In our previous studies, we reported the constitutive activation of STAT1 in Wilms tumors. Activation of a STAT family member occurs in a variety of human neoplasms but generally involves Y705 phosphorylation of STAT3. While investigating the possible dysregulation of STATs in Wilms tumors, we discovered the constitutive phosphorylation of S727 in STAT1 and demonstrated its role in Wilms tumorigenesis. Since this neoplasm is a prototype for arrested cellular differentiation in cancer, we hypothesized that STAT1 signaling might also play an important role in the developing kidney. In order to determine if STAT signaling is active during metanephric development, we analyzed a temporal series of embryonic kidneys and discovered that both STAT1 and STAT3 are activated by tyrosine and serine phosphorylation during the peak of nephronic development. STAT5 is also expressed and may function in this process. To examine their roles in nephrogenesis, we utilized explant cultures of metanephric mesenchyme (MM), the progenitor of nephronic epithelia and putative origin of Wilms tumor. Interferon-gamma (IFNg), unlike differentiation-inducing leukemia inhibitory factor (LIF), caused STAT1, but not STAT3, activation in MM. In functional opposition to LIF, which induces tubulogenesis in MM, IFNg stimulated proliferation in MMs and inhibited nephron formation following induction. Furthermore, a peptide inhibitor was designed by our NCI-Frederick collaborator Nadya Tarasova, to target the STAT1 N-domain, and it specifically disrupts STAT1-dependent transcription and induces tubule formation in explanted MMs. These findings indicate that STAT1 activation provides a critical regulatory signal in MM specification, consistent with its role in Wilms tumorigenesis, and further suggest that disruption of STAT1 activation may provide a novel target for Wilms tumor therapy. We are currently evaluating the efficacy of using this novel STAT1 peptide as an anti-tumor agent in vivo. To directly examine the role of STAT signaling in renal development, we are also developing mouse genetic models for mutant forms of STAT1 that conditionally target S727 activity (floxed stop codons upstream of the mutant Stat form). These include a S727-to-E727 mutation, which is constitutively active, and a S727-to-A727, which is inactive for phosphorylation at this site. We also have generated a floxed-stop codon form of IFNg, which causes increased expression of this cytokine conditionally with Cre activation. We speculate that this will increase the progenitor population in the developing kidney or at least disrupt normal development. A similar strategy has been used by others to induce medulloblastomas which are STAT1 dependent in mice. Finally, we have also obtained mouse lines with floxed Stat1 or Stat3 alleles to assess the impact of the loss of each of these genes either separately or in combination. In addition to these mechanistic studies, we are attempting to leverage our findings with LIF and IFNg in tissue culture in order to massively expand populations of MM for biochemical analyses. A longstanding problem in working with embryonic tissues is the small numbers of cells that can be harvested for study. We estimate that the average MM contains about 4000 cells. By culturing these cells on a matrix using LIF and IFNg, we have now been able to expand 30 MMs (120,000 cells) into over 20 million cells in a weeks time without sacrificing their ability to be induced to differentiate. We are currently evaluating subsequent passages to determine if they retain this capacity. This would provide an invaluable resource for the research community if we are successful in achieving this. In addition, we have applied similar conditions to Wilms tumor tissues in order to establish cell lines from these as well. There is currently only one cell line that has been successfully established from a Wilms tumor, and clinicians have voiced a need for more appropriate lines for drug trial xenograft studies. With our conditions, we have established an early passage cell line from a Wilms tumor that expresses characteristics of Wilms including NCam, cadherin11 and Osr1. We are in the process of acquiring more tumors for culturing and will eventually use them in xenograft studies to demonstrate tumorigenicity and evaluate histology. There has been considerable effort and resource committed by others in the pursuit of this goal, so it would be a major step forward for this research area to have a replenishable biological source for study and an improved model for xenograft drug trials. Another feature of Wilms tumor is the frequent activation of the Wnt pathway, as demonstrated in our recent publication of nephroblastomas, which show nuclear localization of beta-catenin in most tumors. This pathway is also critical to nephronic development;however, the exact mechanism involved is not clear. Studies have shown an essential role for Wnt/beta-catenin in mesenchymal-epithelial transition (MET) during nephron formation. Intriguingly, analysis of BATlacZ transgenic mice expressing a beta-catenin-transactivated lacZ reporter revealed that the reporter is strongly expressed in the ureteric bud but not in surrounding metanephric mesenchyme (MM) or newly formed nephronic tubules, raising the possibility that beta-catenin may act through a TCF/LEF-independent mechanism to promote MET and tubule formation. To explore the mechanism, isolated MMs were treated with a variety of small molecules that affect the Wnt/beta-catenin pathway. An activator of Wnt signaling (WA), which stimulates TCF/LEF-dependent transactivation without altering beta-catenin levels, could not induce MET. Three glycogen synthase kinase-3beta inhibitors, including bromoindirubin-3-oxime (BIO), were able to increase beta-catenin levels and induce MET. Proteasome inhibitor MG132 also increased beta-catenin levels, but did not induce MET. Using our newly developed approach for expanding the population of MM cells, we are now able to perform GST pull-down studies of beta-catenin to determine how the various chemical treatments affect its binding form. We have found that BIO treatment increased both TCF- and cadherin-binding forms of beta-catenin, while MG132 elevated only the TCF-binding form. Expression of a mutated cadherin that disrupts normal beta-catenin/cadherin complex formation prevented BIO-induced tight junction formation. Consistent with these observations, neutralizing antibody against cadherin abrogated BIO-induced tubulogenesis in cultured MMs. These data indicate that a beta-catenin interaction with cadherin plays an essential role in MET and tubule formation during renal development. We are currently testing peptides generated by Dr. Tarasova that target beta-catenin and its interaction with TCF/LEF1 in efforts to specifically block canonical activation. Such peptides may have the ability to biochemically dissect Wnt signaling, separating its role in transcriptional activation through TCF from its function in adherens junctions. Since many different types of tumors show nuclear localization of beta-catenin and aberrant activation of Wnt signaling, such an approach could impact the development of effective therapies directed at this target. We have also investigated the susceptibility of rat to develop nephroblastomas. The Noble rat is exquisitely sensitive to chemical induction of these tumors, while the F344 rat is insensitive. When crossed, animals develop a tumor incidence suggestive of the involvemen [summary truncated at 7800 characters]
Showing the most recent 10 out of 20 publications
