Our initial goal was to develop cell culture systems in which PAX3-FOXO1 expression can be modulated and in which the downstream effects of PAX3-FOXO1 can be assessed at the gene expression and phenotypic levels. In previous studies, we developed a doxycycline-inducible PAX3-FOXO1 expression construct. This construct was introduced into immortalized human myoblasts (with and without a constitutive MYCN expression construct) to generate cell culture systems in which PAX3-FOXO1 expression can be up-regulated by drug treatment and then down-regulated by drug withdrawal. In contrast to the brisk growth of these cells in the absence of doxycycline, there is little or no growth when PAX3-FOXO1 expression was induced (with or without endogenous MYCN expression), thus confirming the toxic effects of PAX3-FOXO1. Despite these toxic effects, cell culture evidence of oncogenic transformation was detected in cells with induced PAX3-FOXO1 expression, but only in the cells constitutively co-expressing exogenous MYCN. During the past year, we further investigated the phenotypic changes resulting from induction of PAX3-FOXO1 expression in these cells. To characterize the toxic effects of PAX3-FOXO1, we assessed PARP cleavage as a marker of apoptotic cell death. Though PAX3-FOXO1 induction is associated with diminished growth with or without exogenous MYCN expression, PARP cleavage was only detected in cells constitutively co-expressing exogenous MYCN. Based on the hypothesis that tolerance to the toxic effects may be present in a small subset of cells in the population, cells with inducible PAX3-FOXO1 and constitutive MYCN expression were subcloned, and individual subclones were tested for PAX3-FOXO1-induced toxic effects. Diminished growth in the presence of doxycycline was detected in all examined subclones except for one in which there was greatly reduced induction of PAX3-FOXO1 expression. To detect the presence of tolerant cells at a lower frequency, we developed a strategy to enrich for these cells by growing several of the subclones for six weeks in the presence of doxycycline. Though doxycycline treatment was still able to induce robust PAX3-FOXO1 expression, there were no detectable toxic effects associated with PAX3-FOXO1 expression in these """"""""selected"""""""" cells, and thus we conclude that these cells developed tolerance to these effects. The presence of functional PAX3-FOXO1 in these tolerant cells was evidenced by the finding of oncogenic transformation in doxycycline-treated cells. In a related set of studies, we analyzed myogenic differentiation in PAX3-FOXO1-inducible myoblasts in the presence of standard differentiation conditions (growth factor-deficient medium). In accord with previous studies of cells with constitutive PAX3-FOXO1 expression, doxycycline induction of PAX3-FOXO1 expression inhibited myogenic differentiation as determined by western blot detection of differentiation products and cytologic detection of elongated myotubes. Though doxycycline induction of PAX3-FOXO1 expression was also associated with inhibition of differentiation in the tolerant cell populations, the untreated tolerant cells demonstrated more marked myogenic differentiation, particularly at earlier times and even in normal culture conditions. Therefore, it appears that the selection of cells with higher tolerance also selected for cells with a greater dependence on continued PAX3-FOXO1 expression to prevent myogenic differentiation. In a final set of experiments, we assessed the tumorigenic capability of these myoblasts with inducible PAX3-FOXO1 expression (with and without constitutive MYCN expression). Though tumors did not form in control or test cells without doxycycline treatment, progressively growing tumors formed in mice injected with PAX3-FOXO1-inducible myoblasts when treated with doxycycline. Furthermore, the cells expressing both inducible PAX3-FOXO1 and constitutive MYCN formed tumors several weeks earlier than cells expressing inducible PAX3-FOXO1 without exogenous MYCN. For inducible PAX3-FOXO1-expressing cells with or without constitutive MYCN expression, tumor formation stopped and often regressed when doxycycline was withdrawn. However, progressively growing tumors eventually re-formed several weeks after the drug was stopped, suggesting that rare cells with alternative oncogenic changes were selected. Similar to our cell culture study of cells grown continuously in doxycycline, cell lines derived from these doxycycline-induced tumors demonstrated tolerance to PAX3-FOXO1-induced toxicity.
Showing the most recent 10 out of 20 publications
