Project description:Purpose: The aim of the present study was to examine the biological differences between seminomas with occult and clinically apparent metastases at the time of diagnosis of the primary tumor to gain insight into the biology of these tumors and facilitate the identification of novel predictors of seminoma metastasis. Materials and Methods: Total RNA including small RNAs was isolated from testicular tumors of patients with pure seminoma presenting with lymphogenic metastasis (n=5, clinical stage IIb/c) and occult metastasis (n=5, clinical stage I). The regulation of biological processes was examined (1) throughout the mRNA transcriptome (whole genome microarrays, 8x60K Array, Agilent and 4 samples per group) and (2) the miRNA transcriptome employing small RNA next generation sequencing (SOLID, Life Technologies). Protein coding genes (mRNAs) and small RNAs showing a significant (> 2-fold) difference between the groups were identified. Finally (3), we examined 95 candidate miRNAs in 36 apparent metastasized and another 5 occult metastasized seminoma using logistic regression analysis. Results: Among 19,596 genes, on average 12,894 mRNAs appeared expressed (65.8%, SD +/- 2.4; range, 62.0M-bM-^@M-^S69.3%) and 16.99M-CM-^W106/13.94M-CM-^W106 small RNA reads were identified for apparent/occult metastasized seminoma. These reads on average convert into 9,901/9,675 small RNAs including 422/404 mature microRNAs. None of these mRNAs/small RNAs met our selection criteria for candidate genes. From 95 candidate miRNAs 44 appeared expressed, with 3 of them showing weak but significant (p=0.047-0.049) differences among both groups. Conclusions: Occult and apparent metastasized seminomas are biologically almost indistinguishable and probably represent no separate tumor entities. These findings may simplify future research on seminoma metastasis. Regarding the genome-wide expression profiling we analyzed gene expression changes in human seminoma samples with occult or lymphatic metastasis using Agilent SurePrint G3 Human Gene Expression 8x60K v2 Microarrays in 8x60K format in combination with a one-color based hybridization protocol. Microarray signals were detected using the Agilent DNA Microarray Scanner. Differential RNA expression was assessed by applying appropriate biostatistics to the data set. GeneSpring GX12.6 analysis software was used to normalize and analyze the raw data. Different filtering approaches were applied to identify up- and downregulated transcripts in the comparisons of interest. Identified differentially expressed genes were further characterized for their involvement in biological pathways and processes based on the Panther database. For mRNA transcriptome analysis we used 4 samples per group. Only significant and > 2-fold differences in gene expression between the two groups were considered for comparisons, and unadjusted p-values and those adjusted for multiple comparisons (false discovery rate) were calculated. All gene candidates underwent gene set enrichment analyses using PANTHER pathway software (http://www.pantherdb.org/), which groups genes with similar biological functions based on their annotation.

Project description:Germ cell tumours (GCTs) are a complex group of malignancies. They are unique in developing from a pluripotent progenitor cell. The initial lesion is believed to be the relatively benign precursor lesion (ICGNU), from which either highly chemosensitive seminomas or the more aggressive non-seminomas develop. Previous analyses of selected genes have suggested that non-seminomas exhibit much higher levels of DNA methylation than seminomas. However, the genomic targets that are methylated, the extent to which this results in gene silencing and the identity of the silenced genes most likely to play a role in the tumoursâ biology have not yet been established. Here genome-wide methylation and expression analysis of GCT cell lines was combined with gene expression data. This demonstrated that the seminoma cells exhibited very little aberrant DNA methylation while non-seminoma cells exhibited very high levels of DNA methylation. Large differences in the level of methylation of the CpG islands of individual genes between tumour cell lines correlated well with differential gene expression. Treatment of non-seminoma cells with 5-aza-2-deoxycytidine verified that methylation of all genes tested did indeed play a role in their silencing and many of these genes were also differentially expressed in primary tumours. Through this approach the genes silenced in the various GCT cell lines were identified. Conclusions: Several pluripotency-associated genes, never before implicated in this type of cancer, were identified as a major functional group of silenced genes. Silencing of these factors that normally suppress somatic differentiation might play an important role in the progression to non-seminoma formation. Genomic DNA and RNA was extracted from cell lines representing four subtypes of GCT. RNA was subjected to Affymetric expression array analysis while DNA was bisulfite treated and analysed using Illumina Infinium 450K arrays. Statistical approaches were used to correlate methylation and expression for each gene.

Dataset Information

Gene expression analysis of human seminoma samples

Dataset's files

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets