Building flexible and robust analysis frameworks for molecular subtyping of cancers.

Molecular subtyping is essential to infer tumor aggressiveness and predict prognosis. In practice, tumor profiling requires in-depth knowledge of bioinformatics tools involved in the processing and analysis of the generated data. Additionally, data incompatibility (e.g., microarray versus RNA sequencing data) and technical and uncharacterized biological variance between training and test data can pose challenges in classifying individual samples. In this article, we provide a roadmap for implementing bioinformatics frameworks for molecular profiling of human cancers in a clinical diagnostic setting. We describe a framework for integrating several methods for quality control, normalization, batch correction, classification and reporting, and develop a use case of the framework in breast cancer.

Expression and splicing mediate distinct biological signals.

BACKGROUND: Through alternative splicing, most human genes produce multiple isoforms in a cell-, tissue-, and disease-specific manner. Numerous studies show that alternative splicing is essential for development, diseases, and their treatments. Despite these important examples, the extent and biological relevance of splicing are currently unknown. RESULTS: To solve this problem, we developed pairedGSEA and used it to profile transcriptional changes in 100 representative RNA-seq datasets. Our systematic analysis demonstrates that changes in splicing, on average, contribute to 48.1% of the biological signal in expression analyses. Gene-set enrichment analysis furthermore indicates that expression and splicing both convey shared and distinct biological signals. CONCLUSIONS: These findings establish alternative splicing as a major regulator of the human condition and suggest that most contemporary RNA-seq studies likely miss out on critical biological insights. We anticipate our results will contribute to the transition from a gene-centric to an isoform-centric research paradigm.

Most protein domains exist as variants with distinct functions across cells, tissues and diseases.

Protein domains are the active subunits that provide proteins with specific functions through precise three-dimensional structures. Such domains facilitate most protein functions, including molecular interactions and signal transduction. Currently, these protein domains are described and analyzed as invariable molecular building blocks with fixed functions. Here, I show that most human protein domains exist as multiple distinct variants termed 'domain isotypes'. Domain isotypes are used in a cell, tissue and disease-specific manner and have surprisingly different 3D structures. Accordingly, domain isotypes, compared to each other, modulate or abolish the functionality of protein domains. These results challenge the current view of protein domains as invariable building blocks and have significant implications for both wet- and dry-lab workflows. The extensive use of protein domain isotypes within protein isoforms adds to the literature indicating we need to transition to an isoform-centric research paradigm.

The HOPX and BLBP landscape and gliogenic regions in developing human brain.

Outer radial glial cells (oRGs) give rise to neurons and glial cells and contribute to cell migration and expansion in developing neocortex. HOPX has been described as a marker of oRGs and possible actor in glioblastomas. Recent years' evidence points to spatiotemporal differences in brain development which may have implications for the classification of cell types in the central nervous system and understanding of a range of neurological diseases. Using the Human Embryonic/Fetal Biobank, Institute of Cellular and Molecular Medicine, Faculty of Health and Medical Sciences, University of Copenhagen, Denmark, HOPX and BLBP immunoexpression was investigated in developing frontal, parietal, temporal and occipital human neocortex, other cortical areas and brain stem regions to interrogate oRG and HOPX regional heterogeneity. Furthermore, usage of high-plex spatial profiling (Nanostring GeoMx® DSP) was tested on the same material. HOPX marked oRGs in several human developing brain regions as well as cells in known gliogenic areas but did not completely overlap with BLBP or GFAP. Interestingly, limbic structures (e.g. olfactory bulb, indusium griseum, entorhinal cortex, fimbria) showed more intense HOPX immunoreactivity than adjacent neocortex and in cerebellum and brain stem, HOPX and BLBP seemed to stain different cell populations in cerebellar cortex and corpus pontobulbare. DSP screening of corresponding regions indicated differences in cell type composition, vessel density and presence of apolipoproteins within and across regions and thereby confirming the importance of acknowledging time and place in developmental neuroscience.

Non-metabolic functions of phosphofructokinase-1 orchestrate tumor cellular invasion and genome maintenance under bevacizumab therapy.

