TGF-β control of stem cell differentiation genes

Abstract

The canonical TGF-β/Smad signaling pathway was delineated in the mid 90s and enriched over the past decade with many findings about its specificity, regulation, networking, and malfunctions in disease. However, a growing understanding of the chromatin status of a critical class of TGF-β target genes - the genes controlling differentiation of embryonic stem cells - recently prompted a reexamination of this pathway and its critical role in the regulation of stem cell differentiation. The new findings reveal master regulators of the pluripotent state set the stage for Smad-mediated activation of master regulators of the next differentiation stage. Furthermore, a novel branch of the TGF-β/Smad pathway has been identified in which a chromatin-reading Smad complex makes the master differentiation genes accessible to canonical Smad complexes for transcriptional activation. These findings provide exciting new insights into the global role of TGF-β signaling in the regulators of stem cell fate.

Figure 1. Model of nodal regulation of gene expression in embryonic stem cells

A. Active target genes. Nodal target genes that control cell homeostasis or Smad pathway feedback (e.g. SerpinE1 and Smad7, respectively) are in a transcriptionally active state with the Smad binding sites exposed. Signal-driven Smad4-Smad2/3 complexes, in associated with DNA binding cofactors (“xyz”), readily bind to these sites for stimulation of transcription. B. Poised target genes. Nodal target genes that control ESC differentiation (e.g. Gsc and Mixl1) have Smad binding sites secluded by repressive chromatin. The Oct4-Sox2-nanog complex, which is an enforcer of pluripotency in ESCs, recruits SetDB1 to differentiation genes. SetDB1 catalyzes Lys9 trimethylation on histone H3 (H3K9me3 motif), which binds HP1 for chromatin compaction and gene repression. When H3K9me3 is accompanied with unmodified K4 and acetylated K18 (H3K4-K9me3-K18ac motif), it provides a platform for activation of Gsc and Mixl1 by nodal signals. The TRIM33-Smad2/3 complex binds to H3K4-K9me3-K18ac, displacing HP1 to relax the chromatin and provide Smad4-Smad2/3 with access to its binding sites. FoxH1 is a Smad cofactor for specific recognition of these sites. Thus, Gsc and Mixl1 in ESCs are in a “poised” state that is silent but ready for nodal-driven activation through the joint action of a TRIM33-Smad2/3 chromatin-binding complex and a Smad4-Smad2/3 Pol II activating complex.

Figure 2. Complementary branches of the TGF-β/Smad pathway in embryonic stem cells

TGF-βsignaling involves the binding of ligand to two pairs of receptor serine/threonine kinases for the assembly of a receptor complex that phosphorylates RSmad proteins (Smads 2 and 3 in the case of TGF-β and nodal receptors). The receptor-phosphorylated RSmads bind to Smad4, assembling complexes that recognize specific promoter elements to stimulate or inhibit transcription. In a newfound second branch of this pathway, receptor-phosphorylated Smad2/3 bind to TRIM33, assembling complexes that recognize certain repressive marks on the chromatin. The Smad4-Smad2/3 complex is necessary and sufficient for TGF-β regulation of cell homeostasis genes that are in an active state. However, master differentiation genes in ESCs are secluded by repressive chromatin marks. Activation of these genes by TGF-β signals requires TRIM33-Smad2/3 in addition to the Smad4-Smad2/3. Smad4-independent gene responses may also exist.