John Patrick Mpindi

Systematic bioinformatic analysis of expression levels of 17,330 human genes across 9,783 samples from 175 types of healthy and pathological tissues

Our knowledge on tissue- and disease-specific functions of human genes is rather limited and highly context-specific. Here, we have developed a method for the comparison of mRNA expression levels of most human genes across 9,783 Affymetrix gene expression array experiments representing 43 normal human tissue types, 68 cancer types, and 64 other diseases. This database of gene expression patterns in normal human tissues and pathological conditions covers 113 million datapoints and is available from the GeneSapiens website.

FZD4 as a Mediator of ERG Oncogene–Induced WNT Signaling and Epithelial-to-Mesenchymal Transition in Human Prostate Cancer Cells

TMPRSS2-ERG and other gene fusions involving ETS factors and genes with strong promoter elements are common in prostate cancer. Although ERG activation has been linked to invasive properties of prostate cancers, the precise mechanisms and pathways of ERG-mediated oncogenesis remain poorly understood. Here, we show that ERG knockdown in VCaP prostate cancer cells causes an activation of cell adhesion, resulting in strongly induced active beta(1)-integrin and E-cadherin expression as well as changes in WNT signaling. These observations were corroborated by data from ERG-overexpressing nontransformed prostate epithelial cells as well as gene expression data from clinical prostate cancer samples, which both indicated a link between ERG and epithelial-to-mesenchymal transition (EMT). Upregulation of several WNT pathway members was seen in ERG-positive prostate cancers, with frizzled-4 (FZD4) showing the strongest overexpression as verified by both reverse transcription-PCR and immunostaining. Both ERG knockin and knockdown modulated the levels of FZD4 expression. FZD4 silencing could mimic the ERG knockdown phenotype by inducing active beta(1)-integrin and E-cadherin expression, whereas FZD4 overexpression reversed the phenotypic effects seen with ERG knockdown. Taken together, our results provide mechanistic insights to ERG oncogenesis in prostate cancer, involving activation of WNT signaling through FZD4, leading to cancer-promoting phenotypic effects, including EMT and loss of cell adhesion.

GTI: A Novel Algorithm for Identifying Outlier Gene Expression Profiles from Integrated Microarray Datasets

Background Meta-analysis of gene expression microarray datasets presents significant challenges for statistical analysis. We developed and validated a new bioinformatic method for the identification of genes upregulated in subsets of samples of a given tumour type (‘outlier genes’), a hallmark of potential oncogenes. Methodology A new statistical method (the gene tissue index, GTI) was developed by modifying and adapting algorithms originally developed for statistical problems in economics. We compared the potential of the GTI to detect outlier genes in meta-datasets with four previously defined statistical methods, COPA, the OS statistic, the t-test and ORT, using simulated data. We demonstrated that the GTI performed equally well to existing methods in a single study simulation. Next, we evaluated the performance of the GTI in the analysis of combined Affymetrix gene expression data from several published studies covering 392 normal samples of tissue from the central nervous system, 74 astrocytomas, and 353 glioblastomas. According to the results, the GTI was better able than most of the previous methods to identify known oncogenic outlier genes. In addition, the GTI identified 29 novel outlier genes in glioblastomas, including TYMS and CDKN2A. The over-expression of these genes was validated in vivo by immunohistochemical staining data from clinical glioblastoma samples. Immunohistochemical data were available for 65% (19 of 29) of these genes, and 17 of these 19 genes (90%) showed a typical outlier staining pattern. Furthermore, raltitrexed, a specific inhibitor of TYMS used in the therapy of tumour types other than glioblastoma, also effectively blocked cell proliferation in glioblastoma cell lines, thus highlighting this outlier gene candidate as a potential therapeutic target. Conclusions/Significance Taken together, these results support the GTI as a novel approach to identify potential oncogene outliers and drug targets. The algorithm is implemented in an R package (Text S1).

