Defining novel causal SNPs and linked phenotypes at melanoma-associated loci.

A number of genomic regions have been associated with melanoma risk through genome-wide association studies; however, the causal variants underlying the majority of these associations remain unknown. Here, we sequenced either the full locus or the functional regions including exons of 19 melanoma-associated loci in 1959 British melanoma cases and 737 controls. Variant filtering followed by Fisher's exact test analyses identified 66 variants associated with melanoma risk. Sequential conditional logistic regression identified the distinct haplotypes on which variants reside, and massively parallel reporter assays provided biological insights into how these variants influence gene function. We performed further analyses to link variants to melanoma risk phenotypes and assessed their association with melanoma-specific survival. Our analyses replicate previously known associations in the melanocortin 1 receptor (MC1R) and tyrosinase (TYR) loci, while identifying novel potentially causal variants at the MTAP/CDKN2A and CASP8 loci. These results improve our understanding of the architecture of melanoma risk and outcome.

Oncolytic virus treatment differentially affects the CD56dim and CD56bright NK cell subsets in vivo and regulates a spectrum of human NK cell activity.

Natural killer (NK) cells protect against intracellular infection and cancer. These properties are exploited in oncolytic virus (OV) therapy, where antiviral responses enhance anti-tumour immunity. We have analysed the mechanism by which reovirus, an oncolytic dsRNA virus, modulates human NK cell activity. Reovirus activates NK cells in a type I interferon (IFN-I) dependent manner, inducing STAT1 and STAT4 signalling in both CD56dim and CD56bright NK cell subsets. Gene expression profiling revealed the dominance of IFN-I responses and identified induction of genes associated with NK cell cytotoxicity and cell cycle progression, with distinct responses in the CD56dim and CD56bright subsets. However, reovirus treatment inhibited IL-15 induced NK cell proliferation in an IFN-I dependent manner and was associated with reduced AKT signalling. In vivo, human CD56dim and CD56bright NK cells responded with similar kinetics to reovirus treatment, but CD56bright NK cells were transiently lost from the peripheral circulation at the peak of the IFN-I response, suggestive of their redistribution to secondary lymphoid tissue. Coupled with the direct, OV-mediated killing of tumour cells, the activation of both CD56dim and CD56bright NK cells by antiviral pathways induces a spectrum of activity that includes the NK cell-mediated killing of tumour cells and modulation of adaptive responses via the trafficking of IFN-γ expressing CD56bright NK cells to lymph nodes.

Chemically induced neurite-like outgrowth reveals a multicellular network function in patient-derived glioblastoma cells.

Tumor stem cells and malignant multicellular networks have been separately implicated in the therapeutic resistance of glioblastoma multiforme (GBM), the most aggressive type of brain cancer in adults. Here, we show that small-molecule inhibition of RHO-associated serine/threonine kinase proteins (ROCKi) significantly promoted the outgrowth of neurite-like cell projections in cultures of heterogeneous patient-derived GBM stem-like cells. These projections formed de novo-induced cellular network (iNet) 'webs', which regressed after withdrawal of ROCKi. Connected cells within the iNet web exhibited long range Ca2+ signal transmission, and significant lysosomal and mitochondrial trafficking. In contrast to their less-connected vehicle control counterparts, iNet cells remained viable and proliferative after high-dose radiation. These findings demonstrate a link between ROCKi-regulated cell projection dynamics and the formation of radiation-resistant multicellular networks. Our study identifies means to reversibly induce iNet webs ex vivo, and may thereby accelerate future studies into the biology of GBM cellular networks.

The clinicopathologic spectrum and genomic landscape of de-/trans-differentiated melanoma.

Dedifferentiation and transdifferentiation are rare and only poorly understood phenomena in cutaneous melanoma. To study this disease more comprehensively we have retrieved 11 primary cutaneous melanomas from our pathology archives showing biphasic features characterized by a conventional melanoma and additional areas of de-/trans-differentiation as defined by a lack of immunohistochemical expression of all conventional melanocytic markers (S-100 protein, SOX10, Melan-A, and HMB-45). The clinical, histologic, and immunohistochemical findings were recorded and follow-up was obtained. The patients were mostly elderly (median: 81 years; range: 42-86 years) without significant gender predilection, and the sun-exposed skin of the head and neck area was most commonly affected. The tumors were deeply invasive with a mean depth of 7 mm (range: 4-80 mm). The dedifferentiated component showed atypical fibroxanthoma-like features in the majority of cases (7), while additional rhabdomyosarcomatous and epithelial transdifferentiation was noted histologically and/or immunohistochemically in two tumors each. The background conventional melanoma component was of desmoplastic (4), superficial spreading (3), nodular (2), lentigo maligna (1), or spindle cell (1) types. For the seven patients with available follow-up data (median follow-up period of 25 months; range: 8-36 months), two died from their disease, and three developed metastases. Next-generation sequencing of the cohort revealed somatic mutations of established melanoma drivers including mainly NF1 mutations (5) in the conventional component, which was also detected in the corresponding de-/trans-differentiated component. In summary, the diagnosis of primary cutaneous de-/trans-differentiated melanoma is challenging and depends on the morphologic identification of conventional melanoma. Molecular analysis is diagnostically helpful as the mutated gene profile is shared between the conventional and de-/trans-differentiated components. Importantly, de-/trans-differentiation does not appear to confer a more aggressive behavior.

Simulation of heterogeneous tumour genomes with HeteroGenesis and in silico whole exome sequencing.

SUMMARY: Tumour evolution results in progressive cancer phenotypes such as metastatic spread and treatment resistance. To better treat cancers, we must characterize tumour evolution and the genetic events that confer progressive phenotypes. This is facilitated by high coverage genome or exome sequencing. However, the best approach by which, or indeed whether, these data can be used to accurately model and interpret underlying evolutionary dynamics is yet to be confirmed. Establishing this requires sequencing data from appropriately heterogeneous tumours in which the exact trajectory and combination of events occurring throughout its evolution are known. We therefore developed HeteroGenesis: a tool to generate realistically evolved tumour genomes, which can be sequenced using weighted-Wessim (w-Wessim), an in silico exome sequencing tool that we have adapted from previous methods. HeteroGenesis simulates more complex and realistic heterogeneous tumour genomes than existing methods, can model different evolutionary dynamics, and enables the creation of multi-region and longitudinal data. AVAILABILITY AND IMPLEMENTATION: HeteroGenesis and w-Wessim are freely available under the GNU General Public Licence from https://github.com/GeorgetteTanner, implemented in Python and supported on linux and MS Windows. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

How to analyse the spatiotemporal tumour samples needed to investigate cancer evolution: A case study using paired primary and recurrent glioblastoma.

Many traits of cancer progression (e.g., development of metastases or resistance to therapy) are facilitated by tumour evolution: Darwinian selection of subclones with distinct genotypes or phenotypes that enable such progression. Characterising these subclones provide an opportunity to develop drugs to better target their specific properties but requires the accurate identification of somatic mutations shared across multiple spatiotemporal tumours from the same patient. Current best practices for calling somatic mutations are optimised for single samples, and risk being too conservative to identify shared mutations with low prevalence in some samples. We reasoned that datasets from multiple matched tumours can be used for mutual validation and thus propose an adapted two-stage approach: (1) low-stringency mutation calling to identify mutations shared across samples irrespective of the weight of evidence in a single sample; (2) high-stringency mutation calling to further characterise mutations present in a single sample. We applied our approach to three-independent cohorts of paired primary and recurrent glioblastoma tumours, two of which have previously been analysed using existing approaches, and found that it significantly increased the amount of biologically relevant shared somatic mutations identified. We also found that duplicate removal was detrimental when identifying shared somatic mutations. Our approach is also applicable when multiple datasets e.g. DNA and RNA are available for the same tumour.

