Sarcomas are cancers of connective tissues, such as bone, adipose and cartilage, and are thought to arise from the aberrant development of the mesenchyme. As such, mesenchymal stem cells are thought to be the cell of origin for sarcomas. Genetic or epigenetic lesions at particular points during the differentiation of a mesenchymal stem cell into its terminal mesenchymal cell type are able to give rise to specific subtypes of sarcomas.
Recently, a number of reports have identified elevated expression of the human Piwi homolog–called Hiwi–in a variety of human cancers, including gastric cancer, pancreatic cancer, gliomas and, most relevant for this dissertation, sarcomas. In sarcomas, Hiwi is highly expressed and elevated Hiwi prognosticates shorter patient survival. Hiwi is the human homolog of the Piwi family of proteins, which are members of the Paz-Piwi Doman (PPD) family.
During normal development, Piwis are thought to maintain stem cells of the germline, and indeed their expression is limited to early development and to the adult germline. Piwis are thought to maintain stem cells in the germline with small RNA partners, called piwi-interacting RNAs (piRNAs). More specifically, Piwi/piRNA complexes in the germline are thought to maintain transposon silencing, and thus ensure genomic stability.
A detailed mechanism by with Piwis suppress transposon migration in the germline remains an area of active investigation, but is thought to occur via DNA methylation of transposon regions. In this way, Piwis are critical for maintenance of genomic integrity of germline stem cells during normal development. Thus, the finding that Piwis are elevated in human cancers is directly in conflict with its known role in ensuring genomic stability during development. Piwi homologs are critical for maintenance of germline stem cells during development but aberrant Hiwi expression has also been identified in all cancers examined, including in sarcomas.
A potential connection between mesenchymal stem cells, sarcomas and Hiwi remains unexplored. Moreover, the role of Hiwi in sarcomas is unknown. In the studies presented here, we demonstrate that over-expressing Hiwi in mesenchymal stem cells inhibits their differentiation in vitro and generates sarcomas in vivo. Secondly, transgenic mice expressing Hiwi (mesodermally-restricted) develop sarcomas. Conversely, inducible down-regulation of Hiwi in human sarcomas inhibits growth and re-establishes differentiation. These data reveal that Hiwi is directly tumorigenic.
We have also identified the presence of piRNAs in our Hiwi-expressing models. We further show that DNA methylation correlates with Hiwi expression and that cyclin-dependent kinase inhibitor (CDKI) tumor suppressor genes are silenced upon Hiwi over-expression. Moreover, Hiwi’s tumorigenic effects are reversible using DNA de-methylating agents. These studies reveal for the first time not only a novel oncogenic role for Hiwi as a driver of tumorigenesis, but also suggest that the use of epigenetic agents may be clinically beneficial for treatment of tumors that express Hiwi.
Additionally, our data showing that Hiwi-associated DNA hyper-methylation with subsequent genetic and epigenetic changes favoring a tumorigenic state reconciles the conundrum of how Hiwi may act appropriately to promote genomic integrity during early development (via transposon silencing) and inappropriately in adult tissues with subsequent tumorigenesis.