Mouse Model of Fragile X Syndrome Analyzed by Quantitative Proteomics: A Comparison of Methods

Multiple methods for quantitative proteomics are available for proteome profiling. It is unclear which methods are most useful in situations involving deep proteome profiling and the detection of subtle distortions in the proteome. Here, we compared the performance of seven different strategies in the analysis of a mouse model of Fragile X Syndrome, involving the knockout of the fmr1 gene that is the leading cause of autism spectrum disorder. Focusing on the cerebellum, we show that data-independent acquisition (DIA) and the tandem mass tag (TMT)-based real-time search method (RTS) generated the most informative profiles, generating 334 and 329 significantly altered proteins, respectively, although the latter still suffered from ratio compression. Label-free methods such as BoxCar and a conventional data-dependent acquisition were too noisy to generate a reliable profile, while TMT methods that do not invoke RTS showed a suppressed dynamic range. The TMT method using the TMTpro reagents together with complementary ion quantification (ProC) overcomes ratio compression, but current limitations in ion detection reduce sensitivity. Overall, both DIA and RTS uncovered known regulators of the syndrome and detected alterations in calcium signaling pathways that are consistent with calcium deregulation recently observed in imaging studies. Data are available via ProteomeXchange with the identifier PXD039885.

The proteomic landscape of glioblastoma recurrence reveals novel and targetable immunoregulatory drivers

Glioblastoma (GBM) is characterized by extensive cellular and genetic heterogeneity. Its initial presentation as primary disease (pGBM) has been subject to exhaustive molecular and cellular profiling. By contrast, our understanding of how GBM evolves to evade the selective pressure of therapy is starkly limited. The proteomic landscape of recurrent GBM (rGBM), which is refractory to most treatments used for pGBM, are poorly known. We, therefore, quantified the transcriptome and proteome of 134 patient-derived pGBM and rGBM samples, including 40 matched pGBM-rGBM pairs. GBM subtypes transition from pGBM to rGBM towards a preferentially mesenchymal state at recurrence, consistent with the increasingly invasive nature of rGBM. We identified immune regulatory/suppressive genes as important drivers of rGBM and in particular 2-5-oligoadenylate synthase 2 (OAS2) as an essential gene in recurrent disease. Our data identify a new class of therapeutic targets that emerge from the adaptive response of pGBM to therapy, emerging specifically in recurrent disease and may provide new therapeutic opportunities absent at pGBM diagnosis.

MODL-16. SOMATICALLY ENGINEERED MOUSE MODELS RECAPITULATING CELLULAR AND MOLECULAR FEATURES OF HUMAN GBM

INTRODUCTION

Mouse models are instrumental in advancing our understanding of glioblastoma (GBM), however many commonly used models have features that limit their practicality or applicability. These limitations can include an incomplete immune context, a requirement for complex breeding strategies, high cost of specialized lines, and unpredictability in where tumors arise. We sought to develop and validate a mouse modeling system for some of the major human GBM subtypes that overcame these practical challenges.

METHODS

In vivo electroporation was used to introduce genetic alterations into periventricular cells of early postnatal C57Bl6/j mice. PiggyBac transposon/transposase and CRISPR-Cas9 systems were used to overexpress (OE) or knock out (KO) genes associated with mesenchymal (Nf1-KO/Pten-KO/p53-KO), classical (EGFRvIII-OE/Cdkn2a-KO/Pten-KO), and proneural (Pdgfra-OE/Cdk4-OE/p53-KO) IDH-WT GBM subtypes. KO or OE was confirmed by Indel Detection by Amplicon Analysis and immunofluorescent staining (IF). Tumours were analyzed for histologic, immunohistochemical, and transcriptional (RNAseq) features.

