Bladder cancer is the second most common urological malignancy in the United States and is usually not curable at its advanced stage. However, basic research funding geared specifically to diseases in this organ site is relatively meager. It has long been observed that human tumors are heterogenous, as defined by histology or biological properties (i.e. anchorage independent growth property, proliferation and differentiation status). One model hypothesizes the existence of a hierarchy in tumors, which may explain the heterogeneity of tumor cell population within a tumor. This model proposes that rare tumor-initiating cells with stem cell-like properties (cancer stem cells) exist, which can self-renew and differentiate into phenotypically diverse tumor cells that reconstitute the heterogeneity of original tumor. In the current proposal, I describe for the first time the successful isolation and initial characterization of a rare CD44 positive, bladder cancer stem cells (CSCs) from patient tumor. I hypothesize that CSCs exist in all patient bladder tumors and therefore can be prospectively isolated based on their unique biological properties. The primary experimental approaches will be utilizing a combination of cell surface markers (Aim 1A.B&C) and constructs that report on self-renewal signaling pathways (Aim2A). Localization of bladder CSCs and the correlation of their expression level to disease progression will be determined by immunohistochemical analysis in tissue sections and high throughput tissue microarrays (Aim1 D). Further, in order to define a molecular signature"""""""" for bladder CSCs, two separate approaches will be taken (Aim3): (I) a small scale molecular screen to look at specific """"""""sternness genes"""""""" at mRNA level by quantitative RT-PCR (AimSAi), and utilization of cutting edge microfluidics western technologies to look at these candidates at protein level (AimSAii);(II) a genome wide screen using HEEBO microarrays (AimSB). Preclinical targeting of selfpathways unique to bladder CSCs (Aim2B) in comparison to the MARK pathway (Aim2C) will reveal drug targets in vivo. I hypothesize that these CSCs are in fact the """"""""roots"""""""", and therefore ideal targets for future anti-cancer therapies. My long term goal is to enrich for a highly purified population of bladder CSCs, further characterize their biological properties (e.g. ability to invade, and migrate to distal organs), and to develop specific strategies to target such CSCs in attempt to eradicate advanced stage bladder cancers.