High-Throughput Transcriptomic and RNAi Analysis Identifies AIM1, ERGIC1, TMED3 and TPX2 as Potential Drug Targets in Prostate Cancer

Prostate cancer is a heterogeneous group of diseases and there is a need for more efficient and targeted methods of treatment. In this study, the potential of gene expression data and RNA interference technique were combined to advance future personalized prostate cancer therapeutics. To distinguish the most promising in vivo prevalidated prostate cancer drug targets, a bioinformatic analysis was carried out using genome-wide gene expression data from 9873 human tissue samples. In total, 295 genes were selected for further functional studies in cultured prostate cancer cells due to their high mRNA expression in prostate, prostate cancer or in metastatic prostate cancer samples. Second, RNAi based cell viability assay was performed in VCaP and LNCaP prostate cancer cells. Based on the siRNA results, gene expression patterns in human tissues and novelty, endoplasmic reticulum function associated targets AIM1, ERGIC1 and TMED3, as well as mitosis regulating TPX2 were selected for further validation. AIM1, ERGIC1, and TPX2 were shown to be highly expressed especially in prostate cancer tissues, and high mRNA expression of ERGIC1 and TMED3 associated with AR and ERG oncogene expression. ERGIC1 silencing specifically regulated the proliferation of ERG oncogene positive prostate cancer cells and inhibited ERG mRNA expression in these cells, indicating that it is a potent drug target in ERG positive subgroup of prostate cancers. TPX2 expression associated with PSA failure and TPX2 silencing reduced PSA expression, indicating that TPX2 regulates androgen receptor mediated signaling. In conclusion, the combinatorial usage of microarray and RNAi techniques yielded in a large number of potential novel biomarkers and therapeutic targets, for future development of targeted and personalized approaches for prostate cancer management.

Comprehensive data-driven analysis of the impact of chemoinformatic structure on the genome-wide biological response profiles of cancer cells to 1159 drugs

BackgroundDetailed and systematic understanding of the biological effects of millions of available compounds on living cells is a significant challenge. As most compounds impact multiple targets and pathways, traditional methods for analyzing structure-function relationships are not comprehensive enough. Therefore more advanced integrative models are needed for predicting biological effects elicited by specific chemical features. As a step towards creating such computational links we developed a data-driven chemical systems biology approach to comprehensively study the relationship of 76 structural 3D-descriptors (VolSurf, chemical space) of 1159 drugs with the microarray gene expression responses (biological space) they elicited in three cancer cell lines. The analysis covering 11350 genes was based on data from the Connectivity Map. We decomposed the biological response profiles into components, each linked to a characteristic chemical descriptor profile.ResultsIntegrated analysis of both the chemical and biological space was more informative than either dataset alone in predicting drug similarity as measured by shared protein targets. We identified ten major components that link distinct VolSurf chemical features across multiple compounds to specific cellular responses. For example, component 2 (hydrophobic properties) strongly linked to DNA damage response, while component 3 (hydrogen bonding) was associated with metabolic stress. Individual structural and biological features were often linked to one cell line only, such as leukemia cells (HL-60) specifically responding to cardiac glycosides.ConclusionsIn summary, our approach identified several novel links between specific chemical structure properties and distinct biological responses in cells incubated with these drugs. Importantly, the analysis focused on chemical-biological properties that emerge across multiple drugs. The decoding of such systematic relationships is necessary to build better models of drug effects, including unanticipated types of molecular properties having strong biological effects.

