Comprehensive transcriptomic analysis of cell lines as models of primary tumors across 22 tumor types

Serum Mac2-binding protein glycosylated isomer (M2BPGi) as a prognostic biomarker in pancreatic ductal adenocarcinoma: iCAFs-derived M2BPGi drives tumor invasion

BACKGROUND

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive malignancy with a poor prognosis. Mac2-binding protein glycosylated isomer (M2BPGi), a known biomarker for liver fibrosis, is also elevated in other fibrotic tissues. However, its role in PDAC remains unexplored. This study investigates the potential of M2BPGi as a prognostic biomarker for PDAC and elucidates its role in cancer progression.

METHODS

We analyzed serum M2BPGi levels in 83 PDAC patients and 60 healthy controls, examining the relationship with clinical outcomes. Tissue immunostaining and in vitro experiments were conducted to investigate M2BPGi-secreting cells and its role.

RESULTS

Serum M2BPGi levels were significantly higher in PDAC patients than in controls (0.98 vs. 0.59, p < 0.0001). Notably, elevated serum M2BPGi was associated with worse progression-free survival (144 days vs. 260 days, p = 0.017) and overall survival (OS) (245 days vs. 541 days, p < 0.001) following chemotherapy. Multivariable Cox regression analysis further confirmed that a high serum M2BPGi level is an independent risk factor for OS (HR: 2.44, 95% CI 1.26-4.74, p = 0.008). Immunostaining revealed that M2BPGi is secreted by both cancer cells and cancer-associated fibroblasts (CAFs), with high M2BP expression in CAFs correlating with poor prognosis. Furthermore, M2BPGi-secreting CAFs exhibited characteristics of inflammatory CAFs. M2BPGi directly activated mTOR signaling and epithelial-mesenchymal transition in PDAC cells, enhancing their invasive and migratory capabilities.

CONCLUSIONS

Our findings identify M2BPGi as a promising prognostic biomarker for PDAC. Moreover, we demonstrate that inflammatory CAFs promote tumor invasion and contribute to poor outcomes by secreting M2BPGi, revealing a novel mechanism of PDAC progression.

Identification of a Selective Anticancer Agent from a Collection of Complex-And-Diverse Compounds Synthesized from Stevioside

Compounds constructed by distorting the ring systems of natural products serve as a ready source of complex and diverse molecules, useful for a variety of applications. Herein is presented the use of the diterpenoids steviol and isosteviol as starting points for the construction of >50 new compounds through this complexity-to-diversity approach, featuring novel ring system distortions and a noteworthy thallium(III) nitrate (TTN)-mediated ring fusion. Evaluation of this collection identified SteviX4 as a potent and selective anticancer compound, inducing cell death at low nanomolar concentrations against some cancer cell lines in culture, compared to micromolar activity against others. SteviX4 induces ferroptotic cell death in susceptible cell lines, and target identification experiments reveal SteviX4 acts as an inhibitor of glutathione peroxidase 4 (GPX4), a critical protein that protects cancer cells against ferroptosis. In its induction of cell death, SteviX4 displays enhanced cell line selectivity relative to most known GPX4 inhibitors. SteviX4 was used to reveal dependency on GPX4 as a vulnerability of certain cancer cell lines, not tied to any one type of cancer, suggesting GPX4 inhibition as a cancer type-agnostic anticancer strategy. With its high fraction of sp3-hybridized carbons and considerable cell line selectivity and potency, SteviX4 is unique among GPX4 inhibitors, serving as an outstanding probe compound and basis for further translational development.

A hybrid machine learning framework for functional annotation of mitochondrial glutathione transport and metabolism proteins in cancers

BACKGROUND

Alterations of metabolism, including changes in mitochondrial metabolism as well as glutathione (GSH) metabolism are a well appreciated hallmark of many cancers. Mitochondrial GSH (mGSH) transport is a poorly characterized aspect of GSH metabolism, which we investigate in the context of cancer. Existing functional annotation approaches from machine (ML) or deep learning (DL) models based only on protein sequences, were unable to annotate functions in biological contexts.

RESULTS

We develop a flexible ML framework for functional annotation from diverse feature data. This hybrid ML framework leverages cancer cell line multi-omics data and other biological knowledge data as features, to uncover potential genes involved in mGSH metabolism and membrane transport in cancers. This framework achieves strong performance across functional annotation tasks and several cell line and primary tumor cancer samples. For our application, classification models predict the known mGSH transporter SLC25A39 but not SLC25A40 as being highly probably related to mGSH metabolism in cancers. SLC25A10, SLC25A50, and orphan SLC25A24, SLC25A43 are predicted to be associated with mGSH metabolism in multiple biological contexts and structural analysis of these proteins reveal similarities in potential substrate binding regions to the binding residues of SLC25A39.

CONCLUSION

These findings have implications for a better understanding of cancer cell metabolism and novel therapeutic targets with respect to GSH metabolism through potential novel functional annotations of genes. The hybrid ML framework proposed here can be applied to other biological function classifications or multi-omics datasets to generate hypotheses in various biological contexts. Code and a tutorial for generating models and predictions in this framework are available at: https://github.com/lkenn012/mGSH_cancerClassifiers .

Squamous cell cancers of the aero-upper digestive tract: A unified perspective on biology, genetics, and therapy

Squamous cell cancers (SCCs) of the head and neck, esophagus, and lung, referred to as aero-upper digestive SCCs, are prevalent in the United States and worldwide. Their incidence and mortality are projected to increase at alarming rates, posing diagnostic, prognostic, and therapeutic challenges. These SCCs share certain epigenetic, genomic, and genetic alterations, immunologic properties, environmental exposures, as well as lifestyle and nutritional risk factors, which may underscore common complex gene-environmental interactions across them. This review focuses upon the frequent shared epigenetic, genomic, and genetic alterations, emerging preclinical model systems, and how this collective knowledge can be leveraged into perspectives on standard of care therapies and mechanisms of resistance, nominating new potential directions in translational therapeutics.

Integrative analysis of miRNA expression data reveals a minimal signature for tumour cells classification

MicroRNAs (miRNAs) are pivotal biomarkers for cancer screening. Identifying distinctive expression patterns of miRNAs in specific cancer types can serve as an effective strategy for classification and characterization. However, the development of a minimal signature of miRNAs for accurate cancer classification remains challenging, hindered by the lack of integrated approaches that systematically analyse miRNA expression levels of miRNAs alongside their associated biological pathways. In this study, we present a comprehensive integrative approach that utilizes transcriptomic data from lung, breast, and melanoma cancer cell lines to identify specific expression patterns. By combining bioinformatics, dimensionality reduction techniques, machine learning, and experimental validation, we pinpoint miRNAs linked to critical biological pathways. Our results demonstrate a highly significant differentiation of cancer types, achieving 100 % classification accuracy with minimal training time using a streamlined miRNA signature. Validation of the miRNA profile confirms that each of the three identified miRNAs regulates distinct biological pathways with minimal overlap. This specificity highlights their unique roles in tumour biology and set the stage for further exploration of miRNAs interactions and their contributions to tumourigenesis across diverse cancer types. Our work paves the way for multi-cancer classification, emphasizing the transformative potential of miRNA research in oncology. Beyond advancing the understanding of tumour biology, our step-by-step guide offers a robust tool for a wide range of users to investigate precise diagnostics and promising therapeutic strategies.

Discovery of therapeutic targets in cancer using chromatin accessibility and transcriptomic data

Most cancer types lack targeted therapeutic options, and when first-line targeted therapies are available, treatment resistance is a huge challenge. Recent technological advances enable the use of assay for transposase-accessible chromatin with sequencing (ATAC-seq) and RNA sequencing (RNA-seq) on patient tissue in a high-throughput manner. Here, we present a computational approach that leverages these datasets to identify drug targets based on tumor lineage. We constructed gene regulatory networks for 371 patients of 22 cancer types using machine learning approaches trained with three-dimensional genomic data for enhancer-to-promoter contacts. Next, we identified the key transcription factors (TFs) in these networks, which are used to find therapeutic vulnerabilities, by direct targeting of either TFs or the proteins that they interact with. We validated four candidates identified for neuroendocrine, liver, and renal cancers, which have a dismal prognosis with current therapeutic options.

Evaluating the Effects of Perinatal Exposures to BPSIP on Hepatic Cholesterol Metabolism in Female and Male Offspring ICR Mice

BACKGROUND

A broad suite of bisphenol S (BPS) derivatives as alternatives for BPS have been identified in various human biological samples, including 4-hydroxyphenyl 4-isopropoxyphenylsulfone (BPSIP) detected in human umbilical cord plasma and breast milk. However, very little is known about the health outcomes of prenatal BPS derivative exposure to offspring.