RESULTS

Tumors arose with near complete penetrance and median survivals of 36 to 116 days amongst models. All tumors were GFAP-expressing high-grade gliomas, with distinct phenotypes associated with different genetic combinations. Nf1-KO/Pten-KO/p53-KO tumors were composed of spindle cells. EGFRvIII-OE/Cdkn2a-KO/Pten-KO and Pdgfra-OE/Cdk4-OE/p53-KO tumors showed prevalence of small, round cells. Molecularly, Nf1-KO/Pten-KO/p53-KO tumors were enriched for human mesenchymal signature and displayed more differentiated phenotype. The Nf1-KO/Pten-KO/p53-KO model also showed increased stromal cells and macrophages. The other models had mixed proneural and classical profiles and were enriched for OPC- and NPC-like gene signatures found in human GBM. Once tumors formed, cells from each model were transplantable into C57Bl6/j mice to generate subsequent tumors.

CONCLUSION

We developed and validated a rapid, versatile, and reproducible system to model GBMs. These models allow for controlled study of GBM pathogenesis, progression, and treatment response, and allow for robust generation of syngeneic implantable mouse models that can serve as valuable tools for preclinical testing.

Abstract 6397: Understanding the glioblastoma microenvironment with spatial resolution in PDX models

Glioblastoma multiforme (GBM) is the most common adult brain tumour, and despite aggressive treatment, it recurs fatally. GBM tumours include diverse populations of malignant and non-neoplastic cells with distinct molecular capabilities and with differential levels of sensitivity to treatment. Understanding the cell dynamics that occur during the development of GBM resistance to therapy could reveal key aspects of this process, including how resistance is acquired in time and how the diverse cell types of the tumour microenvironment (TME) contribute to this phenotype. In addition to the role of the TME, GBM exhibits significant tumour heterogeneity with diverse genetic clones coexisting in the same tumor, as well as cells with similar genetic backgrounds capable of adopting distinct transcriptional states and subtypes. The complex and dynamic interactions between tumor and TME remain to be fully studied. This work focuses on the in vivo spatial organization in GBM during disease progression. We generated spatial transcriptomic data from a set of adult GBM samples grown as patient-derived xenograft (PDX) models, profiled at different time points of the disease. Three PDX lines from one GBM patient (derived from tumor core, vascularized area, and infiltrating front) were used to recapitulate the genetic and phenotypic heterogeneity observed in the human disease. Two replicates from each of 8 PDX mice were collected from early, mid, and late time points of tumour growth and data was generated using the 10X Genomics Visium platform. We developed a robust computational pipeline capable of distinguishing admixture of human (tumour) and mouse cells (TME), using state-of-the-art tools. Human and mouse cell types and states were identified using pooled and separate single-cell references of human GBM states, and mouse brain cells from both normal and tumour conditions. With this approach we observe spatially distinct patterns of both (a) tumour infiltration patterns specific to the each PDX line that includes non-random distribution of GBM transcriptional states and genetic clones, and (b) spatially distinct infiltration of TME components including microglial and macrophage populations. Overall, our approach addresses the challenge of understanding the tumor-TME relationship by application of spatial profiling in PDX models, and provides a computational pipeline complex multi-species analysis in the spatial transcriptomic field.

Citation Format: Aly O. Abdelkareem, Katalin Osz, Donna Senger, Jennifer A. Chan, Sorana Morrissy. Understanding the glioblastoma microenvironment with spatial resolution in PDX models [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6397.

Copy-scAT: Deconvoluting single-cell chromatin accessibility of genetic subclones in cancer

Single-cell epigenomic assays have tremendous potential to illuminate mechanisms of transcriptional control in functionally diverse cancer cell populations. However, application of these techniques to clinical tumor specimens has been hampered by the current inability to distinguish malignant from nonmalignant cells, which potently confounds data analysis and interpretation. Here, we describe Copy-scAT, an R package that uses single-cell epigenomic data to infer copy number variants (CNVs) that define cancer cells. Copy-scAT enables studies of subclonal chromatin dynamics in complex tumors like glioblastoma. By deploying Copy-scAT, we uncovered potent influences of genetics on chromatin accessibility profiles in individual subclones. Consequently, some genetic subclones were predisposed to acquire stem-like or more differentiated molecular phenotypes, reminiscent of developmental paradigms. Copy-scAT is ideal for studies of the relationships between genetics and epigenetics in malignancies with high levels of intratumoral heterogeneity and to investigate how cancer cells interface with their microenvironment.

Histology-based molecular profiling improves mutation detection for advanced thyroid cancer

Advanced cancers frequently show histologic and molecular intratumoral heterogeneity. Therefore, we comprehensively characterized advanced, metastatic, radioiodine-resistant (RAIR) thyroid carcinomas at the molecular level in the context of histologic heterogeneity with the aim to identify potentially actionable mutations that may guide the use of specific tyrosine kinase inhibitor (TKI) treatment. Whole exome sequencing (WES) was applied to 29 macrodissected tissue samples of histologically heterogeneous and homogeneous areas, lymph node and lung metastases from six clinically and histologically well-characterized metastatic RAIR thyroid cancer patients with structural incomplete response to treatment. WES data were analyzed to identify potential driver mutations in oncogenic pathways, copy number alterations, microsatellite instability, mutant-allele tumor heterogeneity, and the relevance of histologic heterogeneity to molecular profiling. In addition to known driver mutations in BRAF, NRAS, EIF1AX, NCOA4-RET, and TERT, further potentially actionable drivers were identified in AKT1, ATM, E2F1, HTR2A, and MLH3. The analysis of the evolutionary history of the mutations and the reconstruction of the molecular phylogeny of the cancers show a remarkable association between histologic and molecular heterogeneity. A comprehensive molecular analysis of the primary tumor guided by histologic analysis may help to better stratify patients for precision medicine approaches. Given the association between the molecular and the histologic heterogeneity, the selection of tumor samples for molecular analysis should be based on meticulous histologic evaluation of the entire tumor.

Profiling Chromatin Accessibility at Single-cell Resolution

How distinct transcriptional programs are enacted to generate cellular heterogeneity and plasticity, and enable complex fate decisions are important open questions. One key regulator is the cell’s epigenome state that drives distinct transcriptional programs by regulating chromatin accessibility. Genome-wide chromatin accessibility measurements can impart insights into regulatory sequences (in)accessible to DNA-binding proteins at a single-cell resolution. This review outlines molecular methods and bioinformatic tools for capturing cell-to-cell chromatin variation using single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq) in a scalable fashion. It also covers joint profiling of chromatin with transcriptome/proteome measurements, computational strategies to integrate multi-omic measurements, and predictive bioinformatic tools to infer chromatin accessibility from single-cell transcriptomic datasets. Methodological refinements that increase power for cell discovery through robust chromatin coverage and integrate measurements from multiple modalities will further expand our understanding of gene regulation during homeostasis and disease.

High-resolution structural genomics reveals new therapeutic vulnerabilities in glioblastoma

We investigated the role of 3D genome architecture in instructing functional properties of glioblastoma stem cells (GSCs) by generating sub-5-kb resolution 3D genome maps by in situ Hi-C. Contact maps at sub-5-kb resolution allow identification of individual DNA loops, domain organization, and large-scale genome compartmentalization. We observed differences in looping architectures among GSCs from different patients, suggesting that 3D genome architecture is a further layer of inter-patient heterogeneity for glioblastoma. Integration of DNA contact maps with chromatin and transcriptional profiles identified specific mechanisms of gene regulation, including the convergence of multiple super enhancers to individual stemness genes within individual cells. We show that the number of loops contacting a gene correlates with elevated transcription. These results indicate that stemness genes are hubs of interaction between multiple regulatory regions, likely to ensure their sustained expression. Regions of open chromatin common among the GSCs tested were poised for expression of immune-related genes, including CD276. We demonstrate that this gene is co-expressed with stemness genes in GSCs and that CD276 can be targeted with an antibody-drug conjugate to eliminate self-renewing cells. Our results demonstrate that integrated structural genomics data sets can be employed to rationally identify therapeutic vulnerabilities in self-renewing cells.

Dual Regulatory Functions of SUFU and Targetome of GLI2 in SHH Subgroup Medulloblastoma

SUFU alterations are common in human Sonic Hedgehog (SHH) subgroup medulloblastoma (MB). However, its tumorigenic mechanisms have remained elusive. Here, we report that loss of Sufu alone is unable to induce MB formation in mice, due to insufficient Gli2 activation. Simultaneous loss of Spop, an E3 ubiquitin ligase targeting Gli2, restores robust Gli2 activation and induces rapid MB formation in Sufu knockout background. We also demonstrated a tumor-promoting role of Sufu in Smo-activated MB (∼60% of human SHH MB) by maintaining robust Gli activity. Having established Gli2 activation as a key driver of SHH MB, we report a comprehensive analysis of its targetome. Furthermore, we identified Atoh1 as a target and molecular accomplice of Gli2 that activates core SHH MB signature genes in a synergistic manner. Overall, our work establishes the dual role of SUFU in SHH MB and provides mechanistic insights into transcriptional regulation underlying Gli2-mediated SHH MB tumorigenesis.

Opposing Effects of CREBBP Mutations Govern the Phenotype of Rubinstein-Taybi Syndrome and Adult SHH Medulloblastoma

Recurrent mutations in chromatin modifiers are specifically prevalent in adolescent or adult patients with Sonic hedgehog-associated medulloblastoma (SHH MB). Here, we report that mutations in the acetyltransferase CREBBP have opposing effects during the development of the cerebellum, the primary site of origin of SHH MB. Our data reveal that loss of Crebbp in cerebellar granule neuron progenitors (GNPs) during embryonic development of mice compromises GNP development, in part by downregulation of brain-derived neurotrophic factor (Bdnf). Interestingly, concomitant cerebellar hypoplasia was also observed in patients with Rubinstein-Taybi syndrome, a congenital disorder caused by germline mutations of CREBBP. By contrast, loss of Crebbp in GNPs during postnatal development synergizes with oncogenic activation of SHH signaling to drive MB growth, thereby explaining the enrichment of somatic CREBBP mutations in SHH MB of adult patients. Together, our data provide insights into time-sensitive consequences of CREBBP mutations and corresponding associations with human diseases.

Medulloblastoma-associated DDX3 variant selectively alters the translational response to stress

DDX3X encodes a DEAD-box family RNA helicase (DDX3) commonly mutated in medulloblastoma, a highly aggressive cerebellar tumor affecting both children and adults. Despite being implicated in several facets of RNA metabolism, the nature and scope of DDX3′s interactions with RNA remain unclear. Here, we show DDX3 collaborates extensively with the translation initiation machinery through direct binding to 5′UTRs of nearly all coding RNAs, specific sites on the 18S rRNA, and multiple components of the translation initiation complex. Impairment of translation initiation is also evident in primary medulloblastomas harboring mutations in DDX3X, further highlighting DDX3′s role in this process. Arsenite-induced stress shifts DDX3 binding from the 5′UTR into the coding region of mRNAs concomitant with a general reduction of translation, and both the shift of DDX3 on mRNA and decreased translation are blunted by expression of a catalytically-impaired, medulloblastoma-associated DDX3^R534H variant. Furthermore, despite the global repression of translation induced by arsenite, translation is preserved on select genes involved in chromatin organization in DDX3^R534H-expressing cells. Thus, DDX3 interacts extensively with RNA and ribosomal machinery to help remodel the translation landscape in response to stress, while cancer-related DDX3 variants adapt this response to selectively preserve translation.

