Stem cell therapy has the potential to dramatically change the treatment of human disease. The Yamaguchi laboratory is studying how the Wnt family of signaling molecules regulates the growth and development of stem cells during early embryogenesis and tumorigenesis. The laboratory is particularly interested in stem cell populations that form the neural and musculoskeletal cells of the mammalian trunk and in the gut stem cells that form the colon. Wnt signaling has profound effects on the development of these populations. In the absence of a Wnt signal, the muscles, cartilage, and bone of the trunk fail to form, as well as the colon section of the intestinal tract. Aberrant activation of Wnt signaling causes major abnormalities in musculoskeletal development and promotes colon cancer. Understanding how Wnt signaling regulates stem cell pathways may reveal new mechanisms to program stem cells to form neural and musculoskeletal cells valuable in cell based therapies or how to target effective treatments for colon cancer. We are studying a unique stem cell known as the neuromesodermal (NM) stem cell that resides in the primitive streak (PS) of the gastrulating embryo and gives rise to the spinal cord and musculoskeletal progenitors of the body. Wnt3a is expressed in the PS and functions there to maintain NM stem cells and to control their differentiation however how Wnt3a simultaneously regulates both processes remains poorly understood. Wnt3a regulates cellular behavior by stabilizing beta-catenin, which interacts with members of the Lef/Tcf family of DNA-binding factors to transcriptionally activate target genes. The goals of the laboratory are: 1) to understand how NM stem cells in vivo make cell fate decisions based on the exposure to a single growth factor, Wnt3a. 2) to understand how Wnt3a regulates NM stem cell self-renewal and differentiation by elucidating Wnt3a-dependent transcription factors and gene regulatory networks (GRN). 3) to identify and understand the role of gut stem cells in colon formation and tumorigenesis We have made significant progress in achieving our goals: 1) We utilized cell specific markers and Tamoxifen-induced Cre-based lineage tracings to locate putative NM stem cells in specific germ layers of wildtype embryos. These data show that NM stem cells form the bulk of musculoskeletal and neural progenitors and provide functional evidence for NM stem cell location in both the epithelial and ingressed primitive streak. We further traced these stem cell populations in Wnt pathway mouse mutants to examine how the Wnt3a/beta-catenin signaling pathway regulates NM stem cell behavior. We have provided definitive proof that trunk progenitors can undergo """"""""transfating"""""""", i.e., progeny can enter the mesodermal germ layer but be redirected towards the neural lineage, in the absence of Wnt3a. Finally, we found that Wnt3a/beta-catenin signaling directs NM stem cells towards paraxial mesoderm cell fates;however, our data suggests a new view that Wnt3a positively supports both musculoskeletal and neural progenitors. 2) We have generated genome-wide transcriptional profiles of wildtype (wt) and Wnt3a-/- PS and identified 729 differentially expressed genes, including several previously characterized Wnt target genes. Although this approach has successfully identified many new Wnt3a targets, it remains a significant challenge to identify the genes that possess functional Wnt3a-like activity. We have therefore generated a series of ESCs carrying Doxycycline (Dox)-inducible epitope-tagged transgenes, to screen for Wnt target genes that regulate NM stem cell development. Since stimulation of ES cells with Wnt3a rapidly induces PSM and NM stem cells, we reasoned that the overexpression of transcriptional effectors of Wnt3a, in the absence of exogenous Wnt3a, should induce these cells. These studies have led us to focus on two particularly interesting downstream transcription factors, Mesogenin (Msgn1) and Sp5. Msgn1: We are studying Msgn1, a bHLH transcription factor, since Msgn1-/- mutants indicate that it is required for PSM differentiation, a process known to be controlled by Wnt3a. We have discovered that Msgn1 is a direct Wnt3a target gene and is sufficient to drive ESCs to undergo EMT, migration and differentiation into PSM. Using genomic approaches, we have found that Msgn1 directly binds enhancers to activate EMT and PSM-specific gene expression and have defined the specific Msgn1 binding site. Importantly, we have used gain-of-function (GOF) transgenics to show that Msgn1 induces PSM in vivo, and that the forced expression of Msgn1 in Wnt3a-/- mutants is capable of rescuing the PSM deficit. By combining Msgn1 overexpression with cell lineage tracing we have been able to show that ectopic expression of Msgn1 in NM stem cells in vivo reduced the contribution of progeny to the neural tube, and dramatically expanded the unsegmented mesenchymal PSM while blocking somitogenesis and notochord differentiation. This remarkable finding argues that Msgn1 is a master regulator of PSM differentiation. Our recent finding is highly significant since we now know how Wnt3a elicits PSM differentiation and cell motility from NM stem cells - by activating the expression of the PSM master regulator Msgn1. Sp5: The laboratory has also focused on the Sp1 family of Zinc-finger transcription factors. Sp5 is regulated by Wnt3a/beta-catenin signaling during gastrulation, and is expressed at multiple sites of Wnt activity, including normal and diseased embryonic and adult tissues, suggesting that Sp5 is a universal Wnt/beta-catenin target gene. We have shown that Sp5 functions redundantly with the closely related family member Sp8. Double mutants display a phenotype remarkably similar to Wnt3a mutants indicating that Sp5/Sp8 function in the self-renewal of NM stem cells. To construct the GRN that transduces Wnt3a activity in the PS, we have undertaken a large-scale RNA-Seq project to define the transcriptomes of control, Wnt3a, Lef1, Msgn1, and Sp5/Sp8 loss and gain-of-function mutants and ESCs and are currently integrating this data with Msgn1, Lef1, Sp5 and Sp8 ChIP-Seq data. Although still a work-in-progress, we have found that Sp5/8 bind directly to promoters of """"""""classic"""""""" Wnt target genes such as Axin2, and often at the same sites that Lef1 binds, to elevate transcription. Recent mechanistic studies indicate that Sp5 and Sp8 exist in a large multiprotein complex with beta-catenin and Mediator to bridge the enhanceosome and PolII and the basal transcriptional complex. Thus we have discovered that Sp5/8 are new components of the Wnt/beta-catenin signal transduction pathway. 3) We have identified a novel colon stem cell (CoSC) population that uses Wnt3a to signal directional growth to elongate the colon portion of the gut from the small intestine region. Surprisingly, Wnt3a has a specific and potent activity to stimulate growth of CoSC but not small intestine. This is particularly interesting since most human colon cancers are caused by altered Wnt signaling. Our proposal that Wnt3a could be the major growth stimulator for the colon region of the intestine could shed light on why aberrant Wnt signaling primarily causes colon cancer. We screened our embryonic Wnt target genes for expression in the murine adult intestine and in intestinal tumors. Remarkably, many of our genes are expressed in the stem cell compartment of the GI tract, and are highly expressed in adenomas caused by Wnt gain of function mutations. We have found that Sp5 alone is not required for spontaneous adenoma formation in the Apcmin mouse model of human intestinal cancer, suggesting that Sp5 may also have redundant functions during intestinal tumorigenesis. We are currently addressing the potential redundancy of Sp5 and Sp8 in intestinal cancer stem cells.
Showing the most recent 10 out of 20 publications
