ATG5 and ATG7 Expression Levels Are Reduced in Cutaneous Melanoma and Regulated by NRF1.

Autophagy is a highly conserved cellular process in which intracellular proteins and organelles are sequestered and degraded after the fusion of double-membrane vesicles known as autophagosomes with lysosomes. The process of autophagy is dependent on autophagy-related (ATG) proteins. The role of autophagy in cancer is very complex and still elusive. We investigated the expression of ATG proteins in benign nevi, primary and metastatic melanoma tissues using customized tissue microarrays (TMA). Results from immunohistochemistry show that the expression of ATG5 and ATG7 is significantly reduced in melanoma tissues compared to benign nevi. This reduction correlated with changes in the expression of autophagic activity markers, suggesting decreased basal levels of autophagy in primary and metastatic melanomas. Furthermore, the analysis of survival data of melanoma patients revealed an association between reduced ATG5 and ATG7 levels with an unfavourable clinical outcome. Currently, the mechanisms regulating ATG expression levels in human melanoma remains unknown. Using bioinformatic predictions of transcription factor (TF) binding motifs in accessible chromatin of primary melanocytes, we identified new TFs involved in the regulation of core ATGs. We then show that nuclear respiratory factor 1 (NRF1) stimulates the production of mRNA and protein as well as the promoter activity of ATG5 and ATG7. Moreover, NRF1 deficiency increased in vitro migration of melanoma cells. Our results support the concept that reduced autophagic activity contributes to melanoma development and progression, and identifies NRF1 as a novel TF involved in the regulation of both ATG5 and ATG7 genes.

Hypoxia Routes Tryptophan Homeostasis Towards Increased Tryptamine Production.

The liver is the central hub for processing and maintaining homeostatic levels of dietary nutrients especially essential amino acids such as tryptophan (Trp). Trp is required not only to sustain protein synthesis but also as a precursor for the production of NAD, neurotransmitters and immunosuppressive metabolites. In light of these roles of Trp and its metabolic products, maintaining homeostatic levels of Trp is essential for health and well-being. The liver regulates global Trp supply by the immunosuppressive enzyme tryptophan-2,3-dioxygenase (TDO2), which degrades Trp down the kynurenine pathway (KP). In the current study, we show that isolated primary hepatocytes when exposed to hypoxic environments, extensively rewire their Trp metabolism by reducing constitutive Tdo2 expression and differentially regulating other Trp pathway enzymes and transporters. Mathematical modelling of Trp metabolism in liver cells under hypoxia predicted decreased flux through the KP while metabolic flux through the tryptamine branch significantly increased. In line, the model also revealed an increased accumulation of tryptamines under hypoxia, at the expense of kynurenines. Metabolic measurements in hypoxic hepatocytes confirmed the predicted reduction in KP metabolites as well as accumulation of tryptamine. Tdo2 expression in cultured primary hepatocytes was reduced upon hypoxia inducible factor (HIF) stabilisation by dimethyloxalylglycine (DMOG), demonstrating that HIFs are involved in the hypoxic downregulation of hepatic Tdo2. DMOG abrogated hepatic luciferase signals in Tdo2 reporter mice, indicating that HIF stability also recapitulates hypoxic rewiring of Trp metabolism in vivo. Also in WT mice HIF stabilization drove homeostatic Trp metabolism away from the KP towards enhanced tryptamine production, leading to enhanced levels of tryptamine in liver, serum and brain. As tryptamines are the most potent hallucinogens known, the observed upregulation of tryptamine in response to hypoxic exposure of hepatocytes may be involved in the generation of hallucinations occurring at high altitude. KP metabolites are known to activate the aryl hydrocarbon receptor (AHR). The AHR-activating properties of tryptamines may explain why immunosuppressive AHR activity is maintained under hypoxia despite downregulation of the KP. In summary our results identify hypoxia as an important factor controlling Trp metabolism in the liver with possible implications for immunosuppressive AHR activation and mental disturbances.

L-plastin Ser5 phosphorylation is modulated by the PI3K/SGK pathway and promotes breast cancer cell invasiveness.

BACKGROUND: Metastasis is the predominant cause for cancer morbidity and mortality accounting for approximatively 90% of cancer deaths. The actin-bundling protein L-plastin has been proposed as a metastatic marker and phosphorylation on its residue Ser5 is known to increase its actin-bundling activity. We recently showed that activation of the ERK/MAPK signalling pathway leads to L-plastin Ser5 phosphorylation and that the downstream kinases RSK1 and RSK2 are able to directly phosphorylate Ser5. Here we investigate the involvement of the PI3K pathway in L-plastin Ser5 phosphorylation and the functional effect of this phosphorylation event in breast cancer cells. METHODS: To unravel the signal transduction network upstream of L-plastin Ser5 phosphorylation, we performed computational modelling based on immunoblot analysis data, followed by experimental validation through inhibition/overexpression studies and in vitro kinase assays. To assess the functional impact of L-plastin expression/Ser5 phosphorylation in breast cancer cells, we either silenced L-plastin in cell lines initially expressing endogenous L-plastin or neoexpressed L-plastin wild type and phosphovariants in cell lines devoid of endogenous L-plastin. The established cell lines were used for cell biology experiments and confocal microscopy analysis. RESULTS: Our modelling approach revealed that, in addition to the ERK/MAPK pathway and depending on the cellular context, the PI3K pathway contributes to L-plastin Ser5 phosphorylation through its downstream kinase SGK3. The results of the transwell invasion/migration assays showed that shRNA-mediated knockdown of L-plastin in BT-20 or HCC38 cells significantly reduced cell invasion, whereas stable expression of the phosphomimetic L-plastin Ser5Glu variant led to increased migration and invasion of BT-549 and MDA-MB-231 cells. Finally, confocal image analysis combined with zymography experiments and gelatin degradation assays provided evidence that L-plastin Ser5 phosphorylation promotes L-plastin recruitment to invadopodia, MMP-9 activity and concomitant extracellular matrix degradation. CONCLUSION: Altogether, our results demonstrate that L-plastin Ser5 phosphorylation increases breast cancer cell invasiveness. Being a downstream molecule of both ERK/MAPK and PI3K/SGK pathways, L-plastin is proposed here as a potential target for therapeutic approaches that are aimed at blocking dysregulated signalling outcome of both pathways and, thus, at impairing cancer cell invasion and metastasis formation. Video abstract.

Machine learning-assisted neurotoxicity prediction in human midbrain organoids.

INTRODUCTION: Brain organoids are highly complex multi-cellular tissue proxies, which have recently risen as novel tools to study neurodegenerative diseases such as Parkinson's disease (PD). However, with increasing complexity of the system, usage of quantitative tools becomes challenging. OBJECTIVES: The primary objective of this study was to develop a neurotoxin-induced PD organoid model and to assess the neurotoxic effect on dopaminergic neurons using microscopy-based phenotyping in a high-content fashion. METHODS: We describe a pipeline for a machine learning-based analytical method, allowing for detailed image-based cell profiling and toxicity prediction in brain organoids treated with the neurotoxic compound 6-hydroxydopamine (6-OHDA). RESULTS: We quantified features such as dopaminergic neuron count and neuronal complexity and built a machine learning classifier with the data to optimize data processing strategies and to discriminate between different treatment conditions. We validated the approach with high content imaging data from PD patient derived midbrain organoids. CONCLUSIONS: The here described model is a valuable tool for advanced in vitro PD modeling and to test putative neurotoxic compounds.

Computational models of melanoma.

Genes, proteins, or cells influence each other and consequently create patterns, which can be increasingly better observed by experimental biology and medicine. Thereby, descriptive methods of statistics and bioinformatics sharpen and structure our perception. However, additionally considering the interconnectivity between biological elements promises a deeper and more coherent understanding of melanoma. For instance, integrative network-based tools and well-grounded inductive in silico research reveal disease mechanisms, stratify patients, and support treatment individualization. This review gives an overview of different modeling techniques beyond statistics, shows how different strategies align with the respective medical biology, and identifies possible areas of new computational melanoma research.

Cancer Cells Employ Nuclear Caspase-8 to Overcome the p53-Dependent G2/M Checkpoint through Cleavage of USP28.

Cytosolic caspase-8 is a mediator of death receptor signaling. While caspase-8 expression is lost in some tumors, it is increased in others, indicating a conditional pro-survival function of caspase-8 in cancer. Here, we show that tumor cells employ DNA-damage-induced nuclear caspase-8 to override the p53-dependent G2/M cell-cycle checkpoint. Caspase-8 is upregulated and localized to the nucleus in multiple human cancers, correlating with treatment resistance and poor clinical outcome. Depletion of caspase-8 causes G2/M arrest, stabilization of p53, and induction of p53-dependent intrinsic apoptosis in tumor cells. In the nucleus, caspase-8 cleaves and inactivates the ubiquitin-specific peptidase 28 (USP28), preventing USP28 from de-ubiquitinating and stabilizing wild-type p53. This results in de facto p53 protein loss, switching cell fate from apoptosis toward mitosis. In summary, our work identifies a non-canonical role of caspase-8 exploited by cancer cells to override the p53-dependent G2/M cell-cycle checkpoint.

Impaired serine metabolism complements LRRK2-G2019S pathogenicity in PD patients.

Parkinson's disease (PD) is a multifactorial disorder with complex etiology. The most prevalent PD associated mutation, LRRK2-G2019S is linked to familial and sporadic cases. Based on the multitude of genetic predispositions in PD and the incomplete penetrance of LRRK2-G2019S, we hypothesize that modifiers in the patients' genetic background act as susceptibility factors for developing PD. To assess LRRK2-G2019S modifiers, we used human induced pluripotent stem cell-derived neuroepithelial stem cells (NESCs). Isogenic controls distinguish between LRRK2-G2019S dependent and independent cellular phenotypes. LRRK2-G2019S patient and healthy mutagenized lines showed altered NESC self-renewal and viability, as well as impaired serine metabolism. In patient cells, phenotypes were only partly LRRK2-G2019S dependent, suggesting a significant contribution of the genetic background. In this context we identified the gene serine racemase (SRR) as a novel patient-specific, developmental, genetic modifier contributing to the aberrant phenotypes. Its enzymatic product, d-serine, rescued altered cellular phenotypes. Susceptibility factors in the genetic background, such as SRR, could be new targets for early PD diagnosis and treatment.

Clinical update on head and neck cancer: molecular biology and ongoing challenges.

Head and neck squamous cell carcinomas (HNSCCs) are an aggressive, genetically complex and difficult to treat group of cancers. In lieu of truly effective targeted therapies, surgery and radiotherapy represent the primary treatment options for most patients. But these treatments are associated with significant morbidity and a reduction in quality of life. Resistance to both radiotherapy and the only available targeted therapy, and subsequent relapse are common. Research has therefore focussed on identifying biomarkers to stratify patients into clinically meaningful groups and to develop more effective targeted therapies. However, as we are now discovering, the poor response to therapy and aggressive nature of HNSCCs is not only affected by the complex alterations in intracellular signalling pathways but is also heavily influenced by the behaviour of the extracellular microenvironment. The HNSCC tumour landscape is an environment permissive of these tumours' aggressive nature, fostered by the actions of the immune system, the response to tumour hypoxia and the influence of the microbiome. Solving these challenges now rests on expanding our knowledge of these areas, in parallel with a greater understanding of the molecular biology of HNSCC subtypes. This update aims to build on our earlier 2014 review by bringing up to date our understanding of the molecular biology of HNSCCs and provide insights into areas of ongoing research and perspectives for the future.

Systemic network analysis identifies XIAP and IκBα as potential drug targets in TRAIL resistant BRAF mutated melanoma.

Metastatic melanoma remains a life-threatening disease because most tumors develop resistance to targeted kinase inhibitors thereby regaining tumorigenic capacity. We show the 2nd generation hexavalent TRAIL receptor-targeted agonist IZI1551 to induce pronounced apoptotic cell death in mutBRAF melanoma cells. Aiming to identify molecular changes that may confer IZI1551 resistance we combined Dynamic Bayesian Network modelling with a sophisticated regularization strategy resulting in sparse and context-sensitive networks and show the performance of this strategy in the detection of cell line-specific deregulations of a signalling network. Comparing IZI1551-sensitive to IZI1551-resistant melanoma cells the model accurately and correctly predicted activation of NFκB in concert with upregulation of the anti-apoptotic protein XIAP as the key mediator of IZI1551 resistance. Thus, the incorporation of multiple regularization functions in logical network optimization may provide a promising avenue to assess the effects of drug combinations and to identify responders to selected combination therapies.

Stroma-induced phenotypic plasticity offers phenotype-specific targeting to improve melanoma treatment.

Cancer cells' phenotypic plasticity, promoted by stromal cells, contributes to intra-tumoral heterogeneity and affects response to therapy. We have disclosed an association between fibroblast-stimulated phenotype switching and resistance to the clinically used BRAF inhibitor (BRAFi) vemurafenib in malignant melanoma, revealing a challenge in targeting the fibroblast-induced phenotype. Here we compared molecular features and drug sensitivity in melanoma cells grown as co-cultures with fibroblasts versus mono-cultures. In the presence of fibroblasts, melanoma cells switched to the dedifferentiated, mesenchymal-like, inflammatory phenotype that showed reduced sensitivity to the most of 275 tested cancer drugs. Fibroblasts, however, sensitized melanoma cells to PI3K inhibitors (PI3Ki) and particularly the inhibitor of GSK3, AR-A014418 (GSK3i), that showed superior efficacy in co-cultures. The proteome changes induced by the BRAFi + GSK3i combination mimicked changes induced by BRAFi in mono-cultures, and GSK3i in co-cultures. This suggests that the single drug drives the response to the combination treatment, depending on fibroblast presence or absence, consequently, phenotype. We propose that the BRAFi and GSK3i (or PI3Ki) combination exemplifies phenotype-specific combinatorial treatment that should be beneficial in phenotypically heterogeneous tumors rich in stromal interactions.

Using Regularization to Infer Cell Line Specificity in Logical Network Models of Signaling Pathways.

Understanding the functional properties of cells of different origins is a long-standing challenge of personalized medicine. Especially in cancer, the high heterogeneity observed in patients slows down the development of effective cures. The molecular differences between cell types or between healthy and diseased cellular states are usually determined by the wiring of regulatory networks. Understanding these molecular and cellular differences at the systems level would improve patient stratification and facilitate the design of rational intervention strategies. Models of cellular regulatory networks frequently make weak assumptions about the distribution of model parameters across cell types or patients. These assumptions are usually expressed in the form of regularization of the objective function of the optimization problem. We propose a new method of regularization for network models of signaling pathways based on the local density of the inferred parameter values within the parameter space. Our method reduces the complexity of models by creating groups of cell line-specific parameters which can then be optimized together. We demonstrate the use of our method by recovering the correct topology and inferring accurate values of the parameters of a small synthetic model. To show the value of our method in a realistic setting, we re-analyze a recently published phosphoproteomic dataset from a panel of 14 colon cancer cell lines. We conclude that our method efficiently reduces model complexity and helps recovering context-specific regulatory information.

Resolving the Combinatorial Complexity of Smad Protein Complex Formation and Its Link to Gene Expression.

Upon stimulation of cells with transforming growth factor ß (TGF-ß), Smad proteins form trimeric complexes and activate a broad spectrum of target genes. It remains unresolved which of the possible Smad complexes are formed in cellular contexts and how these contribute to gene expression. By combining quantitative mass spectrometry with a computational selection strategy, we predict and provide experimental evidence for the three most relevant Smad complexes in the mouse hepatoma cell line Hepa1-6. Utilizing dynamic pathway modeling, we specify the contribution of each Smad complex to the expression of representative Smad target genes, and show that these contributions are conserved in human hepatoma cell lines and primary hepatocytes. We predict, based on gene expression data of patient samples, increased amounts of Smad2/3/4 proteins and Smad2 phosphorylation as hallmarks of hepatocellular carcinoma and experimentally verify this prediction. Our findings demonstrate that modeling approaches can disentangle the complexity of transcription factor complex formation and its impact on gene expression.

Model Based Targeting of IL-6-Induced Inflammatory Responses in Cultured Primary Hepatocytes to Improve Application of the JAK Inhibitor Ruxolitinib.

IL-6 is a central mediator of the immediate induction of hepatic acute phase proteins (APP) in the liver during infection and after injury, but increased IL-6 activity has been associated with multiple pathological conditions. In hepatocytes, IL-6 activates JAK1-STAT3 signaling that induces the negative feedback regulator SOCS3 and expression of APPs. While different inhibitors of IL-6-induced JAK1-STAT3-signaling have been developed, understanding their precise impact on signaling dynamics requires a systems biology approach. Here we present a mathematical model of IL-6-induced JAK1-STAT3 signaling that quantitatively links physiological IL-6 concentrations to the dynamics of IL-6-induced signal transduction and expression of target genes in hepatocytes. The mathematical model consists of coupled ordinary differential equations (ODE) and the model parameters were estimated by a maximum likelihood approach, whereas identifiability of the dynamic model parameters was ensured by the Profile Likelihood. Using model simulations coupled with experimental validation we could optimize the long-term impact of the JAK-inhibitor Ruxolitinib, a therapeutic compound that is quickly metabolized. Model-predicted doses and timing of treatments helps to improve the reduction of inflammatory APP gene expression in primary mouse hepatocytes close to levels observed during regenerative conditions. The concept of improved efficacy of the inhibitor through multiple treatments at optimized time intervals was confirmed in primary human hepatocytes. Thus, combining quantitative data generation with mathematical modeling suggests that repetitive treatment with Ruxolitinib is required to effectively target excessive inflammatory responses without exceeding doses recommended by the clinical guidelines.

FALCON: a toolbox for the fast contextualization of logical networks.

MOTIVATION: Mathematical modelling of regulatory networks allows for the discovery of knowledge at the system level. However, existing modelling tools are often computation-heavy and do not offer intuitive ways to explore the model, to test hypotheses or to interpret the results biologically. RESULTS: We have developed a computational approach to contextualize logical models of regulatory networks with biological measurements based on a probabilistic description of rule-based interactions between the different molecules. Here, we propose a Matlab toolbox, FALCON, to automatically and efficiently build and contextualize networks, which includes a pipeline for conducting parameter analysis, knockouts and easy and fast model investigation. The contextualized models could then provide qualitative and quantitative information about the network and suggest hypotheses about biological processes. AVAILABILITY AND IMPLEMENTATION: FALCON is freely available for non-commercial users on GitHub under the GPLv3 licence. The toolbox, installation instructions, full documentation and test datasets are available at https://github.com/sysbiolux/FALCON. FALCON runs under Matlab (MathWorks) and requires the Optimization Toolbox. CONTACT: thomas.sauter@uni.lu. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Downregulation of the TGFß Pseudoreceptor BAMBI in Non-Small Cell Lung Cancer Enhances TGFß Signaling and Invasion.

Non-small cell lung cancer (NSCLC) is characterized by early metastasis and has the highest mortality rate among all solid tumors, with the majority of patients diagnosed at an advanced stage where curative therapeutic options are lacking. In this study, we identify a targetable mechanism involving TGFß elevation that orchestrates tumor progression in this disease. Substantial activation of this pathway was detected in human lung cancer tissues with concomitant downregulation of BAMBI, a negative regulator of the TGFß signaling pathway. Alterations of epithelial-to-mesenchymal transition (EMT) marker expression were observed in lung cancer samples compared with tumor-free tissues. Distinct alterations in the DNA methylation of the gene regions encoding TGFß pathway components were detected in NSCLC samples compared with tumor-free lung tissues. In particular, epigenetic silencing of BAMBI was identified as a hallmark of NSCLC. Reconstitution of BAMBI expression in NSCLC cells resulted in a marked reduction of TGFß-induced EMT, migration, and invasion in vitro, along with reduced tumor burden and tumor growth in vivo In conclusion, our results demonstrate how BAMBI downregulation drives the invasiveness of NSCLC, highlighting TGFß signaling as a candidate therapeutic target in this setting. Cancer Res; 76(13); 3785-801. ©2016 AACR.