BACKGROUND: Glioblastoma (GBM) is a highly lethal malignancy for which neoangiogenesis serves as a defining hallmark. The anti-VEGF antibody, bevacizumab, has been approved for the treatment of recurrent GBM, but resistance is universal. METHODS: We analyzed expression data of GBM patients treated with bevacizumab to discover potential resistance mechanisms. Patient-derived xenografts (PDXs) and cultures were interrogated for effects of phosphofructokinase-1, muscle isoform (PFKM) loss on tumor cell motility, migration, and invasion through genetic and pharmacologic targeting. RESULTS: We identified PFKM as a driver of bevacizumab resistance. PFKM functions dichotomize based on subcellular location: cytosolic PFKM interacted with KIF11, a tubular motor protein, to promote tumor invasion, whereas nuclear PFKM safeguarded genomic stability of tumor cells through interaction with NBS1. Leveraging differential transcriptional profiling, bupivacaine phenocopied genetic targeting of PFKM, and enhanced efficacy of bevacizumab in preclinical GBM models in vivo. CONCLUSION: PFKM drives novel molecular pathways in GBM, offering a translational path to a novel therapeutic paradigm.

Progression of the pluripotent epiblast depends upon the NMD factor UPF2.

Nonsense-mediated RNA decay (NMD) is a highly conserved RNA turnover pathway that degrades RNAs harboring in-frame stop codons in specific contexts. Loss of NMD factors leads to embryonic lethality in organisms spanning the phylogenetic scale, but the mechanism remains unknown. Here, we report that the core NMD factor, UPF2, is required for expansion of epiblast cells within the inner cell mass of mice in vivo. We identify NMD target mRNAs in mouse blastocysts - both canonical and alternatively processed mRNAs - including those encoding cell cycle arrest and apoptosis factors, raising the possibility that NMD is essential for embryonic cell proliferation and survival. In support, the inner cell mass of Upf2-null blastocysts rapidly regresses with outgrowth and is incompetent for embryonic stem cell derivation in vitro. In addition, we uncovered concordant temporal- and lineage-specific regulation of NMD factors and mRNA targets, indicative of a shift in NMD magnitude during peri-implantation development. Together, our results reveal developmental and molecular functions of the NMD pathway in the early embryo.

Target isoforms are an overlooked challenge and opportunity in chimeric antigen receptor cell therapy.

The development of novel chimeric antigen receptor (CAR) cell therapies is rapidly growing, with 299 new agents being reported and 109 new clinical trials initiated so far this year. One critical lesson from approved CD19-specific CAR therapies is that target isoform switching has been shown to cause tumour relapse, but little is known about the isoforms of CAR targets in solid cancers. Here we assess the protein isoform landscape and identify both the challenges and opportunities protein isoform switching present as CAR therapy is applied to solid cancers.

PTBP1 promotes hematopoietic stem cell maintenance and red blood cell development by ensuring sufficient availability of ribosomal constituents.

Ribosomopathies constitute a range of disorders associated with defective protein synthesis mainly affecting hematopoietic stem cells (HSCs) and erythroid development. Here, we demonstrate that deletion of poly-pyrimidine-tract-binding protein 1 (PTBP1) in the hematopoietic compartment leads to the development of a ribosomopathy-like condition. Specifically, loss of PTBP1 is associated with decreases in HSC self-renewal, erythroid differentiation, and protein synthesis. Consistent with its function as a splicing regulator, PTBP1 deficiency results in splicing defects in hundreds of genes, and we demonstrate that the up-regulation of a specific isoform of CDC42 partly mimics the protein-synthesis defect associated with loss of PTBP1. Furthermore, PTBP1 deficiency is associated with a marked defect in ribosome biogenesis and a selective reduction in the translation of mRNAs encoding ribosomal proteins. Collectively, this work identifies PTBP1 as a key integrator of ribosomal functions and highlights the broad functional repertoire of RNA-binding proteins.

Plasma IL-8 and ICOSLG as prognostic biomarkers in glioblastoma.

