Dataset Information

Transcription profiling by array of human paediatric medulloblastoma samples and normal cerebellum to investigate the occurance of alternative splicing

ABSTRACT: Alternative pre-mRNA splicing critically contributes to the generation of protein variety in a tissue- and development-specific manner. Alterations in the normal pathways of alternative splicing (AS) have been associated with the growth and maintenance of several tumour types, and have been indicated as candidate bio-markers of tumour progression, metastasis and patient survival. In this study we applied genome-wide exon array technology to investigate AS events that may distinguish between human medulloblastoma and normal cerebellum. We initially investigated gene-level expression profiles to identify samples expressing gene signatures characteristic of previously described MB molecular subgroups. 3 medulloblastomas significantly over-expressed genes typically belonging to the Shh signalling pathway or associated with a granule cell progenitor status and were therefore classified as Shh-activated tumours (MB1). The remaining tumour samples were grouped together as medulloblastoma subset 2 (MB2). We then applied the Splicing Index algorithm and identified 1260 unique genes containing at least one candidate exon whose inclusion rate differed between different sample subgroups. Following the analysis of candidate event expression plots and gene structure annotation, we selected 14 examples of differential splicing of cassette exons and successfully validated 11 of them by semi-quantitative RT-PCR in a selection of the initial sample-set and subsequently in an independent set of 10 normal cerebellum and 20 medulloblastoma samples. Through the analysis of AS pathway in in vitro cultures of cerebellar granule cell progenitors (the putative cell of origin of least a subset of medulloblastomas), we showed that medulloblastoma-associated AS patterns could be indicative of a normal cerebellar undifferentiated phenotype and suggested that activation of oncogenic pathways during the development of the cerebellum may lead to a failure of neuronal differentiation also through the disruption of AS programs.

ORGANISM(S): Homo sapiens

SUBMITTER: Menghi F

PROVIDER: S-ECPF-MTAB-292 | biostudies-other | 2011 Mar

REPOSITORIES: biostudies-other

ACCESS DATA

Similar Datasets

Project description:Alternative pre-mRNA splicing critically contributes to the generation of protein variety in a tissue- and development-specific manner. Alterations in the normal pathways of alternative splicing (AS) have been associated with the growth and maintenance of several tumour types, and have been indicated as candidate bio-markers of tumour progression, metastasis and patient survival. In this study we applied genome-wide exon array technology to investigate AS events that may distinguish between human medulloblastoma and normal cerebellum. We initially investigated gene-level expression profiles to identify samples expressing gene signatures characteristic of previously described MB molecular subgroups. 3 medulloblastomas significantly over-expressed genes typically belonging to the Shh signalling pathway or associated with a granule cell progenitor status and were therefore classified as Shh-activated tumours (MB1). The remaining tumour samples were grouped together as medulloblastoma subset 2 (MB2). We then applied the Splicing Index algorithm and identified 1260 unique genes containing at least one candidate exon whose inclusion rate differed between different sample subgroups. Following the analysis of candidate event expression plots and gene structure annotation, we selected 14 examples of differential splicing of cassette exons and successfully validated 11 of them by semi-quantitative RT-PCR in a selection of the initial sample-set and subsequently in an independent set of 10 normal cerebellum and 20 medulloblastoma samples. Through the analysis of AS pathway in in vitro cultures of cerebellar granule cell progenitors (the putative cell of origin of least a subset of medulloblastomas), we showed that medulloblastoma-associated AS patterns could be indicative of a normal cerebellar undifferentiated phenotype and suggested that activation of oncogenic pathways during the development of the cerebellum may lead to a failure of neuronal differentiation also through the disruption of AS programs.

