WT1 Interaction with the Chaperone Protein HSP90 Regulates Its Expression in Myeloid Leukemia.

Introduction: Wilms tumor protein 1 (WT1) is a transcription factor that is universally expressed in AML, and clinical studies have demonstrated a significant correlation between WT1 expression and disease progression. Heat shock proteins (HSP) are ubiquitous molecular chaperones that are involved in the folding, activation and assembly of many proteins. HSP90 may facilitate the maintenance of altered oncogenic client proteins such as flt-3 and bcr-abl in myeloid malignancies. It has been also shown that WT1 is a client protein of Hsp70, and this interaction may be important for the function of WT1. Since HSP70 and HSP90 have similar functions and often complement each other in engaging with client cellular proteins we wanted to determine if WT1 was chaperoned by HSP90 and if so could this HSP90-WT1 interaction be modulated.

Materials and methods: We first looked for a potential interaction between HSP90 and WT1 by confocal microscopy colocalization and then by immunoprecipitation (IP) in the myeloid cell line K562. For confocal studies, cells were grown on coverslips, fixed and permeabilized and stained with the rabbit anti-WT1 and anti- HSP90 antibodies. Samples were examined by using a laser confocal/Zeiss axiovert microscope imaging system. This interaction was further confirmed by co-IP/Western and then confirmed using a GST-pulldown assay. A glutathione S-transferase (GST)-WT1 fusion protein was expressed in E. coli strain BL21(DE3) and immobilized onto glutathione sepharose beads. The full length Hsp90 and different deletion mutant proteins were in vitro translated and [35S] methionine-labeled; the bound proteins were eluted and then separated by SDS-PAGE. The gels were fixed, dried and subjected to fluorography.

Results: Confocal microscopy revealed colocalization of HSP90 and WT1 mainly in the nucleus. Immunoprecipitation of K562 lysates with either anti-WT1 or anti-HSP90 antibody followed by Western analysis for WT1 and HSP90 demonstrated that WT1 and HSP90 could be co-immunoprecipitated. The GST pull down assay confirmed this interaction. Additional mutants of HSP90 from the C to N terminus were prepared to localize the WT1 interacting domain. These studies localized the WT1 interacting domain to the HSP90 region close to the binding pocket of the HSP inhibitor geldanamycin. Treatment of K562 with the geldanamycin analog 17-allylamino-17-demethoxygeldanamycin (17-AAG) resulted in downregulation of WT1 protein expression with an IC50 of 3μM.

Conclusions: Understanding the mechanisms that regulate WT1 expression thus may lead to novel therapeutic strategies to inhibit this expression and result directly in myeloid blast cell killing. We have shown elsewhere that downmodulating WT1 gene expression with interferon or shRNA potentiates myeloid leukemic blast killing by chemotherapeutic agents. We demonstrate here that the WT1 protein interacts with the co-chaperone heat shock protein HSP90. Taken together, our findings support the hypothesis that the regulation of WT1 occurs at the protein level through an interaction with WT1-protein-HSP chaperone levels, and that pharmacologically targeting this interaction with the HSP inhibitor 17AAG results in a significant reduction in WT1 protein expression. Ongoing studies in primary AML cells will confirm these findings and define novel therapeutic strategies to manipulate WT1 protein expression levels and modulate leukemic blast survival.