Consistency in drug response profiling

The comparative analysis by Haibe-Kains et al.1 concluded that data from two large-scale studies of cancer cell lines2,3 showed highly discordant results for drug sensitivity measurements, whereas gene expression data were reasonably concordant. Here, we crosscompared the two original datasets2,3 against our own data of drug response profiles in overlapping cancer cell line panels. Our results indicate that it is possible to achieve concordance between different laboratories for drug response measurements by paying attention to the harmonization of assays and experimental procedures. There is a Reply to this Comment by Safikhani, Z. et al. Nature 540, http://dx.doi.org/10.1038/ nature20172 (2016). Haibe-Kains et al.1 reported on a comparative evaluation of two drug sensitivity and molecular profiling datasets, one from the Cancer Genome Project (CGP)2 and the other from the Cancer Cell Line Encyclopedia (CCLE)3. In their analyses, gene expression profiles between hundreds of common cancer cell lines across all genes showed high consistency between the two studies (median rank correlation (MRC) = 0.85), whereas the drug response data for 15 common compounds were highly discordant (MRC = 0.28 for halfmaximum inhibitory concentration (IC50) values). This report1 and the accompanying commentary4 suggested that differences in laboratory protocols, compounds and their tested concentration ranges, and computational methods may account for the differences, but these reports did not elaborate which of these factors are important and whether they can be controlled for. Here, we reanalysed the dose–response data from both CGP and CCLE using a standardized area under the curve (AUC) response metric, which we call the drug sensitivity score (DSS)5. We then compared the CGP and CCLE data with a new dataset of drug responses profiled using the Institute for Molecular Medicine Finland (FIMM) compound testing assay6, covering 308 drugs across 106 cancer cell lines. The FIMM data included 45 compounds in common with CGP and 14 with the CCLE in 50 cell lines (Supplementary Data 1). In the AUC calculation, we unified the drug concentration ranges across the CGP, CCLE and FIMM assays. We observed a significantly higher level of consistency (P = 4.2 × 10−5), especially between the CCLE and FIMM drug response data (MRC = 0.74), as compared to the consistency between FIMM and CGP data (MRC = 0.54) (Fig. 1a). Similar experimental protocols were applied at FIMM and CCLE, including the same readout (CellTiter-Glo, Promega), similar controls (vehicle as negative control and positive controls of toxic compounds 100 μ M benzethonium chloride or 1 μ M MG132). However, there were also differences, such as the plate format used (1,536 versus 384 wells). Importantly, there was no effort made to standardize cell numbers used or any other parameters between the three laboratories, such as the source, passage number and media used for cells, nor the origin and handling of drugs. Therefore, this observed level of drug response agreement could be substantially improved by further standardization of the laboratory protocols. The CGP experimental protocol differed from the two others in terms of the readout (fluorescent nucleic acid stain Syto 60, Life Technologies), in the use of controls (drug-free cells as negative and no cells as positive controls), and the plate format used (96or 384-well plates). We compared the drug response profiles between the same cell lines from different laboratories, in line with the approach of Haibe-Kains et al.1, in which they showed consistency in gene expression profiles from CGP and CCLE (MRC = 0.85)1. The Haibe-Kains et al.1 approach, in which the correlation is calculated for each drug separately across the cell lines, showed more variability (Fig. 1b), owing to the fact that some drugs show minimal efficacy in all the tested cell lines. Analogously, gene expression correlations vary more widely when analysed at the level of genes across cell lines (MRC = 0.58 between CGP and CCLE), as certain genes are not expressed above technical noise. Although both ways to compare the data are relevant to the overall goal of personalized

Comprehensive Drug Testing of Patient-derived Conditionally Reprogrammed Cells from Castration-resistant Prostate Cancer

BACKGROUND Technology development to enable the culture of human prostate cancer (PCa) progenitor cells is required for the identification of new, potentially curative therapies for PCa. OBJECTIVE We established and characterized patient-derived conditionally reprogrammed cells (CRCs) to assess their biological properties and to apply these to test the efficacies of drugs. DESIGN, SETTING, AND PARTICIPANTS CRCs were established from seven patient samples with disease ranging from primary PCa to advanced castration-resistant PCa (CRPC). The CRCs were characterized by genomic, transcriptomic, protein expression, and drug profiling. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS The phenotypic quantification of the CRCs was done based on immunostaining followed by image analysis with Advanced Cell Classifier using Random Forest supervised machine learning. Copy number aberrations (CNAs) were called from whole-exome sequencing and transcriptomics using in-house pipelines. Dose-response measurements were used to generate multiparameter drug sensitivity scores using R-statistical language. RESULTS AND LIMITATIONS We generated six benign CRC cultures which all had an androgen receptor-negative, basal/transit-amplifying phenotype with few CNAs. In three-dimensional cell culture, these cells could re-express the androgen receptor. The CRCs from a CRPC patient (HUB.5) displayed multiple CNAs, many of which were shared with the parental tumor. We carried out high-throughput drug-response studies with 306 emerging and clinical cancer drugs. Using the benign CRCs as controls, we identified the Bcl-2 family inhibitor navitoclax as the most potent cancer-specific drug for the CRCs from a CRPC patient. Other drug efficacies included taxanes, mepacrine, and retinoids. CONCLUSIONS Comprehensive cancer pharmacopeia-wide drug testing of CRCs from a CRPC patient highlighted both known and novel drug sensitivities in PCa, including navitoclax, which is currently being tested in clinical trials of CRPC. PATIENT SUMMARY We describe an approach to generate patient-derived cancer cells from advanced prostate cancer and apply such cells to discover drugs that could be applied in clinical trials for castration-resistant prostate cancer.