BACKGROUND: CNS immune privilege has been challenged in recent years. Glioblastoma (GBM) immune dysfunction includes complex interactions with the immune system outside the CNS. The aim of this study was to determine diagnostic and prognostic potential of immune-related proteins in plasma in GBM and interrogate biomarker presence in the brain tumor microenvironment (TME). METHODS: One hundred and fifty-eight patients with glioma WHO grade II-IV were included. Plasma collected at surgery was screened for 92 proteins using proximity extension assay technology and related to clinical outcome. Secretion and expression of candidate prognostic biomarkers were subsequently analyzed in 8 GBM cell lines and public RNAseq data. RESULTS: Plasma levels of 20 out of 92 screened proteins were significantly different in patients with GBM compared to patients with astrocytoma WHO grade II-III. High plasma interleukin-8 (IL-8) (hazard ratio [HR] = 1.52; P = .0077) and low CD244 (HR = 0.36; P = .0004) were associated with short progression-free survival and high plasma IL-8 (HR = 1.40; P = .044) and low ICOS ligand (ICOSLG) (HR = 0.17; P = .0003) were associated with short overall survival (OS) in newly diagnosed patients with GBM. A similar trend was found for ICOSLG (HR = 0.34; P = .053) in recurrent GBM. IL-8 was mostly secreted and expressed by mesenchymal GBM cell lines and expressed by vascular cells and immune cells in the TME. This was also the case for ICOSLG, although less consistent, and with additional expression in tumor-associated oligodendrocytes. CONCLUSIONS: High plasma IL-8 and low ICOSLG at surgery are associated with short OS in newly diagnosed GBM. Source of plasma ICOSLG may be found outside the TME.

Perspective: targeting VEGF-A and YKL-40 in glioblastoma - matter matters.

Glioblastomas (GBM) are heterogeneous highly vascular brain tumors exploiting the unique microenvironment in the brain to resist treatment and anti-tumor responses. Anti-angiogenic agents, immunotherapy, and targeted therapy have been studied extensively in GBM patients over a number of decades with minimal success. Despite maximal efforts, prognosis remains dismal with an overall survival of approximately 15 months.Bevacizumab, a humanized anti-vascular endothelial growth factor (VEGF) antibody, underwent accelerated approval by the U.S. Food and Drug Administration in 2009 for the treatment of recurrent GBM based on promising preclinical and early clinical studies. Unfortunately, subsequent clinical trials did not find overall survival benefit. Pursuing pleiotropic targets and leaning toward multitarget strategies may be a key to more effective therapeutic intervention in GBM, but preclinical evaluation requires careful consideration of model choices. In this study, we discuss bevacizumab resistance, dual targeting of pro-angiogenic modulators VEGF and YKL-40 in the context of brain tumor microenvironment, and how model choice impacts study conclusions and its translational significance.

satuRn: Scalable analysis of differential transcript usage for bulk and single-cell RNA-sequencing applications.

Alternative splicing produces multiple functional transcripts from a single gene. Dysregulation of splicing is known to be associated with disease and as a hallmark of cancer. Existing tools for differential transcript usage (DTU) analysis either lack in performance, cannot account for complex experimental designs or do not scale to massive single-cell transcriptome sequencing (scRNA-seq) datasets. We introduce satuRn, a fast and flexible quasi-binomial generalized linear modelling framework that is on par with the best performing DTU methods from the bulk RNA-seq realm, while providing good false discovery rate control, addressing complex experimental designs, and scaling to scRNA-seq applications.

The RNA m6A Reader YTHDF2 Maintains Oncogene Expression and Is a Targetable Dependency in Glioblastoma Stem Cells.

Glioblastoma is a universally lethal cancer driven by glioblastoma stem cells (GSC). Here, we interrogated N 6-methyladenosine (m6A) mRNA modifications in GSCs by methyl RNA immunoprecipitation followed by sequencing and transcriptome analysis, finding transcripts marked by m6A often upregulated compared with normal neural stem cells (NSC). Interrogating m6A regulators, GSCs displayed preferential expression, as well as in vitro and in vivo dependency, of the m6A reader YTHDF2, in contrast to NSCs. Although YTHDF2 has been reported to destabilize mRNAs, YTHDF2 stabilized MYC and VEGFA transcripts in GSCs in an m6A-dependent manner. We identified IGFBP3 as a downstream effector of the YTHDF2-MYC axis in GSCs. The IGF1/IGF1R inhibitor linsitinib preferentially targeted YTHDF2-expressing cells, inhibiting GSC viability without affecting NSCs and impairing in vivo glioblastoma growth. Thus, YTHDF2 links RNA epitranscriptomic modifications and GSC growth, laying the foundation for the YTHDF2-MYC-IGFBP3 axis as a specific and novel therapeutic target in glioblastoma. SIGNIFICANCE: Epitranscriptomics promotes cellular heterogeneity in cancer. RNA m6A landscapes of cancer and NSCs identified cell type-specific dependencies and therapeutic vulnerabilities. The m6A reader YTHDF2 stabilized MYC mRNA specifically in cancer stem cells. Given the challenge of targeting MYC, YTHDF2 presents a therapeutic target to perturb MYC signaling in glioblastoma.This article is highlighted in the In This Issue feature, p. 211.

Personalized B cell response to the Lactobacillus rhamnosus GG probiotic in healthy human subjects: a randomized trial.

The specific effects of administering live probiotics in the human gut are not well characterized. To this end, we investigated the immediate effect of Lactobacillus rhamnosus GG (LGG) in the jejunum of 27 healthy volunteers 2 h after ingestion using a combination of global RNA sequencing of human biopsies and bacterial DNA sequencing in a multi-visit, randomized, cross-over design (ClinicalTrials.gov number NCT03140878). While LGG was detectable in jejunum after 2 h in treated subjects, the gene expression response vs. placebo was subtle if assessed across all subjects. However, clustering analysis revealed that one-third of subjects exhibited a strong and consistent LGG response involving hundreds of genes, where genes related to B cell activation were upregulated, consistent with prior results in mice. Immunohistochemistry and single cell-based deconvolution analyses showed that this B cell signature likely is due to activation and proliferation of existing B cells rather than B cell immigration to the tissue. Our results indicate that the LGG strain has an immediate effect in the human gut in a subpopulation of individuals. In extension, our data strongly suggest that studies on in vivo probiotic effects in humans require large cohorts and must take individual variation into account.

A Functional Link between Nuclear RNA Decay and Transcriptional Control Mediated by the Polycomb Repressive Complex 2.

Pluripotent embryonic stem cells (ESCs) constitute an essential cellular niche sustained by epigenomic and transcriptional regulation. Any role of post-transcriptional processes remains less explored. Here, we identify a link between nuclear RNA levels, regulated by the poly(A) RNA exosome targeting (PAXT) connection, and transcriptional control by the polycomb repressive complex 2 (PRC2). Knockout of the PAXT component ZFC3H1 impairs mouse ESC differentiation. In addition to the upregulation of bona fide PAXT substrates, Zfc3h1-/- cells abnormally express developmental genes usually repressed by PRC2. Such de-repression is paralleled by decreased PRC2 binding to chromatin and low PRC2-directed H3K27 methylation. PRC2 complex stability is compromised in Zfc3h1-/- cells with elevated levels of unspecific RNA bound to PRC2 components. We propose that excess RNA hampers PRC2 function through its sequestration from DNA. Our results highlight the importance of balancing nuclear RNA levels and demonstrate the capacity of bulk RNA to regulate chromatin-associated proteins.

CAGEfightR: analysis of 5'-end data using R/Bioconductor.

BACKGROUND: 5'-end sequencing assays, and Cap Analysis of Gene Expression (CAGE) in particular, have been instrumental in studying transcriptional regulation. 5'-end methods provide genome-wide maps of transcription start sites (TSSs) with base pair resolution. Because active enhancers often feature bidirectional TSSs, such data can also be used to predict enhancer candidates. The current availability of mature and comprehensive computational tools for the analysis of 5'-end data is limited, preventing efficient analysis of new and existing 5'-end data. RESULTS: We present CAGEfightR, a framework for analysis of CAGE and other 5'-end data implemented as an R/Bioconductor-package. CAGEfightR can import data from BigWig files and allows for fast and memory efficient prediction and analysis of TSSs and enhancers. Downstream analyses include quantification, normalization, annotation with transcript and gene models, TSS shape statistics, linking TSSs to enhancers via co-expression, identification of enhancer clusters, and genome-browser style visualization. While built to analyze CAGE data, we demonstrate the utility of CAGEfightR in analyzing nascent RNA 5'-data (PRO-Cap). CAGEfightR is implemented using standard Bioconductor classes, making it easy to learn, use and combine with other Bioconductor packages, for example popular differential expression tools such as limma, DESeq2 and edgeR. CONCLUSIONS: CAGEfightR provides a single, scalable and easy-to-use framework for comprehensive downstream analysis of 5'-end data. CAGEfightR is designed to be interoperable with other Bioconductor packages, thereby unlocking hundreds of mature transcriptomic analysis tools for 5'-end data. CAGEfightR is freely available via Bioconductor: bioconductor.org/packages/CAGEfightR .

Mutant CEBPA directly drives the expression of the targetable tumor-promoting factor CD73 in AML.

The key myeloid transcription factor (TF), CEBPA, is frequently mutated in acute myeloid leukemia (AML), but the direct molecular effects of this leukemic driver mutation remain elusive. To investigate CEBPA mutant AML, we performed microscale, in vivo chromatin immunoprecipitation sequencing and identified a set of aberrantly activated enhancers, exclusively occupied by the leukemia-associated CEBPA-p30 isoform. Comparing gene expression changes in human CEBPA mutant AML and the corresponding Cebpa Lp30 mouse model, we identified Nt5e, encoding CD73, as a cross-species AML gene with an upstream leukemic enhancer physically and functionally linked to the gene. Increased expression of CD73, mediated by the CEBPA-p30 isoform, sustained leukemic growth via the CD73/A2AR axis. Notably, targeting of this pathway enhanced survival of AML-transplanted mice. Our data thus indicate a first-in-class link between a cancer driver mutation in a TF and a druggable, direct transcriptional target.

IsoformSwitchAnalyzeR: analysis of changes in genome-wide patterns of alternative splicing and its functional consequences.

SUMMARY: Alternative splicing is an important mechanism involved in health and disease. Recent work highlights the importance of investigating genome-wide changes in splicing patterns and the subsequent functional consequences. Current computational methods only support such analysis on a gene-by-gene basis. Therefore, we extended IsoformSwitchAnalyzeR R library to enable analysis of genome-wide changes in specific types of alternative splicing and predicted functional consequences of the resulting isoform switches. As a case study, we analyzed RNA-seq data from The Cancer Genome Atlas and found systematic changes in alternative splicing and the consequences of the associated isoform switches. AVAILABILITY AND IMPLEMENTATION: Windows, Linux and Mac OS: http://bioconductor.org/packages/IsoformSwitchAnalyzeR. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Astrogliogenesis in human fetal brain: complex spatiotemporal immunoreactivity patterns of GFAP, S100, AQP4 and YKL-40.

The astroglial lineage consists of heterogeneous cells instrumental for normal brain development, function and repair. Unfortunately, this heterogeneity complicates research in the field, which suffers from lack of truly specific and sensitive astroglial markers. Nevertheless, single astroglial markers are often used to describe astrocytes in different settings. We therefore investigated and compared spatiotemporal patterns of immunoreactivity in developing human brain from 12 to 21 weeks post conception and publicly available RNA expression data for four established and potential astroglial markers - GFAP, S100, AQP4 and YKL-40. In the hippocampal region, we also screened for C3, a complement component highly expressed in A1-reactive astrocytes. We found diverging partly overlapping patterns of the established astroglial markers GFAP, S100 and AQP4, confirming that none of these markers can fully describe and discriminate different developmental forms and subpopulations of astrocytes in human developing brain, although AQP4 seems to be the most sensitive and specific marker for the astroglial lineage at midgestation. AQP4 characterizes a brain-wide water transport system in cerebral cortex with regional differences in immunoreactivity at midgestation. AQP4 distinguishes a vast proportion of astrocytes and subpopulations of radial glial cells destined for the astroglial lineage, including astrocytes determined for the future glia limitans and apical truncated radial glial cells in ganglionic eminences, devoid of GFAP and S100. YKL-40 and C3d, previously found in reactive astrocytes, stain different subpopulations of astrocytes/astroglial progenitors in developing hippocampus at midgestation and may characterize specific subpopulations of 'developmental astrocytes'. Our results clearly reflect that lack of pan-astrocytic markers necessitates the consideration of time, region, context and aim when choosing appropriate astroglial markers.

Characterization of the enhancer and promoter landscape of inflammatory bowel disease from human colon biopsies.

Inflammatory bowel disease (IBD) is a chronic intestinal disorder, with two main types: Crohn's disease (CD) and ulcerative colitis (UC), whose molecular pathology is not well understood. The majority of IBD-associated SNPs are located in non-coding regions and are hard to characterize since regulatory regions in IBD are not known. Here we profile transcription start sites (TSSs) and enhancers in the descending colon of 94 IBD patients and controls. IBD-upregulated promoters and enhancers are highly enriched for IBD-associated SNPs and are bound by the same transcription factors. IBD-specific TSSs are associated to genes with roles in both inflammatory cascades and gut epithelia while TSSs distinguishing UC and CD are associated to gut epithelia functions. We find that as few as 35 TSSs can distinguish active CD, UC, and controls with 85% accuracy in an independent cohort. Our data constitute a foundation for understanding the molecular pathology, gene regulation, and genetics of IBD.