OBJECTIVES

Our study aimed to investigate the response of hepatic cholesterol metabolism by sex in offspring of dams exposed to BPSIP.

METHODS

Pregnant ICR mice were exposed to 5 μ g / kg body weight (BW)/day of BPSIP, BPS, or E2 through drinking water from gestational day one until the pups were weaned. The concentration of BPSIP, BPS, or E2 in the plasma and liver of pups was determined by liquid chromatography-tandem mass spectrometry. Metabolic phenotypes were recorded, and histopathology was examined for liver impairment. Transcriptome analysis was employed to characterize the distribution and expression patterns of differentially expressed genes across sexes. The metabolic regulation was validated by quantitative real-time PCR, immunohistochemistry, and immunoblotting. The role of estrogen receptors (ERs) in mediating sex-dependent effects was investigated using animal models and liver organoids.

RESULTS

Pups of dams exposed to BPSIP showed a higher serum cholesterol level, and liver cholesterol levels were higher in females and lower in males than in the controls. BPSIP concentration in the male liver was 1.22 ± 0.25  ng / g and 0.69 ± 0.27  ng / g in the female liver. Histopathology analysis showed steatosis and lipid deposition in both male and female offspring. Transcriptome and gene expression analyses identified sex-specific differences in cholesterol biosynthesis, absorption, disposal, and efflux between pups of dams exposed to BPSIP and those in controls. In vivo, chromatin immunoprecipitation analysis revealed that the binding of ER α protein to key genes such as Hmgcr, Pcsk9, and Abcg5 was attenuated in BPSIP-exposed females compared to controls, while it was enhanced in males. In vitro, the liver organoid experiments demonstrated that restoration of differential expression induced by BPSIP in key genes, such as Hmgcr, Ldlr, and Cyp7a1, to levels comparable to the controls was only achieved when treated with a combination of ER α agonist and ER β agonist.

DISCUSSION

Findings from this study suggest that perinatal exposure to BPSIP disrupted cholesterol metabolism in a sex-specific manner in a mouse model, in which ER α played a crucial role both in vivo and in vitro. Therefore, it is crucial to systematically evaluate BPS derivatives to protect maternal health during pregnancy and prevent the transmission of metabolic disorders across generations. https://doi.org/10.1289/EHP14643.

TransCell: In Silico Characterization of Genomic Landscape and Cellular Responses by Deep Transfer Learning

Gene expression profiling of new or modified cell lines becomes routine today; however, obtaining comprehensive molecular characterization and cellular responses for a variety of cell lines, including those derived from underrepresented groups, is not trivial when resources are minimal. Using gene expression to predict other measurements has been actively explored; however, systematic investigation of its predictive power in various measurements has not been well studied. Here, we evaluated commonly used machine learning methods and presented TransCell, a two-step deep transfer learning framework that utilized the knowledge derived from pan-cancer tumor samples to predict molecular features and responses. Among these models, TransCell had the best performance in predicting metabolite, gene effect score (or genetic dependency), and drug sensitivity, and had comparable performance in predicting mutation, copy number variation, and protein expression. Notably, TransCell improved the performance by over 50% in drug sensitivity prediction and achieved a correlation of 0.7 in gene effect score prediction. Furthermore, predicted drug sensitivities revealed potential repurposing candidates for new 100 pediatric cancer cell lines, and predicted gene effect scores reflected BRAF resistance in melanoma cell lines. Together, we investigated the predictive power of gene expression in six molecular measurement types and developed a web portal (http://apps.octad.org/transcell/) that enables the prediction of 352,000 genomic and cellular response features solely from gene expression profiles.

Machine learning identifies activation of RUNX/AP-1 as drivers of mesenchymal and fibrotic regulatory programs in gastric cancer

Gastric cancer (GC) is the fifth most common cancer worldwide and is a heterogeneous disease. Among GC subtypes, the mesenchymal phenotype (Mes-like) is more invasive than the epithelial phenotype (Epi-like). Although gene expression of the epithelial-to-mesenchymal transition (EMT) has been studied, the regulatory landscape shaping this process is not fully understood. Here we use ATAC-seq and RNA-seq data from a compendium of GC cell lines and primary tumors to detect drivers of regulatory state changes and their transcriptional responses. Using the ATAC-seq data, we developed a machine learning approach to determine the transcription factors (TFs) regulating the subtypes of GC. We identified TFs driving the mesenchymal (RUNX2, ZEB1, SNAI2, AP-1 dimer) and the epithelial (GATA4, GATA6, KLF5, HNF4A, FOXA2, GRHL2) states in GC. We identified DNA copy number alterations associated with dysregulation of these TFs, specifically deletion of GATA4 and amplification of MAPK9 Comparisons with bulk and single-cell RNA-seq data sets identified activation toward fibroblast-like epigenomic and expression signatures in Mes-like GC. The activation of this mesenchymal fibrotic program is associated with differentially accessible DNA cis-regulatory elements flanking upregulated mesenchymal genes. These findings establish a map of TF activity in GC and highlight the role of copy number driven alterations in shaping epigenomic regulatory programs as potential drivers of GC heterogeneity and progression.

Decoding polygenic diseases: advances in noncoding variant prioritization and validation

Genome-wide association studies (GWASs) provide a key foundation for elucidating the genetic underpinnings of common polygenic diseases. However, these studies have limitations in their ability to assign causality to particular genetic variants, especially those residing in the noncoding genome. Over the past decade, technological and methodological advances in both analytical and empirical prioritization of noncoding variants have enabled the identification of causative variants by leveraging orthogonal functional evidence at increasing scale. In this review, we present an overview of these approaches and describe how this workflow provides the groundwork necessary to move beyond associations toward genetically informed studies on the molecular and cellular mechanisms of polygenic disease.

The transcriptional landscape of glycosylation-related genes in cancer

Changes in glycosylation patterns have been associated with malignant transformation and clinical outcomes in several cancer types, prompting ongoing research into the mechanisms involved and potential clinical applications. In this study, we performed an extensive transcriptomic analysis of glycosylation-related genes and pathways, using publicly available bulk and single cell transcriptomic datasets from tumor samples and cancer cell lines. We identified genes and pathways strongly associated with different tumor types, which may represent novel diagnostic biomarkers. By using single cell RNA-seq data, we characterized the contribution of different cell types to the overall tumor glycosylation. Transcriptomic analysis of cancer cell lines revealed that they present a simplified landscape of genes compared to tissue. Lastly, we describe the association of different genes and pathways with the clinical outcome of patients. These results can serve as a resource for future research aimed to unravel the role of the glyco-code in cancer.

Osteosarcoma PDX-Derived Cell Line Models for Preclinical Drug Evaluation Demonstrate Metastasis Inhibition by Dinaciclib through a Genome-Targeted Approach

PURPOSE

Models to study metastatic disease in rare cancers are needed to advance preclinical therapeutics and to gain insight into disease biology. Osteosarcoma is a rare cancer with a complex genomic landscape in which outcomes for patients with metastatic disease are poor. As osteosarcoma genomes are highly heterogeneous, multiple models are needed to fully elucidate key aspects of disease biology and to recapitulate clinically relevant phenotypes.

EXPERIMENTAL DESIGN

Matched patient samples, patient-derived xenografts (PDX), and PDX-derived cell lines were comprehensively evaluated using whole-genome sequencing and RNA sequencing. The in vivo metastatic phenotype of the PDX-derived cell lines was characterized in both an intravenous and an orthotopic murine model. As a proof-of-concept study, we tested the preclinical effectiveness of a cyclin-dependent kinase inhibitor on the growth of metastatic tumors in an orthotopic amputation model.

RESULTS

PDXs and PDX-derived cell lines largely maintained the expression profiles of the patient from which they were derived despite the emergence of whole-genome duplication in a subset of cell lines. The cell lines were heterogeneous in their metastatic capacity, and heterogeneous tissue tropism was observed in both intravenous and orthotopic models. Single-agent dinaciclib was effective at dramatically reducing the metastatic burden.

CONCLUSIONS

The variation in metastasis predilection sites between osteosarcoma PDX-derived cell lines demonstrates their ability to recapitulate the spectrum of the disease observed in patients. We describe here a panel of new osteosarcoma PDX-derived cell lines that we believe will be of wide use to the osteosarcoma research community.

Evaluation of single-sample network inference methods for precision oncology

A major challenge in precision oncology is to detect targetable cancer vulnerabilities in individual patients. Modeling high-throughput omics data in biological networks allows identifying key molecules and processes of tumorigenesis. Traditionally, network inference methods rely on many samples to contain sufficient information for learning, resulting in aggregate networks. However, to implement patient-tailored approaches in precision oncology, we need to interpret omics data at the level of individual patients. Several single-sample network inference methods have been developed that infer biological networks for an individual sample from bulk RNA-seq data. However, only a limited comparison of these methods has been made and many methods rely on 'normal tissue' samples as reference, which are not always available. Here, we conducted an evaluation of the single-sample network inference methods SSN, LIONESS, SWEET, iENA, CSN and SSPGI using transcriptomic profiles of lung and brain cancer cell lines from the CCLE database. The methods constructed functional gene networks with distinct network characteristics. Hub gene analyses revealed different degrees of subtype-specificity across methods. Single-sample networks were able to distinguish between tumor subtypes, as exemplified by node strength clustering, enrichment of known subtype-specific driver genes among hubs and differential node strength. We also showed that single-sample networks correlated better to other omics data from the same cell line as compared to aggregate networks. We conclude that single-sample network inference methods can reflect sample-specific biology when 'normal tissue' samples are absent and we point out peculiarities of each method.

TTD: Therapeutic Target Database describing target druggability information

Target discovery is one of the essential steps in modern drug development, and the identification of promising targets is fundamental for developing first-in-class drug. A variety of methods have emerged for target assessment based on druggability analysis, which refers to the likelihood of a target being effectively modulated by drug-like agents. In the therapeutic target database (TTD), nine categories of established druggability characteristics were thus collected for 426 successful, 1014 clinical trial, 212 preclinical/patented, and 1479 literature-reported targets via systematic review. These characteristic categories were classified into three distinct perspectives: molecular interaction/regulation, human system profile and cell-based expression variation. With the rapid progression of technology and concerted effort in drug discovery, TTD and other databases were highly expected to facilitate the explorations of druggability characteristics for the discovery and validation of innovative drug target. TTD is now freely accessible at: https://idrblab.org/ttd/.

SERPINB3-MYC axis induces the basal-like/squamous subtype and enhances disease progression in pancreatic cancer

Pancreatic ductal adenocarcinoma (PDAC) exhibits distinct molecular subtypes: classical/progenitor and basal-like/squamous. Our study aimed to identify genes contributing to the development of the basal-like/squamous subtype, known for its aggressiveness. Transcriptome analyses revealed consistent upregulation of SERPINB3 in basal-like/squamous PDAC, correlating with reduced patient survival. SERPINB3 transgene expression in PDAC cells enhanced in vitro invasion and promoted lung metastasis in a mouse PDAC xenograft model. Metabolome analyses unveiled a metabolic signature linked to both SERPINB3 and the basal-like/squamous subtype, characterized by heightened carnitine/acylcarnitine and amino acid metabolism, associated with poor prognosis in patients with PDAC and elevated cellular invasiveness. Further analysis uncovered that SERPINB3 inhibited the cysteine protease calpain, a key enzyme in the MYC degradation pathway, and drove basal-like/squamous subtype and associated metabolic reprogramming through MYC activation. Our findings indicate that the SERPINB3-MYC axis induces the basal-like/squamous subtype, proposing SERPINB3 as a potential diagnostic and therapeutic target for this variant.

Comparative analysis identifies genetic and molecular factors associated with prognostic clusters of PANoptosis in glioma, kidney and melanoma cancer

The importance of inflammatory cell death, PANoptosis, in cancer is increasingly being recognized. PANoptosis can promote or inhibit tumorigenesis in context-dependent manners, and a computational approach leveraging transcriptomic profiling of genes involved in PANoptosis has shown that patients can be stratified into PANoptosis High and PANoptosis Low clusters that have significant differences in overall survival for low grade glioma (LGG), kidney renal cell carcinoma (KIRC) and skin cutaneous melanoma (SKCM). However, the molecular mechanisms that contribute to differential prognosis between PANoptosis clusters require further elucidation. Therefore, we performed a comprehensive comparison of genetic, genomic, tumor microenvironment, and pathway characteristics between the PANoptosis High and PANoptosis Low clusters to determine the relevance of each component in driving the differential associations with prognosis for LGG, KIRC and SKCM. Across these cancer types, we found that activation of the proliferation pathway was significantly different between PANoptosis High and Low clusters. In LGG and SKCM, we also found that aneuploidy and immune cell densities and activations contributed to differences in PANoptosis clusters. In individual cancers, we identified important roles for barrier gene pathway activation (in SKCM) and the somatic mutation profiles of driver oncogenes as well as hedgehog signaling pathway activation (in LGG). By identifying these genetic and molecular factors, we can possibly improve the prognosis for at risk-stratified patient populations based on the PANoptosis phenotype in LGG, KIRC and SKCM. This not only advances our mechanistic understanding of cancer but will allow for the selection of optimal treatment strategies.

CCR9 overexpression promotes T-ALL progression by enhancing cholesterol biosynthesis

Introduction: T-cell acute lymphoblastic leukemia (T-ALL) is an aggressive hematological malignancy of the lymphoid progenitor cells, contributing to ∼ 20% of the total ALL cases, with a higher prevalence in adults than children. Despite the important role of human T-ALL cell lines in understanding the pathobiology of the disease, a detailed comparison of the tumorigenic potentials of two commonly used T-ALL cell lines, MOLT4 and JURKAT cells, is still lacking. Methodology: In the present study, NOD-Prkdc scid IL2rgd ull (NTG) mice were intravenously injected with MOLT4, JURKAT cells, and PBS as a control. The leukemiac cell homing/infiltration into the bone marrow, blood, liver and spleen was investigated for bioluminescence imaging, flow cytometry, and immunohistochemistry staining. Gene expression profiling of the two cell lines was performed via RNA-seq to identify the differentially expressed genes (DEGs). CCR9 identified as a DEG, was further screened for its role in invasion and metastasis in both cell lines in vitro. Moreover, a JURKAT cell line with overexpressed CCR9 (Jurkat-OeCCR9) was investigated for T-ALL formation in the NTG mice as compared to the GFP control. Jurkat-OeCCR9 cells were then subjected to transcriptome analysis to identify the genes and pathways associated with the upregulation of CCR9 leading to enhanced tumirogenesis. The DEGs of the CCR9-associated upregulation were validated both at mRNA and protein levels. Simvastatin was used to assess the effect of cholesterol biosynthesis inhibition on the aggressiveness of T-ALL cells. Results: Comparison of the leukemogenic potentials of the two T-ALL cell lines showed the relatively higher leukemogenic potential of MOLT4 cells, characterized by their enhanced tissue infiltration in NOD-PrkdcscidIL2rgdull (NTG) mice. Transcriptmoe analysis of the two cell lines revealed numerous DEGs, including CCR9, enriched in vital signaling pathways associated with growth and proliferation. Notably, the upregulation of CCR9 also promoted the tissue infiltration of JURKAT cells in vitro and in NTG mice. Transcriptome analysis revealed that CCR9 overexpression facilitated cholesterol production by upregulating the expression of the transcriptional factor SREBF2, and the downstream genes: MSMO1, MVD, HMGCS1, and HMGCR, which was then corroborated at the protein levels. Notably, simvastatin treatment reduced the migration of the CCR9-overexpressing JURKAT cells, suggesting the importance of cholesterol in T-ALL progression. Conclusions: This study highlights the distinct tumorigenic potentials of two T-ALL cell lines and reveals CCR9-regulated enhanced cholesterol biosynthesis in T-ALL.

Systematic transcriptional analysis of human cell lines for gene expression landscape and tumor representation

Cell lines are valuable resources as model for human biology and translational medicine. It is thus important to explore the concordance between the expression in various cell lines vis-à-vis human native and disease tissues. In this study, we investigate the expression of all human protein-coding genes in more than 1,000 human cell lines representing 27 cancer types by a genome-wide transcriptomics analysis. The cell line gene expression is compared with the corresponding profiles in various tissues, organs, single-cell types and cancers. Here, we present the expression for each cell line and give guidance for the most appropriate cell line for a given experimental study. In addition, we explore the cancer-related pathway and cytokine activity of the cell lines to aid human biology studies and drug development projects. All data are presented in an open access cell line section of the Human Protein Atlas to facilitate the exploration of all human protein-coding genes across these cell lines.

Evaluating cancer cell line and patient-derived xenograft recapitulation of tumor and non-diseased tissue gene expression profiles in silico

Preclinical models like cancer cell lines and patient-derived xenografts (PDXs) are vital for studying disease mechanisms and evaluating treatment options. It is essential that they accurately recapitulate the disease state of interest to generate results that will translate in the clinic. Prior studies have demonstrated that preclinical models do not recapitulate all biological aspects of human tissues, particularly with respect to the tissue of origin gene expression signatures. Therefore, it is critical to assess how well preclinical model gene expression profiles correlate with human cancer tissues to inform preclinical model selection and data analysis decisions. Here we evaluated how well preclinical models recapitulate human cancer and non-diseased tissue gene expression patterns in silico with respect to the full gene expression profile as well as subsetting by the most variable genes, genes significantly correlated with tumor purity, and tissue-specific genes by using publicly available gene expression profiles across multiple sources. We found that using the full gene set improves correlations between preclinical model and tissue global gene expression profiles, confirmed that GBM PDX global gene expression correlation to GBM tumor global gene expression outperforms GBM cell line to GBM tumor global gene expression correlations, and demonstrated that preclinical models in our study often failed to reproduce tissue-specific expression. While including additional genes for global gene expression comparison between cell lines and tissues decreases the overall correlation, it improves the relative rank between a cell line and its tissue of origin compared to other tissues. Our findings underscore the importance of using the full gene expression set measured when comparing preclinical models and tissues and confirm that tissue-specific patterns are better preserved in GBM PDX models than in GBM cell lines. Future studies can build on these findings to determine the specific pathways and gene sets recapitulated by particular preclinical models to facilitate model selection for a given study design or goal.

ITGA3 acts as a purity-independent biomarker of both immunotherapy and chemotherapy resistance in pancreatic cancer: bioinformatics and experimental analysis

Contribution of integrin superfamily genes to treatment resistance remains uncertain. Genome patterns of thirty integrin superfamily genes were analyzed of using bulk and single-cell RNA sequencing, mutation, copy number, methylation, clinical information, immune cell infiltration, and drug sensitivity data. To select the integrins that are most strongly associated with treatment resistance in pancreatic cancer, a purity-independent RNA regulation network including integrins were constructed using machine learning. The integrin superfamily genes exhibit extensive dysregulated expression, genome alterations, epigenetic modifications, immune cell infiltration, and drug sensitivity, as evidenced by multi-omics data. However, their heterogeneity varies among different cancers. After constructing a three-gene (TMEM80, EIF4EBP1, and ITGA3) purity-independent Cox regression model using machine learning, ITGA3 was identified as a critical integrin subunit gene in pancreatic cancer. ITGA3 is involved in the molecular transformation from the classical to the basal subtype in pancreatic cancer. Elevated ITGA3 expression correlated with a malignant phenotype characterized by higher PD-L1 expression and reduced CD8+ T cell infiltration, resulting in unfavorable outcomes in patients receiving either chemotherapy or immunotherapy. Our findings suggest that ITGA3 is an important integrin in pancreatic cancer, contributing to chemotherapy resistance and immune checkpoint blockade therapy resistance.

Elucidating the RNA-Protein Interactomes of Target RNAs in Tissue

RNA-protein interactions are key to many aspects of cellular homeostasis and their identification is important to understanding cellular function. Multiple strategies have been developed for the RNA-centric characterization of RNA-protein complexes. However, these studies have all been done in immortalized cell lines that do not capture the complexity of heterogeneous tissue samples. Here, we develop hybridization purification of RNA-protein complexes followed by mass spectrometry (HyPR-MS) for use in tissue samples. We isolated both polyadenylated RNA and the specific long noncoding RNA MALAT1 and characterized their protein interactomes. These results demonstrate the feasibility of HyPR-MS in tissue for the multiplexed characterization of specific RNA-protein complexes.

A systematic evaluation of deep learning methods for the prediction of drug synergy in cancer

One of the main obstacles to the successful treatment of cancer is the phenomenon of drug resistance. A common strategy to overcome resistance is the use of combination therapies. However, the space of possibilities is huge and efficient search strategies are required. Machine Learning (ML) can be a useful tool for the discovery of novel, clinically relevant anti-cancer drug combinations. In particular, deep learning (DL) has become a popular choice for modeling drug combination effects. Here, we set out to examine the impact of different methodological choices on the performance of multimodal DL-based drug synergy prediction methods, including the use of different input data types, preprocessing steps and model architectures. Focusing on the NCI ALMANAC dataset, we found that feature selection based on prior biological knowledge has a positive impact-limiting gene expression data to cancer or drug response-specific genes improved performance. Drug features appeared to be more predictive of drug response, with a 41% increase in coefficient of determination (R2) and 26% increase in Spearman correlation relative to a baseline model that used only cell line and drug identifiers. Molecular fingerprint-based drug representations performed slightly better than learned representations-ECFP4 fingerprints increased R2 by 5.3% and Spearman correlation by 2.8% w.r.t the best learned representations. In general, fully connected feature-encoding subnetworks outperformed other architectures. DL outperformed other ML methods by more than 35% (R2) and 14% (Spearman). Additionally, an ensemble combining the top DL and ML models improved performance by about 6.5% (R2) and 4% (Spearman). Using a state-of-the-art interpretability method, we showed that DL models can learn to associate drug and cell line features with drug response in a biologically meaningful way. The strategies explored in this study will help to improve the development of computational methods for the rational design of effective drug combinations for cancer therapy.

YBX1 integration of oncogenic PI3K/mTOR signalling regulates the fitness of malignant epithelial cells

In heterogeneous head and neck cancer (HNC), subtype-specific treatment regimens are currently missing. An integrated analysis of patient HNC subtypes using single-cell sequencing and proteome profiles reveals an epithelial-mesenchymal transition (EMT) signature within the epithelial cancer-cell population. The EMT signature coincides with PI3K/mTOR inactivation in the mesenchymal subtype. Conversely, the signature is suppressed in epithelial cells of the basal subtype which exhibits hyperactive PI3K/mTOR signalling. We further identify YBX1 phosphorylation, downstream of the PI3K/mTOR pathway, restraining basal-like cancer cell proliferation. In contrast, YBX1 acts as a safeguard against the proliferation-to-invasion switch in mesenchymal-like epithelial cancer cells, and its loss accentuates partial-EMT and in vivo invasion. Interestingly, phospho-YBX1 that is mutually exclusive to partial-EMT, emerges as a prognostic marker for overall patient outcomes. These findings create a unique opportunity to sensitise mesenchymal cancer cells to PI3K/mTOR inhibitors by shifting them towards a basal-like subtype as a promising therapeutic approach against HNC.

Molecular Signature of Long Non-Coding RNA Associated with Areca Nut-Induced Head and Neck Cancer

The areca nut is a high-risk carcinogen for head and neck cancer (HNC) patients in Southeast Asia. The underlying molecular mechanism of areca nut-induced HNC remains unclear, especially regarding the role of long non-coding RNA (lncRNA). This study employed a systemic strategy to identify lncRNA signatures related to areca nut-induced HNC. In total, 84 cancer-related lncRNAs were identified. Using a PCR array method, 28 lncRNAs were identified as being dysregulated in HNC cells treated with areca nut (17 upregulated and 11 downregulated). Using bioinformatics analysis of The Cancer Genome Atlas Head-Neck Squamous Cell Carcinoma (TCGA-HNSC) dataset, 45 lncRNAs were differentially expressed in tumor tissues from HNC patients (39 over- and 6 under-expressions). The integrated evaluation showed 10 lncRNAs dysregulated by the areca nut and altered expression in patients, suggesting that these panel molecules participate in areca nut-induced HNC. Five oncogenic (LUCAT1, MIR31HG, UCA1, HIF1A-AS2, and SUMO1P3) and tumor-suppressive (LINC00312) lncRNAs were independently validated, and three key molecules were further examined. Pathway prediction revealed that LUCAT1, UCA1, and MIR31HG modulate multiple oncogenic mechanisms, including stress response and cellular motility. Clinical assessment showed that these lncRNAs exhibited biomarker potentials in diagnosis (area under the curve = 0.815 for LUCAT1) and a worse prognosis (both p < 0.05, survival analysis). Cellular studies further demonstrated that MIR31HG facilitates areca nut-induced cancer progression, as silencing this molecule attenuated arecoline-induced invasion ability in HNC cells. This study identified lncRNA signatures that play a role in areca nut-induced HNC. These molecules may be further applied in risk assessment, diagnosis, prognosis, and therapeutics for areca nut-associated malignancies.

