Project 2: Self-Renewal and Differentiation: Molecular events that commit ES cells to exit the pluripotent state. (James Thomson, PI) A.
SPECIFIC AIMS I n low (4 ng/ml) or absent exogenous bFGF, BMP4 induces human ES cells to form a homogenous population of trophoblast, the outer layer of the placenta. However, we have recently found that in the presence of high (100 ng/ml) concentrations of bFGF, BMP4 instead induces human ES cells to form a population of cells that no longer express trophoblast markers, but instead transiently express brachyury, a mesoderm or mesendoderm marker, and subsequently express a mixture of endoderm and mesoderm markers. In this project, we will study how BMPs induce human ES cells to exit the pluripotent state and commit to differentiation, and study how FGF mediates these divergent developmental outcomes in response to the same inducer. Understanding how ES cells exit the pluripotent state and why this exit is generally irreversible is central to achieving efficient reprogramming (Project 3), and understanding these key early lineage decisions will allow more efficient differentiation to specific clinically-relevant lineages. We will accomplish the following aims:
Aim 1. We will establish a detailed time course of gene expression in human ES cells upon BMP4- induced differentiation, both in the presence (brachyury positive result) and absence (chorionic gonadotropin positive result) of bFGF, and correlate these changes with the commitment to exit the pluripotent state. Commitment will be measured by adding BMP4 for successively longer time periods, removing BMP4, and then examining how many cells retain markers of pluripotency several days later. The hypothesis of this aim is that the quantitative commitment curve will be most closely correlated with the expression levels of the genes that directly control these differentiation events. We will subsequently overexpress the transcription factors most closely associated with commitment, and we will downregulate genes by RNA interference that are downregulated during commitment to identify those which are sufficient to mediate differentiation to a brachyury-positive population. Preliminary results demonstrate that GATA2 and GATA3 are both individually sufficient to mediate trophoblast differentiation in the absence of bFGF, so we are optimistic that a single transcription factor will be sufficient to mediate human ES cell differentiation to a brachyury positive population in the presence of bFGF.
Aim 2. We will use whole genome chromatin immunoprecipitation on chip (ChlP-chip) to map the genomic binding sites of TGFp/Activin-activated Smad 2/3 and BMP-activated Smad 1/5/8 during BMP4- induced differentiation. The central hypothesis of this aim is that Smad 2/3 directly activates the expression of key pluripotency factors in human ES cells, directly suppresses the expression of genes that would otherwise promote differentiation, and competes at the same promoters with Smad 1/5/8 which has the opposing effects. A second hypothesis is that bFGF will change the DNA binding sites of Smad 1/5/8 during BMP-induced differentiation, leading to the different developmental outcomes observed.
Aim 3. We will use whole genome chromatin immunoprecipitation on chip (ChlP-chip) to map genomic binding sites of GATA2 and GATA3 during BMP4-induced differentiation. The hypothesis of this aim is that GATA2 and GAT A3 directly negatively regulate the transcription of key pluripotency genes, and directly positively regulate their own transcription, so that once they are induced by Smads, GAT A expression becomes self-sustaining and BMP-independent, and thus differentiation continues even upon BMP removal. A second hypothesis is that bFGF will change the DNA binding sites of GATA2/GATA3 during BMP-induced differentiation, leading to the different developmental outcomes observed.
Aim 4. We will identify proteins that are differentially phosphorylated between BMP4-induced trophoblast differentiation (no bFGF) and BMP-induced brachyury positive cellular differentiation (high bFGF) to identify mediators of FGF signaling that cause the switch between these divergent developmental outcomes. The hypothesis of this aim is that differential phosphorylation of a limited number of transcription factors is casually related to the different developmental outcomes observed after BMP induction in the presence or absence of bFGF. Differentiation commitment curves will again become key for focusing attention to relevant phosphorylation events.
Showing the most recent 10 out of 47 publications
