molecular intervention.     

Identification of thioridazine,an antipsychotic drug,as an antiglioblastoma and anticancer stem cell agent using public gene expression data

Glioblastoma (GBM) is a common and malignant tumor with a poor prognosis. Glioblastoma stem cells (GSCs) have been reported to be involved in tumorigenesis, tumor maintenance and therapeutic resistance. Thus, to discover novel candidate therapeutic drugs for anti-GBM and anti-GSCs is an urgent need. We hypothesized that if treatment with a drug could reverse, at least in part, the gene expression signature of GBM and GSCs, this drug may have the potential to inhibit pathways essential in the formation of GBM and thereby treat GBM. Here, we collected 356 GBM gene signatures from public databases and queried the Connectivity Map. We systematically evaluated the in vitro antitumor effects of 79 drugs in GBM cell lines. Of the drugs screened, thioridazine was selected for further characterization because it has potent anti-GBM and anti-GSCs properties. When investigating the mechanisms underlying the cytocidal effects of thioridazine, we found that thioridazine induces autophagy in GBM cell lines, and upregulates AMPK activity. Moreover, LC3-II was upregulated in U87MG sphere cells treated with thioridazine. In addition, thioridazine suppressed GBM tumorigenesis and induced autophagy in vivo. We not only repurposed the antipsychotic drug thioridazine as a potent anti-GBM and anti-GSCs agent, but also provided a new strategy to search for drugs with anticancer and anticancer stem cell properties.Glioblastomas (GBM), the most common and most aggressive primary brain tumors in adults, are classified as grade IV astrocytomas by the World Health Organization and account for 54% of all gliomas.¹ Surgery is typically followed by radiation therapy and chemotherapy with temozolomide (TMZ), which has been in clinical use since 2005.^{2, 3} Despite this multimodal approach, the median survival time of GBM patients is ~14.6 months.³ Therefore, a large number of new drugs are in development for GBM treatment.Instead of focusing on a single drug target, using a batch of genes to query the Connectivity Map (Cmap, http://www.broad.mit.edu/cmap/) may not only allow multiple targets to be considered simultaneously, but it may also identify potential new drugs. Cmap is a database that provides ~7000 microarray expression profiles (conducted on Affymetrix HG-U133A arrays) from four different cancer cell lines treated with 1309 molecular drugs. Of the 1309 drugs included in Cmap, most are currently used in clinical treatment or are well-developed drugs; thus, we can rapidly identify potential drugs and proceed to clinical trial.Thioridazine is an antipsychotic drug and is widely used to treat schizophrenia and psychosis. Recently, it has been shown that patients with schizophrenia have a lower risk of getting cancer (1.93%) than patients without schizophrenia (2.97%).⁴ In addition, inverse cancer comorbidity has been reported in people with certain CNS disorders, and pharmacological treatments is one of possible explanations.⁵Using in silico drug screening via Cmap followed by empirical validations, we discovered that thioridazine can reduce the viability of GBM cells and GBM stem cells, induce autophagy and affect the expressions of related proteins in GBM cells. Thus, thioridazine has potential to treat GBM.     

Gene-Set Local Hierarchical Clustering (GSLHC)—A Gene Set-Based Approach for Characterizing Bioactive Compounds in Terms of Biological Functional Groups

Gene-set-based analysis (GSA), which uses the relative importance of functional gene-sets, or molecular signatures, as units for analysis of genome-wide gene expression data, has exhibited major advantages with respect to greater accuracy, robustness, and biological relevance, over individual gene analysis (IGA), which uses log-ratios of individual genes for analysis. Yet IGA remains the dominant mode of analysis of gene expression data. The Connectivity Map (CMap), an extensive database on genomic profiles of effects of drugs and small molecules and widely used for studies related to repurposed drug discovery, has been mostly employed in IGA mode. Here, we constructed a GSA-based version of CMap, Gene-Set Connectivity Map (GSCMap), in which all the genomic profiles in CMap are converted, using gene-sets from the Molecular Signatures Database, to functional profiles. We showed that GSCMap essentially eliminated cell-type dependence, a weakness of CMap in IGA mode, and yielded significantly better performance on sample clustering and drug-target association. As a first application of GSCMap we constructed the platform Gene-Set Local Hierarchical Clustering (GSLHC) for discovering insights on coordinated actions of biological functions and facilitating classification of heterogeneous subtypes on drug-driven responses. GSLHC was shown to tightly clustered drugs of known similar properties. We used GSLHC to identify the therapeutic properties and putative targets of 18 compounds of previously unknown characteristics listed in CMap, eight of which suggest anti-cancer activities. The GSLHC website http://cloudr.ncu.edu.tw/gslhc/ contains 1,857 local hierarchical clusters accessible by querying 555 of the 1,309 drugs and small molecules listed in CMap. We expect GSCMap and GSLHC to be widely useful in providing new insights in the biological effect of bioactive compounds, in drug repurposing, and in function-based classification of complex diseases.     

