SIGNATURE: a workbench for gene expression signature analysis

NKG2A discriminates natural killer cells with a suppressed phenotype in pediatric acute leukemia

Natural killer (NK) cells are important for early tumor immune surveillance. In patients with hematological cancers, NK cells are generally functional deficient and display dysregulations in their receptor repertoires. Acute leukemia is the most common cancer in children, and we here performed a comparative phenotypic profiling of NK cells from B-cell precursor acute lymphoblastic leukemia (BCP-ALL) patients to identify aberrant NK cell phenotypes. NK cell phenotypes, maturation, and function were analyzed in matched bone marrow and blood NK cells from BCP-ALL patients at diagnosis, during treatment, and at end of treatment and compared with age-matched pediatric control subjects. Expression of several markers were skewed in patients, but with large interindividual variations. Undertaking a multiparameter approach, we found that high expression levels of NKG2A was the single predominant marker distinguishing NK cells in BCP-ALL patients compared with healthy control subjects. Moreover, naïve CD57-NKG2A NK cells dominated in BCP-ALL patients at diagnosis. Further, we found dysregulated expression of the activating receptor DNAM-1 in resident bone marrow CXCR6+ NK cells. CXCR6+ NK cells lacking DNAM-1 expressed NKG2A and had a tendency for lower degranulation activity. In conclusion, high expression of NKG2A dominates NK cell phenotypes from pediatric BCP-ALL patients, indicating that NKG2A could be targeted in therapies for this patient group.

Navigating Transcriptomic Connectivity Mapping Workflows to Link Chemicals with Bioactivities

Screening new compounds for potential bioactivities against cellular targets is vital for drug discovery and chemical safety. Transcriptomics offers an efficient approach for assessing global gene expression changes, but interpreting chemical mechanisms from these data is often challenging. Connectivity mapping is a potential data-driven avenue for linking chemicals to mechanisms based on the observation that many biological processes are associated with unique gene expression signatures (gene signatures). However, mining the effects of a chemical on gene signatures for biological mechanisms is challenging because transcriptomic data contain thousands of noisy genes. New connectivity mapping approaches seeking to distinguish signal from noise continue to be developed, spurred by the promise of discovering chemical mechanisms, new drugs, and disease targets from burgeoning transcriptomic data. Here, we analyze these approaches in terms of diverse transcriptomic technologies, public databases, gene signatures, pattern-matching algorithms, and statistical evaluation criteria. To navigate the complexity of connectivity mapping, we propose a harmonized scheme to coherently organize and compare published workflows. We first standardize concepts underlying transcriptomic profiles and gene signatures based on various transcriptomic technologies such as microarrays, RNA-Seq, and L1000 and discuss the widely used data sources such as Gene Expression Omnibus, ArrayExpress, and MSigDB. Next, we generalize connectivity mapping as a pattern-matching task for finding similarity between a query (e.g., transcriptomic profile for new chemical) and a reference (e.g., gene signature of known target). Published pattern-matching approaches fall into two main categories: vector-based use metrics like correlation, Jaccard index, etc., and aggregation-based use parametric and nonparametric statistics (e.g., gene set enrichment analysis). The statistical methods for evaluating the performance of different approaches are described, along with comparisons reported in the literature on benchmark transcriptomic data sets. Lastly, we review connectivity mapping applications in toxicology and offer guidance on evaluating chemical-induced toxicity with concentration-response transcriptomic data. In addition to serving as a high-level guide and tutorial for understanding and implementing connectivity mapping workflows, we hope this review will stimulate new algorithms for evaluating chemical safety and drug discovery using transcriptomic data.

Predicting clinical response to everolimus in ER+ breast cancers using machine-learning

Endocrine therapy remains the primary treatment choice for ER+ breast cancers. However, most advanced ER+ breast cancers ultimately develop resistance to endocrine. This acquired resistance to endocrine therapy is often driven by the activation of the PI3K/AKT/mTOR signaling pathway. Everolimus, a drug that targets and inhibits the mTOR complex has been shown to improve clinical outcomes in metastatic ER+ breast cancers. However, there are no biomarkers currently available to guide the use of everolimus in the clinic for progressive patients, where multiple therapeutic options are available. Here, we utilized gene expression signatures from 9 ER+ breast cancer cell lines and 23 patients treated with everolimus to develop and validate an integrative machine learning biomarker of mTOR inhibitor response. Our results show that the machine learning biomarker can successfully distinguish responders from non-responders and can be applied to identify patients that will most likely benefit from everolimus treatment.

AKT1 mediates multiple phosphorylation events that functionally promote HSF1 activation

The heat stress response activates the transcription factor heat shock factor 1 (HSF1), which subsequently upregulates heat shock proteins to maintain the integrity of the proteome. HSF1 activation requires nuclear localization, trimerization, DNA binding, phosphorylation and gene transactivation. Phosphorylation at S326 is an important regulator of HSF1 transcriptional activity. Phosphorylation at S326 is mediated by AKT1, mTOR, p38, MEK1 and DYRK2. Here, we observed activation of HSF1 by AKT1 independently of mTOR. AKT2 also phosphorylated S326 of HSF1 but showed weak ability to activate HSF1. Similarly, mTOR, p38, MEK1 and DYRK2 all phosphorylated S326 but AKT1 was the most potent activator. Mass spectrometry showed that AKT1 also phosphorylated HSF1 at T142, S230 and T527 in addition to S326, whereas the other kinases did not. Subsequent investigation revealed that phosphorylation at T142 is necessary for HSF1 trimerization and that S230, S326 and T527 are required for HSF1 gene transactivation and recruitment of TFIIB and CDK9. Interestingly, T527 as a phosphorylated residue has not been previously shown and sits in the transactivation domain, further implying a role for this site in HSF1 gene transactivation. This study suggests that HSF1 hyperphosphorylation is targeted and these specific residues have direct function in regulating HSF1 transcriptional activity.

Systematic identification of ACE2 expression modulators reveals cardiomyopathy as a risk factor for mortality in COVID-19 patients

BACKGROUND

Angiotensin-converting enzyme 2 (ACE2) is the cell-entry receptor for SARS-CoV-2. It plays critical roles in both the transmission and the pathogenesis of COVID-19. Comprehensive profiling of ACE2 expression patterns could reveal risk factors of severe COVID-19 illness. While the expression of ACE2 in healthy human tissues has been well characterized, it is not known which diseases and drugs might be associated with ACE2 expression.

RESULTS

We develop GENEVA (GENe Expression Variance Analysis), a semi-automated framework for exploring massive amounts of RNA-seq datasets. We apply GENEVA to 286,650 publicly available RNA-seq samples to identify any previously studied experimental conditions that could be directly or indirectly associated with ACE2 expression. We identify multiple drugs, genetic perturbations, and diseases that are associated with the expression of ACE2, including cardiomyopathy, HNF1A overexpression, and drug treatments with RAD140 and itraconazole. Our joint analysis of seven datasets confirms ACE2 upregulation in all cardiomyopathy categories. Using electronic health records data from 3936 COVID-19 patients, we demonstrate that patients with pre-existing cardiomyopathy have an increased mortality risk than age-matched patients with other cardiovascular conditions. GENEVA is applicable to any genes of interest and is freely accessible at http://genevatool.org .

CONCLUSIONS

This study identifies multiple diseases and drugs that are associated with the expression of ACE2. The effect of these conditions should be carefully studied in COVID-19 patients. In particular, our analysis identifies cardiomyopathy patients as a high-risk group, with increased ACE2 expression in the heart and increased mortality after SARS-COV-2 infection.

Mammary-specific expression of Trim24 establishes a mouse model of human metaplastic breast cancer

Conditional overexpression of histone reader Tripartite motif containing protein 24 (TRIM24) in mouse mammary epithelia (Trim24COE) drives spontaneous development of mammary carcinosarcoma tumors, lacking ER, PR and HER2. Human carcinosarcomas or metaplastic breast cancers (MpBC) are a rare, chemorefractory subclass of triple-negative breast cancers (TNBC). Comparison of Trim24COE metaplastic carcinosarcoma morphology, TRIM24 protein levels and a derived Trim24COE gene signature reveals strong correlation with human MpBC tumors and MpBC patient-derived xenograft (PDX) models. Global and single-cell tumor profiling reveal Met as a direct oncogenic target of TRIM24, leading to aberrant PI3K/mTOR activation. Here, we find that pharmacological inhibition of these pathways in primary Trim24COE tumor cells and TRIM24-PROTAC treatment of MpBC TNBC PDX tumorspheres decreased cellular viability, suggesting potential in therapeutically targeting TRIM24 and its regulated pathways in TRIM24-expressing TNBC.

TrkA Interacts with and Phosphorylates STAT3 to Enhance Gene Transcription and Promote Breast Cancer Stem Cells in Triple-Negative and HER2-Enriched Breast Cancers

Simple Summary

Breast cancer is the leading cancer in American women. Due to the inherent aggressiveness of triple-negative and HER2-enriched breast cancers, it is imperative to identify novel molecular targets for therapeutic intervention. Due to their abnormal activities in metastatic breast cancers, JAK2–STAT3 and TrkA pathways have been individually implicated in aggressive breast tumors. However, their co-activation and signaling interactions have not been thoroughly investigated. Therefore, our study aimed to elucidate the extent of crosstalk between JAK2–STAT3 and TrkA signaling pathways and its impact on breast cancer. Our data revealed a novel interaction between TrkA and STAT3, and that this interaction results in STAT3 phosphorylation and activation by TrkA, leading to enhanced stemness gene expression and stem cell renewal. We further found that the co-activation of JAK2–STAT3 and TrkA pathways is correlated with shorter time to develop overall and organ-specific metastasis, suggesting that this signaling crosstalk underlies the aggressiveness of triple-negative and HER2-enriched breast cancers.

Abstract

JAK2–STAT3 and TrkA signaling pathways have been separately implicated in aggressive breast cancers; however, whether they are co-activated or undergo functional interaction has not been thoroughly investigated. Herein we report, for the first time that STAT3 and TrkA are significantly co-overexpressed and co-activated in triple-negative breast cancer (TNBC) and HER2-enriched breast cancer, as shown by immunohistochemical staining and data mining. Through immunofluorescence staining–confocal microscopy and immunoprecipitation–Western blotting, we found that TrkA and STAT3 co-localize and physically interact in the cytoplasm, and the interaction is dependent on STAT3-Y705 phosphorylation. TrkA–STAT3 interaction leads to STAT3 phosphorylation at Y705 by TrkA in breast cancer cells and cell-free kinase assays, indicating that STAT3 is a novel substrate of TrkA. β-NGF-mediated TrkA activation induces TrkA–STAT3 interaction, STAT3 nuclear transport and transcriptional activity, and the expression of STAT3 target genes, SOX2 and MYC. The co-activation of both pathways promotes breast cancer stem cells. Finally, we found that TNBC and HER2-enriched breast cancer with JAK2–STAT3 and TrkA co-activation are positively associated with poor overall metastasis-free and organ-specific metastasis-free survival. Collectively, our study uncovered that TrkA is a novel activating kinase of STAT3, and their co-activation enhances gene transcription and promotes breast cancer stem cells in TNBC and HER2-enriched breast cancer.

signatureSearch: environment for gene expression signature searching and functional interpretation

signatureSearch is an R/Bioconductor package that integrates a suite of existing and novel algorithms into an analysis environment for gene expression signature (GES) searching combined with functional enrichment analysis (FEA) and visualization methods to facilitate the interpretation of the search results. In a typical GES search (GESS), a query GES is searched against a database of GESs obtained from large numbers of measurements, such as different genetic backgrounds, disease states and drug perturbations. Database matches sharing correlated signatures with the query indicate related cellular responses frequently governed by connected mechanisms, such as drugs mimicking the expression responses of a disease. To identify which processes are predominantly modulated in the GESS results, we developed specialized FEA methods combined with drug-target network visualization tools. The provided analysis tools are useful for studying the effects of genetic, chemical and environmental perturbations on biological systems, as well as searching single cell GES databases to identify novel network connections or cell types. The signatureSearch software is unique in that it provides access to an integrated environment for GESS/FEA routines that includes several novel search and enrichment methods, efficient data structures, and access to pre-built GES databases, and allowing users to work with custom databases.

Merging microarray studies to identify a common gene expression signature to several structural heart diseases

BACKGROUND

Heart disease is the leading cause of death worldwide. Knowing a gene expression signature in heart disease can lead to the development of more efficient diagnosis and treatments that may prevent premature deaths. A large amount of microarray data is available in public repositories and can be used to identify differentially expressed genes. However, most of the microarray datasets are composed of a reduced number of samples and to obtain more reliable results, several datasets have to be merged, which is a challenging task. The identification of differentially expressed genes is commonly done using statistical methods. Nonetheless, these methods are based on the definition of an arbitrary threshold to select the differentially expressed genes and there is no consensus on the values that should be used.

RESULTS

Nine publicly available microarray datasets from studies of different heart diseases were merged to form a dataset composed of 689 samples and 8354 features. Subsequently, the adjusted p-value and fold change were determined and by combining a set of adjusted p-values cutoffs with a list of different fold change thresholds, 12 sets of differentially expressed genes were obtained. To select the set of differentially expressed genes that has the best accuracy in classifying samples from patients with heart diseases and samples from patients with no heart condition, the random forest algorithm was used. A set of 62 differentially expressed genes having a classification accuracy of approximately 95% was identified.

CONCLUSIONS

We identified a gene expression signature common to different cardiac diseases and supported our findings by showing their involvement in the pathophysiology of the heart. The approach used in this study is suitable for the identification of gene expression signatures, and can be extended to different diseases.

Nrf2 and SQSTM1/p62 jointly contribute to mesenchymal transition and invasion in glioblastoma

Accumulating evidence suggests that constitutively active Nrf2 has a pivotal role in cancer as it induces pro-survival genes that promote cancer cell proliferation and chemoresistance. The mechanisms of Nrf2 dysregulation and functions in cancer have not been fully characterized. Here, we jointly analyzed the Broad-Novartis Cancer Cell Line Encyclopedia (CCLE) and the Cancer Genome Atlas (TCGA) multi-omics data in order to identify cancer types where Nrf2 activation is present. We found that Nrf2 is hyperactivated in a subset of glioblastoma (GBM) patients, whose tumors display a mesenchymal subtype, and uncover several different mechanisms contributing to increased Nrf2 activity. Importantly, we identified a positive feedback loop between SQSTM1/p62 and Nrf2 as a mechanism for activation of the Nrf2 pathway. We also show that autophagy and serine/threonine signaling regulates p62 mediated Keap1 degradation. Our results in glioma cell lines indicate that both Nrf2 and p62 promote proliferation, invasion and mesenchymal transition. Finally, Nrf2 activity was associated with decreased progression free survival in TCGA GBM patient samples, suggesting that treatments have limited efficacy if this transcription factor is overactivated. Overall, our findings place Nrf2 and p62 as the key components of the mesenchymal subtype network, with implications to tumorigenesis and treatment resistance. Thus, Nrf2 activation could be used as a surrogate prognostic marker in mesenchymal subtype GBMs. Furthermore, strategies aiming at either inhibiting Nrf2 or exploiting Nrf2 hyperactivity for targeted gene therapy may provide novel treatment options for this subset of GBM.

Targeting the Interplay between Epithelial-to-Mesenchymal-Transition and the Immune System for Effective Immunotherapy

Over the last decade, both early diagnosis and targeted therapy have improved the survival rates of many cancer patients. Most recently, immunotherapy has revolutionized the treatment options for cancers such as melanoma. Unfortunately, a significant portion of cancers (including lung and breast cancers) do not respond to immunotherapy, and many of them develop resistance to chemotherapy. Molecular characterization of non-responsive cancers suggest that an embryonic program known as epithelial-mesenchymal transition (EMT), which is mostly latent in adults, can be activated under selective pressures, rendering these cancers resistant to chemo- and immunotherapies. EMT can also drive tumor metastases, which in turn also suppress the cancer-fighting activity of cytotoxic T cells that traffic into the tumor, causing immunotherapy to fail. In this review, we compare and contrast immunotherapy treatment options of non-small cell lung cancer (NSCLC) and triple negative breast cancer (TNBC). We discuss why, despite breakthrough progress in immunotherapy, attaining predictable outcomes in the clinic is mostly an unsolved problem for these tumors. Although these two cancer types appear different based upon their tissues of origin and molecular classification, gene expression indicate that they possess many similarities. Patient tumors exhibit activation of EMT, and resulting stem cell properties in both these cancer types associate with metastasis and resistance to existing cancer therapies. In addition, the EMT transition in both these cancers plays a crucial role in immunosuppression, which exacerbates treatment resistance. To improve cancer-related survival we need to understand and circumvent, the mechanisms through which these tumors become therapy resistant. In this review, we discuss new information and complementary perspectives to inform combination treatment strategies to expand and improve the anti-tumor responses of currently available clinical immune checkpoint inhibitors.

Contributions of the RhoA guanine nucleotide exchange factor Net1 to polyoma middle T antigen-mediated mammary gland tumorigenesis and metastasis

Background

The RhoA activating protein Net1 contributes to breast cancer cell proliferation, motility, and invasion in vitro, yet little is known about its roles in mammary gland tumorigenesis and metastasis.

Methods

Net1 knockout (KO) mice were bred to mice with mammary gland specific expression of the polyoma middle T antigen (PyMT) oncogene. Mammary gland tumorigenesis and lung metastasis were monitored. Individual tumors were assessed for proliferation, apoptosis, angiogenesis, RhoA activation, and activation of PyMT-dependent signaling pathways. Primary tumor cells from wild-type and Net1 KO mice were transplanted into the mammary glands of wild-type, nontumor-bearing mice, and tumor growth and metastasis were assessed. Gene expression in wild-type and Net1 KO tumors was analyzed by gene ontology enrichment and for relative activation of gene expression signatures indicative of signaling pathways important for breast cancer initiation and progression. A gene expression signature indicative of Net1 function was identified. Human breast cancer gene expression profiles were screened for the presence of a Net1 gene expression signature.

Results

We show that Net1 makes fundamental contributions to mammary gland tumorigenesis and metastasis. Net1 deletion delays tumorigenesis and strongly suppresses metastasis in PyMT-expressing mice. Moreover, we observe that loss of Net1 reduces cancer cell proliferation, inhibits tumor angiogenesis, and promotes tumor cell apoptosis. Net1 is required for maximal RhoA activation within tumors and for primary tumor cell motility. Furthermore, the ability of PyMT to initiate oncogenic signaling to ERK1/2 and PI3K/Akt1 is inhibited by Net1 deletion. Primary tumor cell transplantation indicates that the reduction in tumor angiogenesis and lung metastasis observed upon Net1 deletion are tumor cell autonomous effects. Using a gene expression signature indicative of Net1 activity, we show that Net1 signaling is activated in 10% of human breast cancers, and that this correlates with elevated proliferation and PI3K pathway activity. We also demonstrate that human breast cancer patients with a high Net1 gene expression signature experience shorter distant metastasis-free survival.

Conclusions

These data indicate that Net1 is required for tumor progression in the PyMT mouse model and suggest that Net1 may contribute to breast cancer progression in humans.

Electronic supplementary material

The online version of this article (10.1186/s13058-018-0966-2) contains supplementary material, which is available to authorized users.

Planning bioinformatics workflows using an expert system

Motivation

Bioinformatic analyses are becoming formidably more complex due to the increasing number of steps required to process the data, as well as the proliferation of methods that can be used in each step. To alleviate this difficulty, pipelines are commonly employed. However, pipelines are typically implemented to automate a specific analysis, and thus are difficult to use for exploratory analyses requiring systematic changes to the software or parameters used.

Results

To automate the development of pipelines, we have investigated expert systems. We created the Bioinformatics ExperT SYstem (BETSY) that includes a knowledge base where the capabilities of bioinformatics software is explicitly and formally encoded. BETSY is a backwards-chaining rule-based expert system comprised of a data model that can capture the richness of biological data, and an inference engine that reasons on the knowledge base to produce workflows. Currently, the knowledge base is populated with rules to analyze microarray and next generation sequencing data. We evaluated BETSY and found that it could generate workflows that reproduce and go beyond previously published bioinformatics results. Finally, a meta-investigation of the workflows generated from the knowledge base produced a quantitative measure of the technical burden imposed by each step of bioinformatics analyses, revealing the large number of steps devoted to the pre-processing of data. In sum, an expert system approach can facilitate exploratory bioinformatic analysis by automating the development of workflows, a task that requires significant domain expertise.

Availability and Implementation

https://github.com/jefftc/changlab.

Contact

jeffrey.t.chang@uth.tmc.edu.

Gene expression signature: a powerful approach for drug discovery in diabetes

Type 2 diabetes (T2D) is increasing in prevalence at an alarming rate around the world. Much effort has gone into the discovery and design of antidiabetic drugs; however, those already available are unable to combat the underlying causes of the disease and instead only moderate the symptoms. The reason for this is that T2D is a complex disease, and attempts to target one biological pathway are insufficient to combat the full extent of the disease. Additionally, the underlying pathophysiology of this disease is yet to be fully elucidated making it difficult to design drugs that target the mechanisms involved. Therefore, the approach of designing new drugs aimed at a specific molecular target is not optimal and a more expansive, unbiased approach is required. In this review, we will look at the current state of diabetes treatments and how these target the disease symptoms but are unable to combat the underlying causes. We will also review how the technique of gene expression signatures (GESs) has been used successfully for other complex diseases and how this may be applied as a powerful tool for the discovery of new drugs for T2D.

Extraction and analysis of signatures from the Gene Expression Omnibus by the crowd

Gene expression data are accumulating exponentially in public repositories. Reanalysis and integration of themed collections from these studies may provide new insights, but requires further human curation. Here we report a crowdsourcing project to annotate and reanalyse a large number of gene expression profiles from Gene Expression Omnibus (GEO). Through a massive open online course on Coursera, over 70 participants from over 25 countries identify and annotate 2,460 single-gene perturbation signatures, 839 disease versus normal signatures, and 906 drug perturbation signatures. All these signatures are unique and are manually validated for quality. Global analysis of these signatures confirms known associations and identifies novel associations between genes, diseases and drugs. The manually curated signatures are used as a training set to develop classifiers for extracting similar signatures from the entire GEO repository. We develop a web portal to serve these signatures for query, download and visualization.

ACLY and ACC1 Regulate Hypoxia-Induced Apoptosis by Modulating ETV4 via α-ketoglutarate

In order to propagate a solid tumor, cancer cells must adapt to and survive under various tumor microenvironment (TME) stresses, such as hypoxia or lactic acidosis. To systematically identify genes that modulate cancer cell survival under stresses, we performed genome-wide shRNA screens under hypoxia or lactic acidosis. We discovered that genetic depletion of acetyl-CoA carboxylase (ACACA or ACC1) or ATP citrate lyase (ACLY) protected cancer cells from hypoxia-induced apoptosis. Additionally, the loss of ACLY or ACC1 reduced levels and activities of the oncogenic transcription factor ETV4. Silencing ETV4 also protected cells from hypoxia-induced apoptosis and led to remarkably similar transcriptional responses as with silenced ACLY or ACC1, including an anti-apoptotic program. Metabolomic analysis found that while α-ketoglutarate levels decrease under hypoxia in control cells, α-ketoglutarate is paradoxically increased under hypoxia when ACC1 or ACLY are depleted. Supplementation with α-ketoglutarate rescued the hypoxia-induced apoptosis and recapitulated the decreased expression and activity of ETV4, likely via an epigenetic mechanism. Therefore, ACC1 and ACLY regulate the levels of ETV4 under hypoxia via increased α-ketoglutarate. These results reveal that the ACC1/ACLY-α-ketoglutarate-ETV4 axis is a novel means by which metabolic states regulate transcriptional output for life vs. death decisions under hypoxia. Since many lipogenic inhibitors are under investigation as cancer therapeutics, our findings suggest that the use of these inhibitors will need to be carefully considered with respect to oncogenic drivers, tumor hypoxia, progression and dormancy. More broadly, our screen provides a framework for studying additional tumor cell stress-adaption mechanisms in the future.

During the development of most solid tumors, there are characteristic physiological differences in the tumor that result from tumor cells outgrowing their local blood supply. Two of these physiological differences, or “stresses,” that occur in the tumor are low oxygen levels (hypoxia) and an accumulation of lactic acidic (lactic acidosis). Cancer cells experiencing hypoxia and lactic acidosis tend to be more resistant to chemo- and radio-therapy and metastasize more readily. Therefore, it is important to understand how tumor cells adapt to and survive these stresses. We used a large scale screening experiment in order to find which genes and proteins are involved in tumor cell adaptation and survival under hypoxia or lactic acidosis. We found that inhibiting either of two genes involved in lipid synthesis allowed tumor cells to survive hypoxia. This occurred because silencing these genes led to an increase in the metabolite α-ketoglutarate, which repressed a transcription factor that contributed to cell death under hypoxia. This research specifically advances our understanding of how tumor cells survive hypoxia and lactic acidosis and more broadly enhances our understanding of the cellular biology of solid tumors.

PhenoMeter: A Metabolome Database Search Tool Using Statistical Similarity Matching of Metabolic Phenotypes for High-Confidence Detection of Functional Links

This article describes PhenoMeter (PM), a new type of metabolomics database search that accepts metabolite response patterns as queries and searches the MetaPhen database of reference patterns for responses that are statistically significantly similar or inverse for the purposes of detecting functional links. To identify a similarity measure that would detect functional links as reliably as possible, we compared the performance of four statistics in correctly top-matching metabolic phenotypes of Arabidopsis thaliana metabolism mutants affected in different steps of the photorespiration metabolic pathway to reference phenotypes of mutants affected in the same enzymes by independent mutations. The best performing statistic, the PM score, was a function of both Pearson correlation and Fisher's Exact Test of directional overlap. This statistic outperformed Pearson correlation, biweight midcorrelation and Fisher's Exact Test used alone. To demonstrate general applicability, we show that the PM reliably retrieved the most closely functionally linked response in the database when queried with responses to a wide variety of environmental and genetic perturbations. Attempts to match metabolic phenotypes between independent studies were met with varying success and possible reasons for this are discussed. Overall, our results suggest that integration of pattern-based search tools into metabolomics databases will aid functional annotation of newly recorded metabolic phenotypes analogously to the way sequence similarity search algorithms have aided the functional annotation of genes and proteins. PM is freely available at MetabolomeExpress (https://www.metabolome-express.org/phenometer.php).

Utilization of the Eμ-Myc mouse to model heterogeneity of therapeutic response

Human aggressive B-cell non-Hodgkin lymphomas (NHL) encompass the continuum between Burkitt lymphoma and diffuse large B-cell lymphoma (DLBCL), and display considerable clinical and biologic heterogeneity, most notably related to therapy response. We previously showed that lymphomas arising in the Eμ-Myc transgenic mouse are heterogeneous, mirroring genomic differences between Burkitt lymphoma and DLBCL. Given clinical heterogeneity in NHL and the need to develop strategies to match therapeutics with discrete forms of disease, we investigated the extent to which genomic variation in the Eμ-Myc model predicts response to therapy. We used genomic analyses to classify Eμ-Myc lymphomas, link Eμ-Myc lymphomas with NHL subtypes, and identify lymphomas with predicted resistance to conventional and NF-κB-targeted therapies. Experimental evaluation of these predictions links genomic profiles with distinct outcomes to conventional and targeted therapies in the Eμ-Myc model, and establishes a framework to test novel targeted therapies or combination therapies in specific genomically defined lymphoma subgroups. In turn, this will rationally inform the design of new treatment options for aggressive human NHL.

An integrated genomics approach identifies drivers of proliferation in luminal-subtype human breast cancer

Elucidating the molecular drivers of human breast cancers requires a strategy capable of integrating multiple forms of data and an ability to interpret the functional consequences of a given genetic aberration. Here we present an integrated genomic strategy based on the use of gene expression signatures of oncogenic pathway activity (n=52) as a framework to analyze DNA copy number alterations in combination with data from a genome-wide RNAi screen. We identify specific DNA amplifications, and importantly, essential genes within these amplicons representing key genetic drivers, including known and novel regulators of oncogenesis. The genes identified include eight that are essential for cell proliferation (FGD5, METTL6, CPT1A, DTX3, MRPS23, EIF2S2, EIF6 and SLC2A10) and are uniquely amplified in patients with highly proliferative luminal breast tumors, a clinical subset of patients for which few therapeutic options are effective. Our results demonstrate that this general strategy has the potential to identify putative therapeutic targets within amplicons through an integrated use of genetic, genomic, and genome-wide RNAi data sets.

A signature of epithelial-mesenchymal plasticity and stromal activation in primary tumor modulates late recurrence in breast cancer independent of disease subtype

Introduction

Despite improvements in adjuvant therapy, late systemic recurrences remain a lethal consequence of both early- and late-stage breast cancer. A delayed recurrence is thought to arise from a state of tumor dormancy, but the mechanisms that govern tumor dormancy remain poorly understood.

Methods

To address the features of breast tumors associated with late recurrence, but not confounded by variations in systemic treatment, we compiled breast tumor gene expression data from 4,767 patients and established a discovery cohort consisting of 743 lymph node-negative patients who did not receive systemic neoadjuvant or adjuvant therapy. We interrogated the gene expression profiles of the 743 tumors and identified gene expression patterns that were associated with early and late disease recurrence among these patients. We applied this classification to a subset of 46 patients for whom expression data from microdissected tumor epithelium and stroma was available, and identified a distinct gene signature in the stroma and also a corresponding tumor epithelium signature that predicted disease recurrence in the discovery cohort. This tumor epithelium signature was then validated as a predictor for late disease recurrence in the entire cohort of 4,767 patients.

Results

We identified a novel 51-gene signature from microdissected tumor epithelium associated with late disease recurrence in breast cancer independent of the molecular disease subtype. This signature correlated with gene expression alterations in the adjacent tumor stroma and describes a process of epithelial to mesenchymal transition (EMT) and tumor-stroma interactions.

Conclusions

Our findings suggest that an EMT-related gene signature in the tumor epithelium is related to both stromal activation and escape from disease dormancy in breast cancer. The presence of a late recurrence gene signature in the primary tumor also suggests that intrinsic features of this tumor regulate the transition of disseminated tumor cells into a dormant phenotype with the ability to outgrowth as recurrent disease.

Electronic supplementary material

The online version of this article (doi:10.1186/s13058-014-0407-9) contains supplementary material, which is available to authorized users.

High CD49f expression is associated with osteosarcoma tumor progression: a study using patient-derived primary cell cultures

Overall prognosis for osteosarcoma (OS) is poor despite aggressive treatment options. Limited access to primary tumors, technical challenges in processing OS tissues, and the lack of well-characterized primary cell cultures has hindered our ability to fully understand the properties of OS tumor initiation and progression. In this study, we have isolated and characterized cell cultures derived from four central high-grade human OS samples. Furthermore, we used the cell cultures to study the role of CD49f in OS progression. Recent studies have implicated CD49f in stemness and multipotency of both cancer stem cells and mesenchymal stem cells. Therefore, we investigated the role of CD49f in osteosarcomagenesis. First, single cell suspensions of tumor biopsies were subcultured and characterized for cell surface marker expression. Next, we characterized the growth and differentiation properties, sensitivity to chemotherapy drugs, and anchorage-independent growth. Xenograft assays showed that cell populations expressing CD49f(hi) /CD90(lo) cell phenotype produced an aggressive tumor. Multiple lines of evidence demonstrated that inhibiting CD49f decreased the tumor-forming ability. Furthermore, the CD49f(hi) /CD90(lo) cell population is generating more aggressive OS tumor growth and indicating this cell surface marker could be a potential candidate for the isolation of an aggressive cell type in OSs.

Circulating IGF system and treatment outcome in epithelial ovarian cancer

Aggressive epithelial ovarian cancers (EOCs) frequently progress and become fatal, even when cytoreduction surgery plus platinum-based chemotherapy are performed. Thus, the early detection of high-risk subgroups is important in order to provide opportunities for better treatment outcomes, using alternative therapeutic strategies. This study aimed to explore the expression of circulating IGF system components and their relationship with treatment outcome in EOC. We included 228 patients with a median follow-up time of 44 months at two tertiary centers. There were 68 cancer deaths and 108 cases of cancer progression in the cohort. Preoperative serum levels of total IGF1, IGF2, IGF-binding protein 2 (IGFBP2), and IGFBP3 were analyzed using an ELISA and were then converted into an IGF1:IGFBP3 molar ratio. The risks of mortality and progression were estimated using Cox regression models in univariate and multivariate analyses. Our results showed that high IGF1, IGF2, and IGFBP3 levels were significantly associated with an early cancer stage, non-serous histology, and optimal cytoreduction. High IGFBP2 levels were associated with an advanced stage and serous histology. Overall and progression-free survival durations were significantly better among patients with high IGF1 (P=0.003 and P=0.001), IGF2 (P=0.003 and P=0.02), or IGFBP3 levels (P=0.02 and P=0.008). In multivariate analysis, serum IGFBP2 levels were significantly associated with increased risk of mortality (hazard ratio=1.84, 95% CI: 1.07-3.18, P=0.03), indicating that IGFBP2 could be used as an early predictor of EOC-related mortality. The combination of elevated IGFBP2 and reduced IGF1 levels at diagnosis could further facilitate the identification of a patient subgroup with the worst prognosis.

Cross-platform prediction of gene expression signatures

Gene expression signatures can predict the activation of oncogenic pathways and other phenotypes of interest via quantitative models that combine the expression levels of multiple genes. However, as the number of platforms to measure genome-wide gene expression proliferates, there is an increasing need to develop models that can be ported across diverse platforms. Because of the range of technologies that measure gene expression, the resulting signal values can vary greatly. To understand how this variation can affect the prediction of gene expression signatures, we have investigated the ability of gene expression signatures to predict pathway activation across Affymetrix and Illumina microarrays. We hybridized the same RNA samples to both platforms and compared the resultant gene expression readings, as well as the signature predictions. Using a new approach to map probes across platforms, we found that the genes in the signatures from the two platforms were highly similar, and that the predictions they generated were also strongly correlated. This demonstrates that our method can map probes from Affymetrix and Illumina microarrays, and that this mapping can be used to predict gene expression signatures across platforms.