Identification of GPC2 as an Oncoprotein and Candidate Immunotherapeutic Target in High-Risk Neuroblastoma

We developed an RNA-sequencing-based pipeline to discover differentially expressed cell-surface molecules in neuroblastoma that meet criteria for optimal immunotherapeutic target safety and efficacy. Here, we show that GPC2 is a strong candidate immunotherapeutic target in this childhood cancer. We demonstrate high GPC2 expression in neuroblastoma due to MYCN transcriptional activation and/or somatic gain of the GPC2 locus. We confirm GPC2 to be highly expressed on most neuroblastomas, but not detectable at appreciable levels in normal childhood tissues. In addition, we demonstrate that GPC2 is required for neuroblastoma proliferation. Finally, we develop a GPC2-directed antibody-drug conjugate that is potently cytotoxic to GPC2-expressing neuroblastoma cells. Collectively, these findings validate GPC2 as a non-mutated neuroblastoma oncoprotein and candidate immunotherapeutic target.

Abstract 685: GPC2 is an oncogene and immunotherapeutic target in high-risk neuroblastoma

Background: GD2-directed immunotherapeutic strategies have improved outcomes in neuroblastoma; however, the majority of patients treated suffer relapse and GD2 expression on pain fibers causes dose-limiting toxicities.

Methods: To identify alternative cell surface immunotherapeutic targets, we compared high-risk neuroblastoma (n=126 tumors) and normal tissue RNA sequencing data (GTEx; n=7859 samples from 31 normal tissues) and prioritized genes by differential and absolute expression and cell surface prediction. Genes were further surveyed for somatic copy number gain and correlative expression with MYCN amplification. Differential protein expression and localization were confirmed in neuroblastoma primary tumors (n=98), patient-derived xenografts (n=32; PDXs), cell lines (n=23), and normal pediatric tissues (n=36). Cell lines were subjected to candidate gene gain and loss of function studies (n=11). Additional pediatric tumor RNA sequencing data was surveyed followed by confirmatory immunohistochemistry (IHC). Finally, candidate specific antibodies were isolated from a human Fab phage library and utilized for antibody-drug conjugate (ADC) engineering followed by cytotoxicity studies.

Results: We identified 33 differentially expressed cell surface molecules from which we prioritized glypican-2 (GPC2) for validation given GPC2’s robust differential expression (log-fold change tumor vs. normal tissue = 1.71-9.22; p=1.99 x 10-9-1.88 x10-300), high-level absolute RNA expression (median FPKM=60), and frequent DNA copy number gain associated with higher GPC2 expression (35%, n=182 tumors; p<0.005). GPC2 expression was also higher in MYCN amplified neuroblastomas (p<0.05), MYCN binds the GPC2 promoter shown by chromatin immunoprecipitation (ChIP) sequencing and reporter assays, and MYCN depletion resulted in decreased GPC2 expression. Immunoblot, flow cytometry, immunofluorescence, and IHC analysis of primary tumors, PDXs, and cell lines confirmed dense cell surface GPC2 expression. Medulloblastomas (n=62) were also found to have high GPC2 expression that positively correlated with MYC, MYCN, and GPC2 loci gain (p<0.0001). Pediatric normal tissues had very restricted cell surface GPC2 expression, with only low levels found in the esophagus and skin. GPC2 depletion in neuroblastoma cell lines resulted in apoptosis and growth inhibition and GPC2 forced over-expression increased neuroblastoma cell proliferation (p<0.001 for all assays). Finally, a human GPC2 antibody, D3-GPC2-Fab, was developed and shown to bind GPC2 with high affinity and specificity. D3-GPC2-IgG1 induced internalization of GPC2 and was conjugated to pyrrolobenzodiazepine (PBD) dimers to form an ADC which induced potent and specific cytotoxicity to GPC2 expressing neuroblastoma cells (IC50 = 1.7-11 pM).

Conclusions: GPC2 is an oncogene and immunotherapeutic target in neuroblastoma and potentially other cancers.

Citation Format: Kristopher R. Bosse, Pichai Raman, Maria Lane, Robyn T. Sussman, Jo Lynne Harenza, Daniel Martinez, Sabine Heitzeneder, Zhongyu Zhu, Komal Rathi, Michael Randall, Laura Donovan, Sorana Morrissy, Doncho V. Zhelev, Yang Feng, Jennifer Hwang, Yanping Wang, Bruce Pawel, Tricia Bhatti, Mariarita Santi, Javed Khan, Michael Taylor, Dimiter S. Dimitrov, Crystal Mackall, John M. Maris. GPC2 is an oncogene and immunotherapeutic target in high-risk neuroblastoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 685. doi:10.1158/1538-7445.AM2017-685

Highlights of Children with Cancer UK's Workshop on Drug Delivery in Paediatric Brain Tumours

The first Workshop on Drug Delivery in Paediatric Brain Tumours was hosted in London by the charity Children with Cancer UK. The goals of the workshop were to break down the barriers to treating central nervous system (CNS) tumours in children, leading to new collaborations and further innovations in this under-represented and emotive field. These barriers include the physical delivery challenges presented by the blood-brain barrier, the underpinning reasons for the intractability of CNS cancers, and the practical difficulties of delivering cancer treatment to the brains of children. Novel techniques for overcoming these problems were discussed, new models brought forth, and experiences compared.

HDAC and PI3K Antagonists Cooperate to Inhibit Growth of MYC-Driven Medulloblastoma

Medulloblastoma (MB) is a highly malignant pediatric brain tumor. Despite aggressive therapy, many patients succumb to the disease, and survivors experience severe side effects from treatment. MYC-driven MB has a particularly poor prognosis and would greatly benefit from more effective therapies. We used an animal model of MYC-driven MB to screen for drugs that decrease viability of tumor cells. Among the most effective compounds were histone deacetylase inhibitors (HDACIs). HDACIs potently inhibit survival of MYC-driven MB cells in vitro, in part by inducing expression of the FOXO1 tumor suppressor gene. HDACIs also synergize with phosphatidylinositol 3-kinase inhibitors to inhibit tumor growth in vivo. These studies identify an effective combination therapy for the most aggressive form of MB.

Biochemical and molecular characteristics of patients with organic acidaemias and urea cycle disorders identified through newborn screening

BACKGROUND

In recent years it has become clear that newborn screening (NBS) programmes using tandem mass spectrometry identify "patients" with "classical" inborn errors of metabolism who are asymptomatic. This observation raises issues regarding medicalization of "non-diseases," potentially unnecessary treatment and unnecessary anxiety to parents.

AIMS

This study aims to identify possible markers that may assist in predicting the need for treatment of infants with "classical" organic acidaemias (OA) and urea cycle disorders (UCD) diagnosed through NBS.

METHODS

Medical records of all patients with classical OA and UCD detected through the Victorian NBS programme from February 2002 to January 2014, or diagnosed clinically between 1990 and January 2002 were retrospectively reviewed.

RESULTS

Neonatal presentation did not always predict the need for on-going strict treatment. Blood concentrations of amino acids and acyl-carnitines and the changes thereof in follow-up samples correlated with severity in citrullinaemia-I, possibly isovaleric acidaemia but not in argininosuccinic aciduria or propionic acidaemia. Some specific mutations correlate with "attenuated" citrullinaemia-I. Gender may affect clinical outcome in propionic acidaemia.

CONCLUSIONS

Changes in blood concentration of certain metabolites (amino acids, acyl-carnitines) in the first weeks of life may be predictive of the need for treatment in some disorders but not in others. Mutation analysis may be predictive in some disorders but whether or not this should be considered as second-tier testing in NBS should be discussed separately.

