Transcription factor PU.1 is required for the development of lymphoid and myeloid progenitors during fetal hematopoiesis. By generating chimeric animals using PU.1-/- ES cells or PU.1(-/-) hematopoietic progenitors, we demonstrate that PU.1 functions in an exclusively cell-autonomous manner to regulate the development of the lymphoid-myeloid system. Multipotential lymphoid-myeloid progenitors (AA4.1+, Lin-) are significantly reduced in PU.1(-/-) embryos and fail to differentiate into B lymphoid or myeloid cells in vitro. These results suggest that the lymphoid and myeloid lineages develop in the fetal liver from a common hematopoietic progenitor not shared with erythrocytes and megakaryocytes. Finally, the Ikaros gene is expressed in PU.1 mutant embryos, suggesting that PU.1 and Ikaros are independently required for specification of embryonic lymphoid cell fates.

The ETS-family transcription factor PU.1 is expressed in hematopoietic tissues, with significant levels of expression in the monocytic and B lymphocytic lineages. PU.1 is identical to the Spi-1 proto-oncogene which is associated with the generation of spleen focusforming virus-induced erythroleukemias. An extensive body of in vitro gene regulatory studies has implicated PU.1 as an important, versatile regulator of B lymphoid-and myeloid-specific genes. The first half of the review is designed to coalesce data generated from studies examining the two PU.1 "knockout" animals, which have prompted a reevaluation of the proposed function of PU.1 during hematopoiesis. During hematopoiesis, PU.1 is required for development along the lymphoid and myeloid lineages but needs to be downregulated during erythropoiesis. These unique functional characteristics of PU.1 will be exemplified by contrasting the function of PU.1 with other transcription factors required during fetal hematopoiesis. The second half of this review will reexamine the functional characteristics of PU.1 deduced from traditional biochemical and transactivation assays in light of recent experiments examining the functional behavior of PU.1 in an embryonic stem cell in vitro differentiation system. Working models of how PU.1 regulates promoter and enhancer regions in the B cell and myeloid lineage will be presented and discussed.

Ulcerative colitis (UC) increases the risk of colorectal cancer (CRC), but the mechanisms involved in colitis-to-cancer transition (CCT) are not well understood. CCT may involve a inflammation-dysplasia-carcinoma progression sequence compared to the better characterized adenoma-carcinoma progression sequence associated with sporadic CRC. One common thread may be activating mutations in components of the Wnt/β-catenin signaling pathway, which occur commonly as early events in sporadic CRC. To examine this hypothesis, we evaluated possible associations between Wnt/β-catenin signaling and CCT based on the cancer stem cell (CSC) model. Wnt/β-catenin immunostaining indicated that UC patients have a level of Wnt-pathway-active cells that is intermediate between normal colon and CRC. These UC cells exhibiting activation of the Wnt pathway constituted a major subpopulation (52%+7.21) of the colonic epithelial cells positive for aldehyde dehydrogenase (ALDH), a putative marker of precursor colon CSC (pCCSC). We further fractionated this subpopulation of pCCSC using a Wnt pathway reporter assay. Over successive passages, pCCSCs with the highest Wnt activity exhibited higher clonogenic and tumorigenic potential than pCCSCs with the lowest Wnt activity, thereby establishing the key role of Wnt activity in driving CSC-like properties in these cells. Notably, 5/20 single cell injections of high-Wnt pCCSC resulted in tumor formation, suggesting a correlation with CCT. Attenuation of Wnt/β-catenin in high-Wnt pCCSC by shRNA-mediated downregulation or pharmacological inhibition significantly reduced tumor growth rates. Overall, the results of our study indicates (i) that early activation of Wnt/β-catenin signaling is critical for CCT, and (ii) that high levels of Wnt/β-catenin signaling can further demarcate ALDH+ tumor-initiating cells in the non-dysplastic epithelium of UC patients. As such, our findings offer plausible diagnostic markers and therapeutic target in the Wnt signaling pathway for early intervention in CCT.