Phenotype-independent DNA methylation changes in prostate cancer.

BACKGROUND: Human prostate cancers display numerous DNA methylation changes compared to normal tissue samples. However, definitive identification of features related to the cells' malignant status has been compromised by the predominance of cells with luminal features in prostate cancers. METHODS: We generated genome-wide DNA methylation profiles of cell subpopulations with basal or luminal features isolated from matched prostate cancer and normal tissue samples. RESULTS: Many frequent DNA methylation changes previously attributed to prostate cancers are here identified as differences between luminal and basal cells in both normal and cancer samples. We also identified changes unique to each of the two cancer subpopulations. Those specific to cancer luminal cells were associated with regulation of metabolic processes, cell proliferation and epithelial development. Within the prostate cancer TCGA dataset, these changes were able to distinguish not only cancers from normal samples, but also organ-confined cancers from those with extraprostatic extensions. Using changes present in both basal and luminal cancer cells, we derived a new 17-CpG prostate cancer signature with high predictive power in the TCGA dataset. CONCLUSIONS: This study demonstrates the importance of comparing phenotypically matched prostate cell populations from normal and cancer tissues to unmask biologically and clinically relevant DNA methylation changes.

fqtools: an efficient software suite for modern FASTQ file manipulation.

UNLABELLED: Many Next Generation Sequencing analyses involve the basic manipulation of input sequence data before downstream processing (e.g. searching for specific sequences, format conversion or basic file statistics). The rapidly increasing data volumes involved in NGS make any dataset manipulation a time-consuming and error-prone process. I have developed fqtools; a fast and reliable FASTQ file manipulation suite that can process the full set of valid FASTQ files, including those with multi-line sequences, whilst identifying invalid files. Fqtools is faster than similar tools, and is designed for use in automatic processing pipelines. AVAILABILITY AND IMPLEMENTATION: fqtools is open source and is available at: https://github.com/alastair-droop/fqtools SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online. CONTACT: a.p.droop@leeds.ac.uk.

qsubsec: a lightweight template system for defining sun grid engine workflows.

UNLABELLED: The Sun Grid Engine (SGE) high-performance computing batch queueing system is commonly used in bioinformatics analysis. Creating re-usable scripts for the SGE is a common challenge. The qsubsec template language and interpreter described here allow researchers to easily create generic template definitions that encapsulate a particular computational job, effectively separating the process logic from the specific run details. At submission time, the generic template is filled in with specific values. This system provides an intermediate level between simple scripting and complete workflow management tools. AVAILABILITY AND IMPLEMENTATION: Qsubsec is open-source and is available at https://github.com/alastair-droop/qsubsec CONTACT: a.p.droop@leeds.ac.uk SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Retinoic acid and androgen receptors combine to achieve tissue specific control of human prostatic transglutaminase expression: a novel regulatory network with broader significance.

In the human prostate, expression of prostate-specific genes is known to be directly regulated by the androgen-induced stimulation of the androgen receptor (AR). However, less is known about the expression control of the prostate-restricted TGM4 (hTGP) gene. In the present study we demonstrate that the regulation of the hTGP gene depends mainly on retinoic acid (RA). We provide evidence that the retinoic acid receptor gamma (RAR-G) plays a major role in the regulation of the hTGP gene and that presence of the AR, but not its transcriptional transactivation activity, is critical for hTGP transcription. RA and androgen responsive elements (RARE and ARE) were mapped to the hTGP promoter by chromatin immunoprecipitation (ChIP), which also indicated that the active ARE and RARE sites were adjacent, suggesting that the antagonistic effect of androgen and RA is related to the relative position of binding sites. Publicly available AR and RAR ChIP-seq data was used to find gene potentially regulated by AR and RAR. Four of these genes (CDCA7L, CDK6, BTG1 and SAMD3) were tested for RAR and AR binding and two of them (CDCA7L and CDK6) proved to be antagonistically regulated by androgens and RA confirming that this regulation is not particular of hTGP.