Investigating the suitability of in vitro cell lines as models for the major subtypes of epithelial ovarian cancer

Epithelial ovarian cancer (EOC) is the most fatal gynaecological malignancy, accounting for over 200,000 deaths worldwide per year. EOC is a highly heterogeneous disease, classified into five major histological subtypes-high-grade serous (HGSOC), clear cell (CCOC), endometrioid (ENOC), mucinous (MOC) and low-grade serous (LGSOC) ovarian carcinomas. Classification of EOCs is clinically beneficial, as the various subtypes respond differently to chemotherapy and have distinct prognoses. Cell lines are often used as in vitro models for cancer, allowing researchers to explore pathophysiology in a relatively cheap and easy to manipulate system. However, most studies that make use of EOC cell lines fail to recognize the importance of subtype. Furthermore, the similarity of cell lines to their cognate primary tumors is often ignored. Identification of cell lines with high molecular similarity to primary tumors is needed in order to better guide pre-clinical EOC research and to improve development of targeted therapeutics and diagnostics for each distinctive subtype. This study aims to generate a reference dataset of cell lines representative of the major EOC subtypes. We found that non-negative matrix factorization (NMF) optimally clustered fifty-six cell lines into five groups, putatively corresponding to each of the five EOC subtypes. These clusters validated previous histological groupings, while also classifying other previously unannotated cell lines. We analysed the mutational and copy number landscapes of these lines to investigate whether they harboured the characteristic genomic alterations of each subtype. Finally we compared the gene expression profiles of cell lines with 93 primary tumor samples stratified by subtype, to identify lines with the highest molecular similarity to HGSOC, CCOC, ENOC, and MOC. In summary, we examined the molecular features of both EOC cell lines and primary tumors of multiple subtypes. We recommend a reference set of cell lines most suited to represent four different subtypes of EOC for both in silico and in vitro studies. We also identify lines displaying poor overall molecular similarity to EOC tumors, which we argue should be avoided in pre-clinical studies. Ultimately, our work emphasizes the importance of choosing suitable cell line models to maximise clinical relevance of experiments.

Development and characterization of new patient-derived xenograft (PDX) models of osteosarcoma with distinct metastatic capacities

Models to study metastatic disease in rare cancers are needed to advance preclinical therapeutics and to gain insight into disease biology, especially for highly aggressive cancers with a propensity for metastatic spread. Osteosarcoma is a rare cancer with a complex genomic landscape in which outcomes for patients with metastatic disease are poor. As osteosarcoma genomes are highly heterogeneous, a large panel of models is needed to fully elucidate key aspects of disease biology and to recapitulate clinically-relevant phenotypes. We describe the development and characterization of osteosarcoma patient-derived xenografts (PDXs) and a panel of PDX-derived cell lines. Matched patient samples, PDXs, and PDX-derived cell lines were comprehensively evaluated using whole genome sequencing and RNA sequencing. PDXs and PDX-derived cell lines largely maintained the expression profiles of the patient from which they were derived despite the emergence of whole-genome duplication (WGD) in a subset of cell lines. These cell line models were heterogeneous in their metastatic capacity and their tissue tropism as observed in both intravenous and orthotopic models. As proof-of-concept study, we used one of these models to test the preclinical effectiveness of a CDK inhibitor on the growth of metastatic tumors in an orthotopic amputation model. Single-agent dinaciclib was effective at dramatically reducing the metastatic burden in this model.

Engineering and Validation of a Peptide-Stabilized Poly(lactic-co-glycolic) Acid Nanoparticle for Targeted Delivery of a Vascular Disruptive Agent in Cancer Therapy

Developing a biocompatible and biodegradable nanoparticle (NP) carrier that integrates drug-loading capability, active targeting, and imaging modality is extremely challenging. Herein, we report an NP with a core of poly(lactic-co-glycolic) acid (PLGA) chemically modified with the drug combretastatin A4 (CA4), a vascular disrupting agent (VDA) in clinical development for ovarian cancer (OvCA) therapy. The NP is stabilized with a short arginine-glycine-aspartic acid-phenylalanine x3 (RGDFFF) peptide via self-assembly of the peptide on the PLGA surface. Importantly, the use of our RGDFFF coating replaces the commonly used polyethylene glycol (PEG) polymer that itself often induces an unwanted immunogenic response. In addition, the RGD motif of the peptide is well-known to preferentially bind to αvβ3 integrin that is implicated in tumor angiogenesis and is exploited as the NP's targeting component. The NP is enhanced with an optical imaging fluorophore label via chemical modification of the PLGA. The RGDFFF-CA4 NPs are synthesized using a nanoprecipitation method and are ∼75 ± 3.7 nm in diameter, where a peptide coating comprises a 2-3 nm outer layer. The NPs are serum stable for 72 h. In vitro studies using human umbilical cord vascular endothelial cells (HUVEC) confirmed the high uptake and biological activity of the RGDFFF-CA4 NP. NP uptake and viability reduction were demonstrated in OvCA cells grown in culture, and the NPs efficiently accumulated in tumors in a preclinical OvCA mouse model. The RGDFFF NP did not induce an inflammatory response when cultured with immune cells. Finally, the NP was efficiently taken up by patient-derived OvCA cells, suggesting a potential for future clinical applications.

Machine learning multi-omics analysis reveals cancer driver dysregulation in pan-cancer cell lines compared to primary tumors

Cancer cell lines have been widely used for decades to study biological processes driving cancer development, and to identify biomarkers of response to therapeutic agents. Advances in genomic sequencing have made possible large-scale genomic characterizations of collections of cancer cell lines and primary tumors, such as the Cancer Cell Line Encyclopedia (CCLE) and The Cancer Genome Atlas (TCGA). These studies allow for the first time a comprehensive evaluation of the comparability of cancer cell lines and primary tumors on the genomic and proteomic level. Here we employ bulk mRNA and micro-RNA sequencing data from thousands of samples in CCLE and TCGA, and proteomic data from partner studies in the MD Anderson Cell Line Project (MCLP) and The Cancer Proteome Atlas (TCPA), to characterize the extent to which cancer cell lines recapitulate tumors. We identify dysregulation of a long non-coding RNA and microRNA regulatory network in cancer cell lines, associated with differential expression between cell lines and primary tumors in four key cancer driver pathways: KRAS signaling, NFKB signaling, IL2/STAT5 signaling and TP53 signaling. Our results emphasize the necessity for careful interpretation of cancer cell line experiments, particularly with respect to therapeutic treatments targeting these important cancer pathways.

Transcriptomics indicate nuclear division and cell adhesion not recapitulated in MCF7 and MCF10A compared to luminal A breast tumours

Breast cancer (BC) cell lines are useful experimental models to understand cancer biology. Yet, their relevance to modelling cancer remains unclear. To better understand the tumour-modelling efficacy of cell lines, we performed RNA-seq analyses on a combined dataset of 2D and 3D cultures of tumourigenic MCF7 and non-tumourigenic MCF10A. To our knowledge, this was the first RNA-seq dataset comprising of 2D and 3D cultures of MCF7 and MCF10A within the same experiment, which facilitates the elucidation of differences between MCF7 and MCF10A across culture types. We compared the genes and gene sets distinguishing MCF7 from MCF10A against separate RNA-seq analyses of clinical luminal A (LumA) and normal samples from the TCGA-BRCA dataset. Among the 1031 cancer-related genes distinguishing LumA from normal samples, only 5.1% and 15.7% of these genes also distinguished MCF7 from MCF10A in 2D and 3D cultures respectively, suggesting that different genes drive cancer-related differences in cell lines compared to clinical BC. Unlike LumA tumours which showed increased nuclear division-related gene expression compared to normal tissue, nuclear division-related gene expression in MCF7 was similar to MCF10A. Moreover, although LumA tumours had similar cell adhesion-related gene expression compared to normal tissues, MCF7 showed reduced cell adhesion-related gene expression compared to MCF10A. These findings suggest that MCF7 and MCF10A cell lines were limited in their ability to model cancer-related processes in clinical LumA tumours.

NetCellMatch: Multiscale Network-Based Matching of Cancer Cell Lines to Patients Using Graphical Wavelets

Cancer cell lines serve as model in vitro systems for investigating therapeutic interventions. Recent advances in high-throughput genomic profiling have enabled the systematic comparison between cell lines and patient tumor samples. The highly interconnected nature of biological data, however, presents a challenge when mapping patient tumors to cell lines. Standard clustering methods can be particularly susceptible to the high level of noise present in these datasets and only output clusters at one unknown scale of the data. In light of these challenges, we present NetCellMatch, a robust framework for network-based matching of cell lines to patient tumors. NetCellMatch first constructs a global network across all cell line-patient samples using their genomic similarity. Then, a multi-scale community detection algorithm integrates information across topologically meaningful (clustering) scales to obtain Network-Based Matching Scores (NBMS). NBMS are measures of cluster robustness which map patient tumors to cell lines. We use NBMS to determine representative "avatar" cell lines for subgroups of patients. We apply NetCellMatch to reverse-phase protein array data obtained from The Cancer Genome Atlas for patients and the MD Anderson Cell Line Project for cell lines. Along with avatar cell line identification, we evaluate connectivity patterns for breast, lung, and colon cancer and explore the proteomic profiles of avatars and their corresponding top matching patients. Our results demonstrate our framework's ability to identify both patient-cell line matches and potential proteomic drivers of similarity. Our methods are general and can be easily adapted to other'omic datasets.

Transcription profiling of feline mammary carcinomas and derived cell lines reveals biomarkers and drug targets associated with metabolic and cell cycle pathways

The molecular heterogeneity of feline mammary carcinomas (FMCs) represents a prognostic and therapeutic challenge. RNA-Seq-based comparative transcriptomic profiling serves to identify recurrent and exclusive differentially expressed genes (DEGs) across sample types and molecular subtypes. Using mass-parallel RNA-Seq, we identified DEGs and performed comparative function-based analysis across 15 tumours (four basal-like triple-negative [TN], eight normal-like TN, and three luminal B fHER2 negative [LB fHER2-]), two cell lines (CL, TiHo-0906, and TiHo-1403) isolated from the primary tumours (LB fHER2-) of two cats included in this study, and 13 healthy mammary tissue controls. DEGs in tumours were predominantly upregulated; dysregulation of CLs transcriptome was more extensive, including mostly downregulated genes. Cell-cycle and metabolic-related DEGs were upregulated in both tumours and CLs, including therapeutically-targetable cell cycle regulators (e.g. CCNB1, CCNB2, CDK1, CDK4, GTSE1, MCM4, and MCM5), metabolic-related genes (e.g. FADS2 and SLC16A3), heat-shock proteins (e.g. HSPH1, HSP90B1, and HSPA5), genes controlling centrosome disjunction (e.g. RACGAP1 and NEK2), and collagen molecules (e.g. COL2A1). DEGs specifically upregulated in basal-like TN tumours were involved in antigen processing and presentation, in normal-like TN tumours encoded G protein-coupled receptors (GPCRs), and in LB fHER2- tumours were associated with lysosomes, phagosomes, and endosomes formation. Downregulated DEGs in CLs were associated with structural and signalling cell surface components. Hence, our results suggest that upregulation of genes enhancing proliferation and metabolism is a common feature among FMCs and derived CLs. In contrast, the dissimilarities observed in dysregulation of membrane components highlight CLs' disconnection with the tumour microenvironment. Furthermore, recurrent and exclusive DEGs associated with dysregulated pathways might be useful for the development of prognostically and therapeutically-relevant targeted panels.

Systematic molecular profiling of acute leukemia cancer stem cells allows identification of druggable targets

Acute myeloid leukemia (AML) is one of the most prevalent and acute blood cancers with a poor prognosis and low overall survival rate, especially in the elderly. Although several new AML markers and drug targets have been recently identified, the rate of long-term cancer eradication has not improved significantly due to the presence and drug resistance of AML cancer stem cells (CSCs). Here we develop a novel computational pipeline to analyze the transcriptomic profiles of AML cancer (stem) cells and identify novel candidate AML CSC markers and drug targets. In our novel pipeline we apply a top-down meta-analysis strategy to integrate The Cancer Genome Atlas data with CSC datasets to infer cell stemness features. As a result, a set of genes termed the "AML key CSC genes" along with all the available drugs/compounds that could target them were identified. Overall, our novel computational pipeline could retrieve known cancer drugs (Carfilzomib) and predicted novel drugs such as Zonisamide, Amitriptyline, and their targets amongst the top ranked drugs and drug targets for targeting AML. Additionally, the pipeline applied in this study could be used for the identification of CSC-specific markers, drivers and their respective targeting drugs in other cancer types.

Integrated analysis of the tumor microenvironment using a reconfigurable microfluidic cell culture platform

The tumor microenvironment (TME) is a complex network of non-malignant cells and stroma that perform a wide array of vital roles in tumor growth, immune evasion, metastasis, and therapeutic resistance. These highly diverse roles have been shown to be critically important to the progression of cancers and have already shown potential as therapeutic targets. Therefore, there has been a tremendous push to elucidate the pathways that underlie these roles and to develop new TME-directed therapies for cancer treatment. Unfortunately, TME-focused research has been limited by a lack of translational in vitro culture platforms that can model this highly complex niche and can support the integrated analysis of cell biology and function. In the current study, we investigate whether an independently developed reconfigurable microfluidic platform, known as Stacks, can address the critical need for translational multi-cellular tumor models and integrated analytics in TME research. We present data on multi-cellular culture of primary human cells in Stacks as well as the orthogonal analysis of cellular polarization, differentiation, migration, and cytotoxicity in this reconfigurable system. These expanded capabilities of Stacks are highly relevant to the cancer research community with the potential to enhance clinical translation of pre-clinical TME studies and to yield novel biological insight into TME crosstalk, metastasis, and responses to novel drug combinations or immune therapies.

High enhancer activity is an epigenetic feature of HPV negative atypical head and neck squamous cell carcinoma

Head and neck squamous cell carcinoma (HNSCC) is a heterogeneous disease with significant mortality and frequent recurrence. Prior efforts to transcriptionally classify HNSCC into groups of varying prognoses have identified four accepted molecular subtypes of the disease: Atypical (AT), Basal (BA), Classical (CL), and Mesenchymal (MS). Here, we investigate the active enhancer landscapes of these subtypes using representative HNSCC cell lines and identify samples belonging to the AT subtype as having increased enhancer activity compared to the other 3 HNSCC subtypes. Cell lines belonging to the AT subtype are more resistant to enhancer-blocking bromodomain inhibitors (BETi). Examination of nascent transcripts reveals that both AT TCGA tumors and cell lines express higher levels of enhancer RNA (eRNA) transcripts for enhancers controlling BETi resistance pathways, such as lipid metabolism and MAPK signaling. Additionally, investigation of higher-order chromatin structure suggests more enhancer-promoter (E-P) contacts in the AT subtype, including on genes identified in the eRNA analysis. Consistently, known BETi resistance pathways are upregulated upon exposure to these inhibitors. Together, our results identify that the AT subtype of HNSCC is associated with higher enhancer activity, resistance to enhancer blockade, and increased signaling through pathways that could serve as future targets for sensitizing HNSCC to BET inhibition.

Computational estimation of quality and clinical relevance of cancer cell lines

Immortal cancer cell lines (CCLs) are the most widely used system for investigating cancer biology and for the preclinical development of oncology therapies. Pharmacogenomic and genome-wide editing screenings have facilitated the discovery of clinically relevant gene-drug interactions and novel therapeutic targets via large panels of extensively characterised CCLs. However, tailoring pharmacological strategies in a precision medicine context requires bridging the existing gaps between tumours and in vitro models. Indeed, intrinsic limitations of CCLs such as misidentification, the absence of tumour microenvironment and genetic drift have highlighted the need to identify the most faithful CCLs for each primary tumour while addressing their heterogeneity, with the development of new models where necessary. Here, we discuss the most significant limitations of CCLs in representing patient features, and we review computational methods aiming at systematically evaluating the suitability of CCLs as tumour proxies and identifying the best patient representative in vitro models. Additionally, we provide an overview of the applications of these methods to more complex models and discuss future machine-learning-based directions that could resolve some of the arising discrepancies.

Widespread redundancy in -omics profiles of cancer mutation states

BACKGROUND

In studies of cellular function in cancer, researchers are increasingly able to choose from many -omics assays as functional readouts. Choosing the correct readout for a given study can be difficult, and which layer of cellular function is most suitable to capture the relevant signal remains unclear.