Contribution of ARLTS1 Cys148Arg (T442C) Variant with Prostate Cancer Risk and ARLTS1 Function in Prostate Cancer Cells

ARLTS1 is a recently characterized tumor suppressor gene at 13q14.3, a region frequently deleted in both sporadic and hereditary prostate cancer (PCa). ARLTS1 variants, especially Cys148Arg (T442C), increase susceptibility to different cancers, including PCa. In this study the role of Cys148Arg substitution was investigated as a risk factor for PCa using both genetic and functional analysis. Cys148Arg genotypes and expression of the ARLTS1 were explored in a large set of familial and unselected PCa cases, clinical tumor samples, xenografts, prostate cancer cell lines and benign prostatic hyperplasia (BPH) samples. The frequency of the variant genotype CC was significantly higher in familial (OR = 1.67, 95% CI = 1.08–2.56, P = 0.019) and unselected patients (OR = 1.52, 95% CI = 1.18–1.97, P = 0.001) and the overall risk was increased (OR = 1.54, 95% CI = 1.20–1.98, P = 0.0007). Additional analysis with clinicopathological data revealed an association with an aggressive disease (OR = 1.28, 95% CI = 1.05-∞, P = 0.02). The CC genotype of the Cys148Arg variant was also contributing to the lowered ARLTS1 expression status in lymphoblastoid cells from familial patients. In addition significantly lowered ARLTS1 expression was observed in clinical tumor samples compared to BPH samples (P = 0.01). The ARLTS1 co-expression signature based on previously published microarray data was generated from 1587 cancer samples confirming the low expression of ARLTS1 in PCa and showed that ARLTS1 expression was strongly associated with immune processes. This study provides strong confirmation of the important role of ARLTS1 Cys148Arg variant as a contributor in PCa predisposition and a potential marker for aggressive disease outcome.

ARLTS1 and Prostate Cancer Risk - Analysis of Expression and Regulation

Prostate cancer (PCa) is a heterogeneous trait for which several susceptibility loci have been implicated by genome-wide linkage and association studies. The genomic region 13q14 is frequently deleted in tumour tissues of both sporadic and familial PCa patients and is consequently recognised as a possible locus of tumour suppressor gene(s). Deletions of this region have been found in many other cancers. Recently, we showed that homozygous carriers for the T442C variant of the ARLTS1 gene (ADP-ribosylation factor-like tumour suppressor protein 1 or ARL11, located at 13q14) are associated with an increased risk for both unselected and familial PCa. Furthermore, the variant T442C was observed in greater frequency among malignant tissue samples, PCa cell lines and xenografts, supporting its role in PCa tumourigenesis. In this study, 84 PCa cases and 15 controls were analysed for ARLTS1 expression status in blood-derived RNA. A statistically significant (p = 0.0037) decrease of ARLTS1 expression in PCa cases was detected. Regulation of ARLTS1 expression was analysed with eQTL (expression quantitative trait loci) methods. Altogether fourteen significant cis-eQTLs affecting the ARLTS1 expression level were found. In addition, epistatic interactions of ARLTS1 genomic variants with genes involved in immune system processes were predicted with the MDR program. In conclusion, this study further supports the role of ARLTS1 as a tumour suppressor gene and reveals that the expression is regulated through variants localised in regulatory regions.

Systemic Analysis of Gene Expression Profiles Identifies ErbB3 as a Potential Drug Target in Pediatric Alveolar Rhabdomyosarcoma

Pediatric sarcomas, including rhabdomyosarcomas, Ewing’s sarcoma, and osteosarcoma, are aggressive tumors with poor survival rates. To overcome problems associated with nonselectivity of the current therapeutic approaches, targeted therapeutics have been developed. Currently, an increasing number of such drugs are used for treating malignancies of adult patients but little is known about their effects in pediatric patients. We analyzed expression of 24 clinically approved target genes in a wide variety of pediatric normal and malignant tissues using a novel high-throughput systems biology approach. Analysis of the Genesapiens database of human transcriptomes demonstrated statistically significant up-regulation of VEGFC and EPHA2 in Ewing’s sarcoma, and ERBB3 in alveolar rhabdomyosarcomas. In silico data for ERBB3 was validated by demonstrating ErbB3 protein expression in pediatric rhabdomyosarcoma in vitro and in vivo. ERBB3 overexpression promoted whereas ERBB3-targeted siRNA suppressed rhabdomyosarcoma cell gowth, indicating a functional role for ErbB3 signaling in rhabdomyosarcoma. These data suggest that drugs targeting ErbB3, EphA2 or VEGF-C could be further tested as therapeutic targets for pediatric sarcomas.