IDHwt glioblastomas can be stratified by their transcriptional response to standard treatment, with implications for targeted therapy.

BACKGROUND: Glioblastoma (GBM) brain tumors lacking IDH1 mutations (IDHwt) have the worst prognosis of all brain neoplasms. Patients receive surgery and chemoradiotherapy but tumors almost always fatally recur. RESULTS: Using RNA sequencing data from 107 pairs of pre- and post-standard treatment locally recurrent IDHwt GBM tumors, we identify two responder subtypes based on longitudinal changes in gene expression. In two thirds of patients, a specific subset of genes is upregulated from primary to recurrence (Up responders), and in one third, the same genes are downregulated (Down responders), specifically in neoplastic cells. Characterization of the responder subtypes indicates subtype-specific adaptive treatment resistance mechanisms that are associated with distinct changes in the tumor microenvironment. In Up responders, recurrent tumors are enriched in quiescent proneural GBM stem cells and differentiated neoplastic cells, with increased interaction with the surrounding normal brain and neurotransmitter signaling, whereas Down responders commonly undergo mesenchymal transition. ChIP-sequencing data from longitudinal GBM tumors suggests that the observed transcriptional reprogramming could be driven by Polycomb-based chromatin remodeling rather than DNA methylation. CONCLUSIONS: We show that the responder subtype is cancer-cell intrinsic, recapitulated in in vitro GBM cell models, and influenced by the presence of the tumor microenvironment. Stratifying GBM tumors by responder subtype may lead to more effective treatment.

Biglycan and reduced glycolysis are associated with breast cancer cell dormancy in the brain.

Exit of quiescent disseminated cancer cells from dormancy is thought to be responsible for metastatic relapse and a better understanding of dormancy could pave the way for novel therapeutic approaches. We used an in vivo model of triple negative breast cancer brain metastasis to identify differences in transcriptional profiles between dormant and proliferating cancer cells in the brain. BGN gene, encoding a small proteoglycan biglycan, was strongly upregulated in dormant cancer cells in vivo. BGN expression was significantly downregulated in patient brain metastases as compared to the matched primary breast tumors and BGN overexpression in cancer cells inhibited their growth in vitro and in vivo. Dormant cancer cells were further characterized by a reduced expression of glycolysis genes in vivo, and inhibition of glycolysis in vitro resulted in a reversible growth arrest reminiscent of dormancy. Our study identified mechanisms that could be targeted to induce/maintain cancer dormancy and thereby prevent metastatic relapse.

ETS transcription factor ELF3 (ESE-1) is a cell cycle regulator in benign and malignant prostate.

This study aimed to elucidate the role of ELF3, an ETS family member in normal prostate growth and prostate cancer. Silencing ELF3 in both benign prostate (BPH-1) and prostate cancer (PC3) cell lines resulted in decreased colony-forming ability, inhibition of cell migration and reduced cell viability due to cell cycle arrest, establishing ELF3 as a cell cycle regulator. Increased ELF3 expression in more advanced prostate tumours was shown by immunostaining of tissue microarrays and from analysis of gene expression and genetic alteration studies. This study indicates that ELF3 functions not only as a part of normal prostate epithelial growth but also as a potential oncogene in advanced prostate cancers.

Correction: Brüning-Richardson et al. GSK-3 Inhibition Is Cytotoxic in Glioma Stem Cells through Centrosome Destabilization and Enhances the Effect of Radiotherapy in Orthotopic Models. Cancers 2021, 13, 5939.

The authors wish to make the following corrections to this paper [1][...].

The calcium sensor STIM1 is regulated by androgens in prostate stromal cells.

BACKGROUND: Prostate development and maintenance in the adult results from an interaction of stromal and glandular components. Androgens can drive this process by direct action on the stroma. We investigated whether there was a direct link between androgens and another key regulator of stromal cells, intracellular Ca2+ ([Ca2+ ]i ). METHODS: Prostate stromal cells were freshly obtained and cultures derived from patients with benign prostatic hyperplasia. Gene expression in dihydrotestosterone treated and untreated cells was compared using Affymetrix gene expression arrays and Ca2+ regulated features were identified by Gene Ontology (GO). Changes in [Ca2+]i were determined in Fluo-4 loaded cells. Androgen regulation was confirmed by chromatin immunoprecipitaion. RESULTS: Stromal cell cultures were sorted for expression of integrin α1 ß1 , which enriched for cells expressing the androgen receptor (AR). We identified key functional categories, within the androgen-induced gene expression signature, focusing on genes involved in calcium signaling. From this analysis, stromal interaction molecule-1 (STIM1) was identified as a significantly differentially expressed gene with four relevant associated GO terms. DNA sequence analysis showed that the promoter region of STIM1 contained putative androgen response element sequences in which AR binding ability of STIM1 was confirmed. Androgens directly regulated STIM1 expression and STIM1 effects on store-operated calcium entry were inhibited by STIM1 knock-down. Reduced STIM1 expression in prostate stromal cells led to a reduction in basal Ca2+ levels, the amount of Ca2+ released by thapsigargin and a reduction in store filling following TG-induced store depletion. CONCLUSIONS: These results indicate that androgens modulate [Ca2+]i through the direct regulation of the STIM1 gene by AR binding to the STIM1 promoter.

Benchmarking pipelines for subclonal deconvolution of bulk tumour sequencing data.

Intratumour heterogeneity provides tumours with the ability to adapt and acquire treatment resistance. The development of more effective and personalised treatments for cancers, therefore, requires accurate characterisation of the clonal architecture of tumours, enabling evolutionary dynamics to be tracked. Many methods exist for achieving this from bulk tumour sequencing data, involving identifying mutations and performing subclonal deconvolution, but there is a lack of systematic benchmarking to inform researchers on which are most accurate, and how dataset characteristics impact performance. To address this, we use the most comprehensive tumour genome simulation tool available for such purposes to create 80 bulk tumour whole exome sequencing datasets of differing depths, tumour complexities, and purities, and use these to benchmark subclonal deconvolution pipelines. We conclude that i) tumour complexity does not impact accuracy, ii) increasing either purity or purity-corrected sequencing depth improves accuracy, and iii) the optimal pipeline consists of Mutect2, FACETS and PyClone-VI. We have made our benchmarking datasets publicly available for future use.

RNA sequencing and functional studies of patient-derived cells reveal that neurexin-1 and regulators of this pathway are associated with poor outcomes in Ewing sarcoma.

PURPOSE: The development of biomarkers and molecularly targeted therapies for patients with Ewing sarcoma (ES) in order to minimise morbidity and improve outcome is urgently needed. Here, we set out to isolate and characterise patient-derived ES primary cell cultures and daughter cancer stem-like cells (CSCs) to identify biomarkers of high-risk disease and candidate therapeutic targets. METHODS: Thirty-two patient-derived primary cultures were established from treatment-naïve tumours and primary ES-CSCs isolated from these cultures using functional methods. By RNA-sequencing we analysed the transcriptome of ES patient-derived cells (n = 24) and ES-CSCs (n = 11) to identify the most abundant and differentially expressed genes (DEGs). Expression of the top DEG(s) in ES-CSCs compared to ES cells was validated at both RNA and protein levels. The functional and prognostic potential of the most significant gene (neurexin-1) was investigated using knock-down studies and immunohistochemistry of two independent tumour cohorts. RESULTS: ES-CSCs were isolated from all primary cell cultures, consistent with the premise that ES is a CSC driven cancer. Transcriptional profiling confirmed that these cells were of mesenchymal origin, revealed novel cell surface targets for therapy that regulate cell-extracellular matrix interactions and identified candidate drivers of progression and relapse. High expression of neurexin-1 and low levels of regulators of its activity, APBA1 and NLGN4X, were associated with poor event-free and overall survival rates. Knock-down of neurexin-1 decreased viable cell numbers and spheroid formation. CONCLUSIONS: Genes that regulate extracellular interactions, including neurexin-1, are candidate therapeutic targets in ES. High levels of neurexin-1 at diagnosis are associated with poor outcome and identify patients with localised disease that will relapse. These patients could benefit from more intensive or novel treatment modalities. The prognostic significance of neurexin-1 should be validated independently.

Cut-like homeobox 1 (CUX1) tumor suppressor gene haploinsufficiency induces apoptosis evasion to sustain myeloid leukemia.

While oncogenes promote tumorigenesis, they also induce deleterious cellular stresses, such as apoptosis, that cancer cells must combat by coopting adaptive responses. Whether tumor suppressor gene haploinsufficiency leads to such phenomena and their mechanistic basis is unclear. Here, we demonstrate that elevated levels of the anti-apoptotic factor, CASP8 and FADD-like apoptosis regulator (CFLAR), promotes apoptosis evasion in acute myeloid leukemia (AML) cells haploinsufficient for the cut-like homeobox 1 (CUX1) transcription factor, whose loss is associated with dismal clinical prognosis. Genome-wide CRISPR/Cas9 screening identifies CFLAR as a selective, acquired vulnerability in CUX1-deficient AML, which can be mimicked therapeutically using inhibitor of apoptosis (IAP) antagonists in murine and human AML cells. Mechanistically, CUX1 deficiency directly alleviates CUX1 repression of the CFLAR promoter to drive CFLAR expression and leukemia survival. These data establish how haploinsufficiency of a tumor suppressor is sufficient to induce advantageous anti-apoptosis cell survival pathways and concurrently nominate CFLAR as potential therapeutic target in these poor-prognosis leukemias.

Analysis of CRISPR-Cas9 screens identifies genetic dependencies in melanoma.

Targeting the MAPK signaling pathway has transformed the treatment of metastatic melanoma. CRISPR-Cas9 genetic screens provide a genome-wide approach to uncover novel genetic dependencies that might serve as therapeutic targets. Here, we analyzed recently reported CRISPR-Cas9 screens comparing data from 28 melanoma cell lines and 313 cell lines of other tumor types in order to identify fitness genes related to melanoma. We found an average of 1,494 fitness genes in each melanoma cell line. We identified 33 genes, inactivation of which specifically reduced the fitness of melanoma. This set of tumor type-specific genes includes established melanoma fitness genes as well as many genes that have not previously been associated with melanoma growth. Several genes encode proteins that can be targeted using available inhibitors. We verified that genetic inactivation of DUSP4 and PPP2R2A reduces the proliferation of melanoma cells. DUSP4 encodes an inhibitor of ERK, suggesting that further activation of MAPK signaling activity through its loss is selectively deleterious to melanoma cells. Collectively, these data present a resource of genetic dependencies in melanoma that may be explored as potential therapeutic targets.

CRISPR activation screen in mice identifies novel membrane proteins enhancing pulmonary metastatic colonisation.

Melanoma represents ~5% of all cutaneous malignancies, yet accounts for the majority of skin cancer deaths due to its propensity to metastasise. To develop new therapies, novel target molecules must to be identified and the accessibility of cell surface proteins makes them attractive targets. Using CRISPR activation technology, we screened a library of guide RNAs targeting membrane protein-encoding genes to identify cell surface molecules whose upregulation enhances the metastatic pulmonary colonisation capabilities of tumour cells in vivo. We show that upregulated expression of the cell surface protein LRRN4CL led to increased pulmonary metastases in mice. Critically, LRRN4CL expression was elevated in melanoma patient samples, with high expression levels correlating with decreased survival. Collectively, our findings uncover an unappreciated role for LRRN4CL in the outcome of melanoma patients and identifies a potential therapeutic target and biomarker.