RESULTS

We consider prediction of cancer mutation status (presence or absence) from functional -omics data as a representative problem that presents an opportunity to quantify and compare the ability of different -omics readouts to capture signals of dysregulation in cancer. From the TCGA Pan-Cancer Atlas that contains genetic alteration data, we focus on RNA sequencing, DNA methylation arrays, reverse phase protein arrays (RPPA), microRNA, and somatic mutational signatures as -omics readouts. Across a collection of genes recurrently mutated in cancer, RNA sequencing tends to be the most effective predictor of mutation state. We find that one or more other data types for many of the genes are approximately equally effective predictors. Performance is more variable between mutations than that between data types for the same mutation, and there is little difference between the top data types. We also find that combining data types into a single multi-omics model provides little or no improvement in predictive ability over the best individual data type.

CONCLUSIONS

Based on our results, for the design of studies focused on the functional outcomes of cancer mutations, there are often multiple -omics types that can serve as effective readouts, although gene expression seems to be a reasonable default option.

Genomic Analysis of Waterpipe Smoke-Induced Lung Tumor Autophagy and Plasticity

The role of autophagy in lung cancer cells exposed to waterpipe smoke (WPS) is not known. Because of the important role of autophagy in tumor resistance and progression, we investigated its relationship with WP smoking. We first showed that WPS activated autophagy, as reflected by LC3 processing, in lung cancer cell lines. The autophagy response in smokers with lung adenocarcinoma, as compared to non-smokers with lung adenocarcinoma, was investigated further using the TCGA lung adenocarcinoma bulk RNA-seq dataset with the available patient metadata on smoking status. The results, based on a machine learning classification model using Random Forest, indicate that smokers have an increase in autophagy-activating genes. Comparative analysis of lung adenocarcinoma molecular signatures in affected patients with a long-term active exposure to smoke compared to non-smoker patients indicates a higher tumor mutational burden, a higher CD8+ T-cell level and a lower dysfunction level in smokers. While the expression of the checkpoint genes tested-PD-1, PD-L1, PD-L2 and CTLA-4-remains unchanged between smokers and non-smokers, B7-1, B7-2, IDO1 and CD200R1 were found to be higher in non-smokers than smokers. Because multiple factors in the tumor microenvironment dictate the success of immunotherapy, in addition to the expression of immune checkpoint genes, our analysis explains why patients who are smokers with lung adenocarcinoma respond better to immunotherapy, even though there are no relative differences in immune checkpoint genes in the two groups. Therefore, targeting autophagy in lung adenocarcinoma patients, in combination with checkpoint inhibitor-targeted therapies or chemotherapy, should be considered in smoker patients with lung adenocarcinoma.

Elucidating gene expression patterns across multiple biological contexts through a large-scale investigation of transcriptomic datasets

Distinct gene expression patterns within cells are foundational for the diversity of functions and unique characteristics observed in specific contexts, such as human tissues and cell types. Though some biological processes commonly occur across contexts, by harnessing the vast amounts of available gene expression data, we can decipher the processes that are unique to a specific context. Therefore, with the goal of developing a portrait of context-specific patterns to better elucidate how they govern distinct biological processes, this work presents a large-scale exploration of transcriptomic signatures across three different contexts (i.e., tissues, cell types, and cell lines) by leveraging over 600 gene expression datasets categorized into 98 subcontexts. The strongest pairwise correlations between genes from these subcontexts are used for the construction of co-expression networks. Using a network-based approach, we then pinpoint patterns that are unique and common across these subcontexts. First, we focused on patterns at the level of individual nodes and evaluated their functional roles using a human protein-protein interactome as a referential network. Next, within each context, we systematically overlaid the co-expression networks to identify specific and shared correlations as well as relations already described in scientific literature. Additionally, in a pathway-level analysis, we overlaid node and edge sets from co-expression networks against pathway knowledge to identify biological processes that are related to specific subcontexts or groups of them. Finally, we have released our data and scripts at https://zenodo.org/record/5831786 and https://github.com/ContNeXt/ , respectively and developed ContNeXt ( https://contnext.scai.fraunhofer.de/ ), a web application to explore the networks generated in this work.

Salivary gland organoid culture maintains distinct glandular properties of murine and human major salivary glands

Salivary glands that produce and secrete saliva, which is essential for lubrication, digestion, immunity, and oral homeostasis, consist of diverse cells. The long-term maintenance of diverse salivary gland cells in organoids remains problematic. Here, we establish long-term murine and human salivary gland organoid cultures. Murine and human salivary gland organoids express gland-specific genes and proteins of acinar, myoepithelial, and duct cells, and exhibit gland functions when stimulated with neurotransmitters. Furthermore, human salivary gland organoids are established from isolated basal or luminal cells, retaining their characteristics. Single-cell RNA sequencing also indicates that human salivary gland organoids contain heterogeneous cell types and replicate glandular diversity. Our protocol also enables the generation of tumoroid cultures from benign and malignant salivary gland tumor types, in which tumor-specific gene signatures are well-conserved. In this study, we provide an experimental platform for the exploration of precision medicine in the era of tissue regeneration and anticancer treatment.

CNpare: matching DNA copy number profiles

SUMMARY

Selecting the optimal cancer cell line for an experiment can be challenging given the diversity of lines available. Here, we present CNpare, which identifies similar cell line models based on genome-wide DNA copy number.

AVAILABILITY

CNpare is available as an R package at https://github.com/macintyrelab/CNpare. All analysis performed in the manuscript can be reproduced via the code found at https://github.com/macintyrelab/CNpare_analyses.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Comparative characterization of 3D chromatin organization in triple-negative breast cancers

Triple-negative breast cancer (TNBC) is a malignant cancer subtype with a high risk of recurrence and an aggressive phenotype compared to other breast cancer subtypes. Although many breast cancer studies conducted to date have investigated genetic variations and differential target gene expression, how 3D chromatin architectures are reorganized in TNBC has been poorly elucidated. Here, using in situ Hi-C technology, we characterized the 3D chromatin organization in cells representing five distinct subtypes of breast cancer (including TNBC) compared to that in normal cells. We found that the global and local 3D architectures were severely disrupted in breast cancer. TNBC cell lines (especially BT549 cells) showed the most dramatic changes relative to normal cells. Importantly, we detected CTCF-dependent TNBC-susceptible losses/gains of 3D chromatin organization and found that these changes were strongly associated with perturbed chromatin accessibility and transcriptional dysregulation. In TNBC tissue, 3D chromatin disorganization was also observed relative to the 3D chromatin organization in normal tissues. We observed that the perturbed local 3D architectures found in TNBC cells were partially conserved in TNBC tissues. Finally, we discovered distinct tissue-specific chromatin loops by comparing normal and TNBC tissues. In this study, we elucidated the characteristics of the 3D chromatin organization in breast cancer relative to normal cells/tissues at multiple scales and identified associations between disrupted structures and various epigenetic features and transcriptomes. Collectively, our findings reveal important 3D chromatin structural features for future diagnostic and therapeutic studies of TNBC.

The 3D architecture of the genome is dramatically altered in an aggressive form of breast cancer, leading to changes in the regulation of gene expression that can fuel tumor growth. A team from South Korea, led by Hyeong-Gon Moon of Seoul National University College of Medicine and Daeyoup Lee of the Korea Advanced Institute of Science and Technology, Daejeon, detailed how chromosomes are positioned and folded within the nucleus of cell liness from five different subtypes of breast cancer. They found that triple-negative breast cancers displayed the most extreme reorganization of their genomes, a pattern also observed in biopsy tissues taken from patients with this subtype of cancer. Knowledge of these conformational changes could inform future efforts to develop therapies and diagnostics for patients with triple-negative breast tumors.

Subclass-specific Prognosis and Treatment Efficacy Inference in Head and Neck Squamous Carcinoma

Exploring the prognostic classification and biomarkers in Head and Neck Squamous Carcinoma (HNSC) is of great clinical significance. We hybridized three prominent strategies to comprehensively characterize the molecular features of HNSC. We constructed a 15-gene signature to predict patients death risk with an average AUC of 0.744 for 1-, 3-, and 5-year on TCGA-HNSC training set, and average AUCs of 0.636, 0.584, 0.755 in GSE65858, GSE-112026, CPTAC-HNSCC datasets, respectively. By combined with NMF clustering and consensus clustering of fraction of tumor immune cell infiltration (ICI) in the tumor microenvironment (TME), we captured a more refined biological characteristics of HNSC, and observed a prognosis heterogeneity in high tumor immunity patients. By matching tumor subset-specific expression signatures to drug-induced cell line expression profiles from large-scale pharmacogenomic databases in the OCTAD workspace, we identified a group of HNSC patients featured with poor prognosis and demonstrated that the individuals in this group are likely to receive increased drug sensitivity to reverse differentially expressed disease signature genes. This trend is especially highlighted among those with higher death risk and tumour immunity.

Progenitor cells derived from gene-engineered human induced pluripotent stem cells as synthetic cancer cell alternatives for in vitro pharmacology

Limitations in genetic stability and recapitulating accurate physiological disease properties challenge the utility of patient-derived (PD) cancer models for reproducible and translational research. We have genetically engineered a portfolio of isogenic human induced pluripotent stem cells (hiPSCs) with different pan-cancer relevant oncoprotein signatures followed by differentiation into lineage-committed progenitor cells. Characterization on molecular and biological level validated successful stable genetic alterations in pluripotency state as well as upon differentiation to prove the functionality of our approach Meanwhile proposing core molecular networks possibly involved in early dysregulation of stem cell homeostasis, the application of our cell systems in comparative substance testing indicates the potential for cancer research such as identification of augmented therapy resistance of stem cells in response to activation of distinct oncogenic signatures. This article is protected by copyright. All rights reserved.

PCR-based profiling of transcription factor activity in vivo by a virus-based reporter battery

Understanding the molecular mechanisms of gene regulation is pivotal for understanding how cells establish and modify their identities and functions. Multiple transcription factors (TFs) coordinate to alter gene expression in cells; however, a method to quantitatively analyze the activity of each TF is lacking, particularly in vivo. Here, we introduce a viral-vector-based TF reporter battery that can be used to simultaneously analyze the activity of multiple TFs, visualized as the TF activity profile (TFAP) obtained by qPCR. We show that the cells possess distinct TFAPs that dynamically change according to experimental manipulation or physiological activity. We report a practical method to obtain the TFAP of a defined cell population and their experience-dependent changes in the mouse brain in vivo. The TFAP obtained by our method will help bridge the information gap between the genome and transcriptome and aid the multi-omics view of understanding the gene regulation system.

•

A virus-based reporter battery for obtaining the TF activity profile of cells

•

TFAP analysis reveals the dynamic change of TF activity upon stimulation

•

Experience-dependent change of TFAP of the mouse brain in vivo

•

Neural and glial change of TF activity revealed by the cell-type-specific TFAP

A text-based computational framework for patient -specific modeling for classification of cancers

Patient heterogeneity precludes cancer treatment and drug development; hence, development of methods for finding prognostic markers for individual treatment is urgently required. Here, we present Pasmopy (Patient-Specific Modeling in Python), a computational framework for stratification of patients using in silico signaling dynamics. Pasmopy converts texts and sentences on biochemical systems into an executable mathematical model. Using this framework, we built a model of the ErbB receptor signaling network, trained in cultured cell lines, and performed in silico simulation of 377 patients with breast cancer using The Cancer Genome Atlas (TCGA) transcriptome datasets. The temporal dynamics of Akt, extracellular signal-regulated kinase (ERK), and c-Myc in each patient were able to accurately predict the difference in prognosis and sensitivity to kinase inhibitors in triple-negative breast cancer (TNBC). Our model applies to any type of signaling network and facilitates the network-based use of prognostic markers and prediction of drug response.

•

A text file describing biochemical systems is converted into an executable model

•

Patient-specific models incorporate individual gene expression profiles

•

In silico signaling dynamics can be utilized as prognostic biomarkers

•

Personalized kinetic models are capable of predicting potential drug targets

A method establishment and comparison of in vivo lung cancer model development platforms for evaluation of tumour metabolism and pharmaceutical efficacy

BACKGROUND

Currently, the identification of accurate biomarkers for the diagnosis of patients with early-stage lung cancer remains difficult. Fortunately, metabolomics technology can be used to improve the detection of plasma metabolic biomarkers for lung cancer. In a previous study, we successfully utilised machine learning methods to identify significant metabolic markers for early-stage lung cancer diagnosis. However, a related research platform for the investigation of tumour metabolism and drug efficacy is still lacking.

HYPOTHESIS/PURPOSE

A novel methodology for the comprehensive evaluation of the internal tumour-metabolic profile and drug evaluation needs to be established.

METHODS

The optimal location for tumour cell inoculation was identified in mouse chest for the non-traumatic orthotopic lung cancer mouse model. Microcomputed tomography (micro-CT) was applied to monitor lung tumour growth. Proscillaridin A (P.A) and cisplatin (CDDP) were utilised to verify the anti-lung cancer efficacy of the platform. The top five clinically valid biomarkers, including proline, L-kynurenine, spermidine, taurine and palmitoyl-L-carnitine, were selected as the evaluation indices to obtain a suitable lung cancer mouse model for clinical metabolomics research by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS).

RESULTS

The platform was successfully established, achieving 100% tumour development rate and 0% surgery mortality. P.A and CDDP had significant anti-lung cancer efficacy in the platform. Compared with the control group, four biomarkers in the orthotopic model and two biomarkers in the metastatic model had significantly higher abundance. Principal component analysis (PCA) showed a significant separation between the orthotopic/metastatic model and the control/subcutaneous/KRAS transgenic model. The platform was mainly involved in arginine and proline metabolism, tryptophan metabolism, and taurine and hypotaurine metabolism.

CONCLUSION

This study is the first to simulate clinical metabolomics by comparing the metabolic phenotype of plasma in different lung cancer mouse models. We found that the orthotopic model was the most suitable for tumour metabolism. Furthermore, the anti-tumour drug efficacy was verified in the platform. The platform can very well match the clinical reality, providing better lung cancer diagnosis and securing more precise evidence for drug evaluation in the future.

Drug screening and genome editing in human pancreatic cancer organoids identifies drug-gene interactions and candidates for off-label treatment

Pancreatic cancer (PDAC) is a highly aggressive malignancy for which the identification of novel therapies is urgently needed. Here, we establish a human PDAC organoid biobank from 31 genetically distinct lines, covering a representative range of tumor subtypes, and demonstrate that these reflect the molecular and phenotypic heterogeneity of primary PDAC tissue. We use CRISPR-Cas9 genome editing and drug screening to characterize drug-gene interactions with ARID1A and BRCA2. We find that missense- but not frameshift mutations in the PDAC driver gene ARID1A are associated with increased sensitivity to the kinase inhibitors dasatinib (p < 0.0001) and VE-821 (p < 0.0001). We conduct an automated drug-repurposing screen with 1,172 FDA-approved compounds, identifying 26 compounds that effectively kill PDAC organoids, including 19 chemotherapy drugs currently approved for other cancer types. We validate the activity of these compounds in vitro and in vivo. The in vivo validated hits include emetine and ouabain, compounds which are approved for non-cancer indications and which perturb the ability of PDAC organoids to respond to hypoxia. Our study provides proof-of-concept for advancing precision oncology and identifying candidates for drug repurposing via genome editing and drug screening in tumor organoid biobanks.

Cautions should be taken when using cell models for gastric cancer research

Gastric cancer (GC)-derived cell lines were generally used in basic cancer research and drug screening. However, it is always concerned about the difference between cultured cells and primary tumor by oncologists. To address this question, we compared differentially expressed genes (DEGs) in primary cancers, healthy tissues, and cell lines both in vitro and in silico. Seven reported genes with decreased expression in GCs by DNA methylation were analyzed in our cohort studies and experimentally validation. Selected datasets from TCGA (The Cancer Genome Atlas), CCLE (The Broad Institute Cancer Cell Line Encyclopedia), and GTEx (The Genotype-Tissue Expression project) were used to represent GCs, GC-derived cell lines, and healthy tissues respectively in the in silico analysis. Thirty gastric tissues together with six cell lines were used for validations. Unexpectedly, we experimentally found that reported cancer-related downregulated genes were only found in cancer cell lines but not in biopsies. The unchanged gene expressions in primary GCs were generally consistent with our cohort study, using information from cancerous (TCGA) and healthy tissues (GETx). Substantial differences were also found between DEGs of cancer tissues (TGCA)/ healthy tissues (GTEx) pair and cell lines (CCLE)/ healthy tissues (GTEx) pair, which confirmed the significant differences between primary cancer and cancer cell lines. Moreover, elevated expression of YWHAQ (14-3-3 δ) and THBS1 were observed in the GC biopsies, which might be potential biomarkers for GC diagnosis, considering the increased YWHAQ and THBS1 associated with poor survival rates in gastric cancer patients. In sum, it is suggested that cautions should be taken when using GC cell lines to study genes that show great differences between cell lines and tissues.