Studies of gene fusions in rhabdomyosarcoma

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

We investigated the molecular collaborating events that are needed for P3F to exert its oncogenic activity during tumorigenesis. In our past studies with the human myoblast system, we showed that the combination of high level MYCN and P3F expression supports a high level of transformation and induces rapid tumorigenesis (3-4 weeks). In contrast, in the setting of P3F expression without exogenous expression of a Myc family protein, there is no detectable transformation and tumorigenesis only occurs after a significant lag (8-9 weeks). Of note, when cells from the P3F-only tumors are cultured and retested in oncogenic assays, these tumor-derived cells show a high level of transformation and rapid tumorigenesis (3-4 weeks), similar to the activity found in cultured cells derived from P3F/MYCN tumors. We hypothesize that, whereas most P3F/MYCN-expressing myoblasts are transformed and tumorigenic, only rare myoblasts expressing P3F alone are transformed and tumorigenic, and thus additional time is needed for these rare cells to proliferate sufficiently to form a grossly detectable tumor mass. To examine the molecular events occurring in these rare transformed P3F-only myoblasts, we assayed RNA and protein expression of P3F, MYC and MYCN in the parental and tumor-derived cells from both P3F-only and P3F/MYCN myoblasts. P3F expression is relatively comparable in parental and tumor-derived cells from both groups of myoblasts. The MYCN expression level in parental P3F-only myoblasts is near the lower limits of detectability whereas the expression level in the parental P3F/MYCN myoblasts is several orders of magnitude higher. Furthermore, the MYCN expression level in the P3F/MYCN tumor-derived cells is comparable to the P3F/MYCN parental cells whereas the MYCN expression level in the P3F-only tumors is substantially higher than the corresponding parental cells, though still lower than the level in the P3F/MYCN parental and tumor-derived cells. In contrast, MYC expression is higher in P3F-only parental cells than in P3F/MYCN parental cells. Furthermore, MYC expression appears to be higher in P3F-only tumor-derived cells than parental cells (particularly at the protein level), whereas there is no difference in MYC expression between P3F/MYCN parental and tumor-derived cells. These findings are compatible with the presence of a feedback mechanism by which the high level of exogenous MYCN expression in the P3F/MYCN cells results in a lower level of MYC expression in these cells compared to the P3F-only cells. Furthermore, though there is no difference in MYCN or MYC expression between P3F/MYCN parental and tumor-derived cells, there appears to be a selection for cells expressing higher MYCN and MYC levels in the P3F-only tumor-derived cells compared to the parental cells. We hypothesize that the rare transformed cells in the P3F-only population are those with higher Myc family expression, and thus selection for the transformed phenotype effectively selects for higher MYCN and MYC expression. In a final set of experiments, we assessed the functional significance of MYCN and MYC in the tumor-derived cells by using a CRISPR/Cas9 targeting strategy to inactivate MYCN or MYC expression. In the P3F/MYCN tumor-derived cells, MYCN but not MYC is required for transforming activity (as defined by the focus formation assay). In contrast, MYCN knockout partially reduces transforming activity in the P3F-only tumor-derived cells but MYC knockout almost completely abrogates transforming activity. Based on these knockout experiments, we conclude that the P3F/MYCN cells are primarily dependent on MYCN for their oncogenic activity whereas P3F-only cells are primarily dependent on MYC. These two systems provide models to examine the molecular basis and functional significance of heterogeneity within the FP RMS category.