Project description:BackgroundBRCA1 is a key protein in cell network, involved in DNA repair pathways and cell cycle. Recently, the ENIGMA consortium has reported a high number of alternative splicing (AS) events at this locus in blood-derived samples. However, BRCA1 splicing pattern in breast tissue samples is unknown. Here, we provide an accurate description of BRCA1 splicing events distribution in breast tissue samples.MethodsBRCA1 splicing events were scanned in 70 breast tumor samples, 4 breast samples from healthy individuals and in 72 blood-derived samples by capillary electrophoresis (capillary EP). Molecular subtype was identified in all tumor samples. Splicing events were considered predominant if their relative expression level was at least the 10% of the full-length reference signal.Results54 BRCA1 AS events were identified, 27 of them were annotated as predominant in at least one sample. Δ5q, Δ13, Δ9, Δ5 and ▼1aA were significantly more frequently annotated as predominant in breast tumor samples than in blood-derived samples. Predominant splicing events were, on average, more frequent in tumor samples than in normal breast tissue samples (P = 0.010). Similarly, likely inactivating splicing events (PTC-NMDs, Non-Coding, Δ5 and Δ18) were more frequently annotated as predominant in tumor than in normal breast samples (P = 0.020), whereas there were no significant differences for other splicing events (No-Fs) frequency distribution between tumor and normal breast samples (P = 0.689).ConclusionsOur results complement recent findings by the ENIGMA consortium, demonstrating that BRCA1 AS, despite its tremendous complexity, is similar in breast and blood samples, with no evidences for tissue specific AS events. Further on, we conclude that somatic inactivation of BRCA1 through spliciogenic mutations is, at best, a rare mechanism in breast carcinogenesis, albeit our data detects an excess of likely inactivating AS events in breast tumor samples.

Project description:Purpose Integrated genomics approaches have identified at least four distinct biological variants in medulloblastoma: WNT, SHH, group C, and group D. Non-WNT/Non-SHH tumors are associated with metastatic dissemination and an unfavorable prognosis. Additional markers may enhance outcome prediction in Non-WNT/Non-SHH medulloblastomas. Experimental Design We combined transcriptomic and DNA copy-number analyses for 64 primary medulloblastomas. Bioinformatic tools were applied to discover marker genes of molecular variants. Differentially expressed transcripts were evaluated for prognostic value in the screening cohort. Immunopositivity for FSTL5 was correlated with molecular and prognostic subgroups for 235 non-overlapping medulloblastoma samples on two independent tissue microarrays (TMA). Results Unsupervised clustering analyses of transcriptome profiles confirmed four distinct molecular variants. Stable subgroup separation was achieved using only the 300 most varying transcripts. Specific distributions of clinical and molecular characteristics were noted for each cluster. Distinct expression patterns of FSTL5 in each molecular subgroup were confirmed by quantitative real-time PCR. Immunopositivity of FSTL5 identified a large cohort of patients (84 of 235 patients; 36%) at high risk for relapse and death. Importantly, over 50% of Non-WNT/Non-SHH tumors displayed FSTL5 negativity, delineating a large patient cohort with an excellent prognosis who would be considered intermediate/high-risk based on current molecular subtyping. Conclusions Comprehensive analyses of transcriptomic and genetic alterations delineate four distinct variants of medulloblastoma. The addition of FSTL5 immunohistochemistry to existing molecular stratification schemes can effectively identify those Non-WNT/Non-SHH tumors with a poor outcome. Immunohistochemical staining for FSTL5 could be a high-quality and practical tool for stratification and prognostication in future clinical trials of medulloblastoma. Whole-genome transcriptional profiling of human medulloblastomas. Subgrouping based on mRNA expression profiles. Fresh frozen tumor material was collected during tumor resection. Dye-swap design used for expression profiling. Reference was a pool of normal cerebellum tissue from 24 donors. Gene expression profiles illustrate distinct expression pattern at diagnosis. This submission represents the gene expression component of the study.

Dataset Information

Transcription profiling by array of human paediatric medulloblastoma samples and normal cerebellum to investigate the occurance of alternative splicing

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets