High-throughput microRNAome analysis in human germ cell tumors: implications for understanding tumorigenesis.

2968

Testicular germ cell tumors (GCTs) of adolescents and adults represent a heterogeneous group of solid cancers that can be subdivided into a number of clinical and biological entities. These include the seminomas and nonseminomas, originating from carcinoma in situ (CIS) cells, being the malignant counterparts of primordial germ cells (PGC)/gonocytes. The invasive components represent the different lineages of embryogenesis, including the stem cell component embryonal carcinoma (EC), the somatic lineage teratoma (TE) and the two extra-embryonic tissues (yolk sac tumor and choriocarcinoma). These are referred to as type II GCTs. The other one is the so-called spermatocytic seminomas (SS, type III GCTs), composed of neoplastic primary spermatocytes. mRNA expression profiling can distinguish the different types of GCTs. Recently, we found that a specific cluster of miRNAs (hsa-miR 371-373) is involved in overruling cellular senescence allowing the cells to fully transform, demonstrating the relevance of miRNAs in type II GCT tumorigenesis.

Here we report the novel findings on the first high-throughput screen of 157 microRNAs in a series of type II and III GCTs using a quantitative PCR-based approach. In total 70 samples were investigated in duplicate, which clustered together, demonstrating the validity and reproducibility of the Q-PCR approach. The previous finding on the hsa-miRNA 371-373 cluster was confirmed. A normalization method was applied to allow inter-sample analysis. Unsupervised cluster analysis demonstrated that the cell lines showed a different pattern of miRNA expression profile compared to the in vivo samples. The different histologies of in vivo samples, both normal and malignant, are characterized by expression of specific miRNA sets. This finding parallels the expression profile during normal embryogenesis, rather than the presence of chromosomal anomalies in the tumors. The majority of miRNAs within a single genomic cluster showed a similar expression pattern, implying common regulatory mechanisms. Normal testicular tissue expressed most discriminating miRNAs at a higher level than SE and SS. Moreover, differentiated nonseminomas showed overexpression of discriminating miRNAs. These results support the model that miRNAs are involved in regulating differentiation of stem cells.

This study demonstrates the value of high-throughput quantitative expression analysis of miRNAs on normal and tumor tissues to generate information about the pathogenesis of the cancer. In addition, it shows the impact of differentiation within GCTs on the expression level of miRNAs.