Clinical and biological relevance of genomic heterogeneity in chronic lymphocytic leukemia

BACKGROUND

Chronic lymphocytic leukemia (CLL) is typically regarded as an indolent B-cell malignancy. However, there is wide variability with regards to need for therapy, time to progressive disease, and treatment response. This clinical variability is due, in part, to biological heterogeneity between individual patients' leukemias. While much has been learned about this biological variation using genomic approaches, it is unclear whether such efforts have sufficiently evaluated biological and clinical heterogeneity in CLL.

METHODS

To study the extent of genomic variability in CLL and the biological and clinical attributes of genomic classification in CLL, we evaluated 893 unique CLL samples from fifteen publicly available gene expression profiling datasets. We used unsupervised approaches to divide the data into subgroups, evaluated the biological pathways and genetic aberrations that were associated with the subgroups, and compared prognostic and clinical outcome data between the subgroups.

RESULTS

Using an unsupervised approach, we determined that approximately 600 CLL samples are needed to define the spectrum of diversity in CLL genomic expression. We identified seven genomically-defined CLL subgroups that have distinct biological properties, are associated with specific chromosomal deletions and amplifications, and have marked differences in molecular prognostic markers and clinical outcomes.

CONCLUSIONS

Our results indicate that investigations focusing on small numbers of patient samples likely provide a biased outlook on CLL biology. These findings may have important implications in identifying patients who should be treated with specific targeted therapies, which could have efficacy against CLL cells that rely on specific biological pathways.

WNT7B mediates autocrine Wnt/β-catenin signaling and anchorage-independent growth in pancreatic adenocarcinoma

Developmental and cancer models show Wnt/β-catenin-dependent signaling mediates diverse phenotypic outcomes in the pancreas that are dictated by context, duration and strength of activation. While generally assumed to be pro-tumorigenic, it is unclear to what extent dysregulation of Wnt/β-catenin signaling impacts tumor progression in pancreatic adenocarcinoma (PDAC). In the present study, Wnt/β-catenin activity was characterized across a spectrum of PDAC cell lines and primary tumors. Reporter and gene expression based assays revealed wide heterogeneity in Wnt/β-catenin transcriptional activity across PDAC cell lines and patient tumors, as well as variable responsiveness to exogenous Wnt ligand stimulation. An experimentally-generated, pancreas-specific gene expression signature of Wnt/β-catenin transcriptional activation was used to stratify pathway activation across a cohort of resected, early stage PDAC tumors (N=41). In this cohort, higher Wnt/β-catenin activation was found to significantly correlate with lymphvascular invasion and worse disease specific survival (median survival time 20.3 versus 43.9 months, log rank P=0.03). Supporting the importance of Wnt ligand in mediating autocrine Wnt signaling, Wnt/β-catenin activity was significantly inhibited in PDAC cell lines by WLS gene silencing and the small molecule inhibitor IWP-2, both of which functionally block Wnt ligand processing and secretion. Transcriptional profiling revealed elevated expression of WNT7B occurred in PDAC cell lines with high levels of cell autonomous Wnt/β-catenin activity. Gene knockdown studies in AsPC-1 and HPAF-2 cell lines confirmed WNT7B mediated cell autonomous Wnt/β-catenin activation, as well as an anchorage-independent growth phenotype. Our findings indicate WNT7B can serve as a primary determinant of differential Wnt/β-catenin activation in PDAC. Disrupting the interaction between Wnt ligands and their receptors may be a particularly suitable approach for therapeutic modulation of Wnt/β-catenin signaling in PDAC and other cancer contexts where Wnt activation is mediated by ligand expression rather than mutations in canonical pathway members.

FOXC2 expression links epithelial-mesenchymal transition and stem cell properties in breast cancer

Resistance to chemotherapy and metastases are the major causes of breast cancer-related mortality. Moreover, cancer stem cells (CSC) play critical roles in cancer progression and treatment resistance. Previously, it was found that CSC-like cells can be generated by aberrant activation of epithelial-mesenchymal transition (EMT), thereby making anti-EMT strategies a novel therapeutic option for treatment of aggressive breast cancers. Here, we report that the transcription factor FOXC2 induced in response to multiple EMT signaling pathways as well as elevated in stem cell-enriched factions is a critical determinant of mesenchymal and stem cell properties, in cells induced to undergo EMT- and CSC-enriched breast cancer cell lines. More specifically, attenuation of FOXC2 expression using lentiviral short hairpin RNA led to inhibition of the mesenchymal phenotype and associated invasive and stem cell properties, which included reduced mammosphere-forming ability and tumor initiation. Whereas, overexpression of FOXC2 was sufficient to induce CSC properties and spontaneous metastasis in transformed human mammary epithelial cells. Furthermore, a FOXC2-induced gene expression signature was enriched in the claudin-low/basal B breast tumor subtype that contains EMT and CSC features. Having identified PDGFR-β to be regulated by FOXC2, we show that the U.S. Food and Drug Administration-approved PDGFR inhibitor, sunitinib, targets FOXC2-expressing tumor cells leading to reduced CSC and metastatic properties. Thus, FOXC2 or its associated gene expression program may provide an effective target for anti-EMT-based therapies for the treatment of claudin-low/basal B breast tumors or other EMT-/CSC-enriched tumors.