Quiescent sox2(+) cells drive hierarchical growth and relapse in sonic hedgehog subgroup medulloblastoma

Functional heterogeneity within tumors presents a significant therapeutic challenge. Here we show that quiescent, therapy-resistant Sox2(+) cells propagate sonic hedgehog subgroup medulloblastoma by a mechanism that mirrors a neurogenic program. Rare Sox2(+) cells produce rapidly cycling doublecortin(+) progenitors that, together with their postmitotic progeny expressing NeuN, comprise tumor bulk. Sox2(+) cells are enriched following anti-mitotic chemotherapy and Smoothened inhibition, creating a reservoir for tumor regrowth. Lineage traces from Sox2(+) cells increase following treatment, suggesting that this population is responsible for relapse. Targeting Sox2(+) cells with the antineoplastic mithramycin abrogated tumor growth. Addressing functional heterogeneity and eliminating Sox2(+) cells presents a promising therapeutic paradigm for treatment of sonic hedgehog subgroup medulloblastoma.

Hotspot mutations in H3F3A and IDH1 define distinct epigenetic and biological subgroups of glioblastoma

Glioblastoma (GBM) is a brain tumor that carries a dismal prognosis and displays considerable heterogeneity. We have recently identified recurrent H3F3A mutations affecting two critical amino acids (K27 and G34) of histone H3.3 in one-third of pediatric GBM. Here, we show that each H3F3A mutation defines an epigenetic subgroup of GBM with a distinct global methylation pattern, and that they are mutually exclusive with IDH1 mutations, which characterize a third mutation-defined subgroup. Three further epigenetic subgroups were enriched for hallmark genetic events of adult GBM and/or established transcriptomic signatures. We also demonstrate that the two H3F3A mutations give rise to GBMs in separate anatomic compartments, with differential regulation of transcription factors OLIG1, OLIG2, and FOXG1, possibly reflecting different cellular origins.

Abstract 1430: Characterization of the medulloblastoma splice-ome reveals subgroup-specific changes in alternative splicing and isoform expression patterns

Medulloblastoma candidate gene approaches have previously identified alternative splicing and isoform expression changes affecting genes critical for cerebellar development and medulloblastoma pathogenesis. Here we present a bioinformatic characterization of the medulloblastoma splice-ome in 103 primary tumors and 14 normal cerebella samples. Medulloblastomas display a statistically significant increase in alternative splicing relative to normal fetal cerebella (2.3-times; P<6.47E-8), with splicing patterns that are specific to each molecular subgroup. Unsupervised hierarchical clustering of alternatively spliced genes accurately assigns medulloblastomas to their correct subgroup. One-third of all medulloblastomas (n=26) display a ‘hyper-spliced’ phenotype, with median splicing levels four-times greater than non-hyperspliced tumors. Hyperspliced medulloblastomas show a relatively worse overall survival (P<3.08E-2), and are seen across all molecular subgroups. Our analysis identifies previously reported splicing events targeting ERBB4 (mixed MB), GLI1 (SHH) and PTCH1 (SHH), supporting our approach. Subgroup-specific pathway analysis of alternatively spliced genes reveals extensive deregulation of neuronal pathways in non-WNT medulloblastomas, with the specific targeting of genes important for axonal guidance, synaptic transmission and neuronal differentiation. Finally, we present evidence for putative regulation of alternative splicing by antisense transcription. These data further demonstrate the differences between medulloblastoma subgroups, and highlight alternative splicing and isoform expression as a mechanism contributing to the transcriptional heterogeneity between subgroups, and perhaps to subgroup specific pathogenic mechanisms.

Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 1430. doi:1538-7445.AM2012-1430

Digital gene expression by tag sequencing on the illumina genome analyzer

This unit provides a protocol for performing digital gene expression profiling on the Illumina Genome Analyzer sequencing platform. Tag sequencing (Tag-seq) is an implementation of the LongSAGE protocol on the Illumina sequencing platform that increases utility while reducing both the cost and time required to generate gene expression profiles. The ultra-high-throughput sequencing capability of the Illumina platform allows the cost-effective generation of libraries containing an average of 20 million tags, a 200-fold improvement over classical LongSAGE. Tag-seq has less sequence composition bias, leading to a better representation of AT-rich tag sequences, and allows a more accurate profiling of a subset of the transcriptome characterized by AT-rich genes expressed at levels below the threshold of detection of LongSAGE (Morrissy et al., 2009).