Context-specific flow through the MEK/ERK module produces cell- and ligand-specific patterns of ERK single and double phosphorylation.

The same pathway, such as the mitogen-activated protein kinase (MAPK) pathway, can produce different cellular responses, depending on stimulus or cell type. We examined the phosphorylation dynamics of the MAPK kinase MEK and its targets extracellular signal-regulated kinase 1 and 2 (ERK1/2) in primary hepatocytes and the transformed keratinocyte cell line HaCaT A5 exposed to either hepatocyte growth factor or interleukin-6. By combining quantitative mass spectrometry with dynamic modeling, we elucidated network structures for the reversible threonine and tyrosine phosphorylation of ERK in both cell types. In addition to differences in the phosphorylation and dephosphorylation reactions, the HaCaT network model required two feedback mechanisms, which, as the experimental data suggested, involved the induction of the dual-specificity phosphatase DUSP6 and the scaffold paxillin. We assayed and modeled the accumulation of the double-phosphorylated and active form of ERK1/2, as well as the dynamics of the changes in the monophosphorylated forms of ERK1/2. Modeling the differences in the dynamics of the changes in the distributions of the phosphorylated forms of ERK1/2 suggested that different amounts of MEK activity triggered context-specific responses, with primary hepatocytes favoring the formation of double-phosphorylated ERK1/2 and HaCaT A5 cells that produce both the threonine-phosphorylated and the double-phosphorylated form. These differences in phosphorylation distributions explained the threshold, sensitivity, and saturation of the ERK response. We extended the findings of differential ERK phosphorylation profiles to five additional cultured cell systems and matched liver tumor and normal tissue, which revealed context-specific patterns of the various forms of phosphorylated ERK.

Excemplify: a flexible template based solution, parsing and managing data in spreadsheets for experimentalists.

In systems biology, quantitative experimental data is the basis of building mathematical models. In most of the cases, they are stored in Excel files and hosted locally. To have a public database for collecting, retrieving and citing experimental raw data as well as experimental conditions is important for both experimentalists and modelers. However, the great effort needed in the data handling procedure and in the data submission procedure becomes the crucial limitation for experimentalists to contribute to a database, thereby impeding the database to deliver its benefit. Moreover, manual copy and paste operations which are commonly used in those procedures increase the chance of making mistakes. Excemplify, a web-based application, proposes a flexible and adaptable template-based solution to solve these problems. Comparing to the normal template based uploading approach, which is supported by some public databases, rather than predefining a format that is potentiall impractical, Excemplify allows users to create their own experiment-specific content templates in different experiment stages and to build corresponding knowledge bases for parsing. Utilizing the embedded knowledge of used templates, Excemplify is able to parse experimental data from the initial setup stage and generate following stages spreadsheets automatically. The proposed solution standardizes the flows of data traveling according to the standard procedures of applying the experiment, cuts down the amount of manual effort and reduces the chance of mistakes caused by manual data handling. In addition, it maintains the context of meta-data from the initial preparation manuscript and improves the data consistency. It interoperates and complements RightField and SEEK as well.

Dynamics and feedback loops in the transforming growth factor ß signaling pathway.

Transforming growth factor ß (TGF-ß) ligands activate a signaling cascade with multiple cell context dependent outcomes. Disruption or disturbance leads to variant clinical disorders. To develop strategies for disease intervention, delineation of the pathway in further detail is required. Current theoretical models of this pathway describe production and degradation of signal mediating proteins and signal transduction from the cell surface into the nucleus, whereas feedback loops have not exhaustively been included. In this study we present a mathematical model to determine the relevance of feedback regulators (Arkadia, Smad7, Smurf1, Smurf2, SnoN and Ski) on TGF-ß target gene expression and the potential to initiate stable oscillations within a realistic parameter space. We employed massive sampling of the parameters space to pinpoint crucial players for potential oscillations as well as transcriptional product levels. We identified Smad7 and Smurf2 with the highest impact on the dynamics. Based on these findings, we conducted preliminary time course experiments.