Gene expression signature of human HepG2 cell line

Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and is associated with various clinico-pathological characteristics such as genetic mutations and viral infections. Therefore, numerous laboratories look out for identifying always new putative markers for the improvement of HCC diagnosis/prognosis. Many molecular profiling studies investigated gene expression changes related to HCC. HepG2 represents a pure cell line of human liver carcinoma, often used as HCC model due to the absence of viral infection. In this study we compare gene expression profiles associated with HepG2 (as HCC model) and normal hepatocyte cells by microarray technology. Hierarchical cluster analysis of genes evidenced that 2646 genes significantly down-regulated in HepG2 cells compared to hepatocytes whereas a further 3586 genes significantly up-regulated. By using the Ingenuity Pathway Analysis (IPA) program, we have classified the genes that were differently expressed and studied the functional networks correlating these genes in the complete human interactome. Moreover, to confirm the differentially expressed genes as well as the reliability of our microarray data, we performed a quantitative Real time RT-PCR analysis on 9 up-regulated and 11 down-regulated genes, respectively. In conclusion this work i) provides a gene signature of human hepatoma cells showing genes that change their expression as a consequence of liver cancer in the absence of any genetic mutations or viral infection, ii) evidences new differently expressed genes found in our signature compared to previous published studies and iii) suggests some genes on which to focus future studies to understand if they can be used to improve the HCC prognosis/diagnosis.     

Identification and validation of a six-lncRNA prognostic signature with its ceRNA networks and candidate drugs in lower-grade gliomas

Gliomas account for 75% of the primary malignant brain tumors and a majority of lower-grade gliomas (LGG) inevitably develop into glioblastoma. The dysregulation of lncRNAs play a crucial role in LGG. In the present study, we first screened out six differentially expressed lncRNAs (AC021739.2, AL031722.1, AL354740.1, FGD5-AS1, LINC00844, and NEAT1) based on TCGA and GTEx RNA-seq databases. LncRNA prognostic signature was then established by Kaplan–Meier and multivariate Cox proportional hazards regression, with its predictive value validated by time-dependent receiver operating characteristic (ROC) curves. After lncRNA-miRNA-mRNA regulatory networks were established by Cytoscape 3.7.2, Gene Oncology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed, with results enriched in various malignancy-related functions and pathways. Finally, six putative drugs (irinotecan, camptothecin, mitoxantrone, azacitidine, mestranol, and enilconazole) were predicted by Connectivity Map. In conclusion, we identified a 6-lncRNA prognostic signature with its ceRNA networks, and six candidate drugs against LGG.     

Single-cell transcriptomics analysis reveals intratumoral heterogeneity and identifies a gene signature associated with prognosis of hepatocellular carcinoma

Hepatocellular carcinoma (HCC) tumors exhibit high heterogeneity. However, current understanding of tumor cell heterogeneity of HCC and the association with prognosis remains very limited. In the present study, we collected and examined tumor tissue from one HCC patient by single-cell RNA sequencing (scRNA-seq). We identified 5753 cells and 16 clusters including hepatocytes/cancer cells, T cells, macrophages, endothelial cells, fibroblasts, NK cells, neutrophils, and B cells. In six tumor cell subclusters, we identified a cluster of proliferative tumor cells associated with poor prognosis. We downloaded scRNA-seq data of {"type":"entrez-geo","attrs":{"text":"GSE125449","term_id":"125449"}}GSE125449 from the NCBI-GEO as validation dataset, and found that a cluster of hepatocytes exhibited high proliferation activity in HCC. Furthermore, we identified a gene signature related to the proliferation of HCC cells. This gene signature is efficient to classify HCC patients into two groups with distinct prognosis in both TCGA and ICGC database cohorts. Our results reveal the intratumoral heterogeneity of HCC at single cell level and identify a gene signature associated with HCC prognosis.