Abstract Smad2 and Smad3 are intracellular mediators of transforming growth factor β (TGFβ) signaling that share various biochemical properties, but data emerging from functional analyses in several cell types indicate that these two Smad proteins may convey distinct cellular responses. Therefore, we have investigated the individual roles of Smad2 and Smad3 in mediating the cytostatic and proapoptotic effects of TGFβ as well as their function in epithelial-to-mesenchymal transition. For this purpose, we transiently depleted mouse mammary epithelial cells (Nme) of Smad2 and/or Smad3 mainly by a strategy relying on RNaseH-induced degradation of mRNA. The effect of such depletion on hallmark events of TGFβ-driven epithelial-to-mesenchymal transition was analyzed, including dissolution of epithelial junctions, formation of stress fibers and focal adhesions, activation of metalloproteinases, and transcriptional regulation of acknowledged target genes. Furthermore, we investigated the effect of Smad2 and Smad3 knockdown on the TGFβ-regulated transcriptome by microarray analysis. Our results identify Smad3 as a key factor to trigger TGFβ-regulated events and ascribe tumor suppressor as well as oncogenic activities to this protein. (Mol Cancer Res 2009;7(8):1342–53)

Supplementary Data from Smad3 Is a Key Nonredundant Mediator of Transforming Growth Factor β Signaling in Nme Mouse Mammary Epithelial Cells

Insulin/insulin-like growth factor (IGF) signaling constitutes an evolutionarily conserved pathway that controls growth, energy homeostasis, and longevity. In Drosophila melanogaster, key components of this pathway are the insulin-like peptides (Dilps). The major source of Dilps is a cluster of large neurons in the brain, the insulin-producing cells (IPCs). The genetic control of IPC development and function is poorly understood. Here, we demonstrate that the Pax6 homolog Eyeless is required in the IPCs to control their differentiation and function. Loss of eyeless results in phenotypes associated with loss of insulin signaling, including decreased animal size and increased carbohydrate levels in larval hemolymph. We show that mutations in eyeless lead to defective differentiation and morphologically abnormal IPCs. We also demonstrate that Eyeless controls IPC function by the direct transcriptional control of one of the major Dilps, dilp5. We propose that Eyeless has an evolutionarily conserved role in IPCs with remarkable similarities to the role of vertebrate Pax6 in β cells of the pancreas.

Remarkable progress has been made in understanding the molecular mechanisms underlying left-right asymmetry in vertebrate animal models but little is known on left-right axis formation in humans. Previously, we identified SESN1 (also known as PA26) as a candidate gene for heterotaxia by positional cloning of the breakpoint regions of a de novo translocation in a heterotaxia patient. In this study, we show by means of a zebrafish sesn1-knockdown model that Sesn1 is required for normal embryonic left-right determination. In this model, developmental defects and expression data of genes implicated in vertebrate left-right asymmetry indicate a role for Sesn1 in mediating Nodal signaling. In the lateral plate mesoderm, Nodal signaling plays a central role in left-right axis formation in vertebrates and is mediated by FoxH1 transcriptional induction. In line with this, we show that Sesn1 physically interacts with FoxH1 or a FoxH1-containing complex. Mutation analysis in a panel of 234 patients with isolated heterotaxia did not reveal mutations, indicating that these are only exceptional causes of human heterotaxia. In this study, we identify SESN1 as an indispensable gene for vertebrate left-right asymmetry and a new player in mediating Nodal signaling.

Ligand-bound receptors of the Transforming Growth Factor-beta (TGF-beta) family promote the formation of complexes between Smad proteins that subsequently accumulate in the nucleus and interact there with other transcriptional regulators, leading to modulation of target gene expression. We identified a novel nuclear protein, Smicl, which binds to Smad proteins. Smicl and Smads cooperate and enhance TGF-beta mediated activation of a Smad-responsive reporter gene. A domain with five CCCH-type zinc fingers in Smicl is structurally and functionally, at least in vitro, similar to a domain in CPSF-30, the 30 kDa subunit of Cleavage and Polyadenylation Specificity Factor (CPSF). Like CPSF-30, Smicl can associate with some other CPSF subunits characterized previously. Its effect on the induction of a reporter gene for TGF-beta requires the cleavage/polyadenylation signal downstream of the coding sequence of that gene. Thus, Smicl is a novel protein that displays CPSF-30-like activities, interacts in the nucleus with activated Smads, and potentiates in TGF-beta stimulated cells Smad-dependent transcriptional responses, possibly in conjunction with the activity of CPSF complexes.

BACKGROUND Smad proteins are intracellular mediators of transforming growth factor-beta (TGFbeta) signalling that regulate gene expression by interacting with different classes of transcription factors including DNA-binding multi-zinc finger proteins. One of these, Smad interacting protein 1 (SIP1), is a novel two-handed zinc-finger protein that displays strong similarity with the transcriptional repressor delta-crystallin enhancer binding factor (deltaEF1). Here, we summarize what is known about the mechanism of action of both proteins and their role in vertebrate embryogenesis. Our data are discussed together with the present knowledge on other zinc-finger containing Smad interacting proteins. METHODS The activities and function of SIP1 have been analysed through documentation of expression patterns, the effect of over-expression of SIP1 on target-gene expression, and promoter studies using Xenopus embryos. Moreover, S1P1/Smad complexes and their association with target promoter DNA were analyzed in biochemical studies. RESULTS SIP1 is a transcriptional repressor displaying overlapping DNA binding specificities with deltaEF1. An in vivo target of SIP1 in Xenopus is a gene required for the formation of mesoderm, Brachyury (XBra). Our data indicate that SIP1 is required to confine XBra gene expression to the mesoderm. Furthermore, the expression pattern in Xenopus invites us to speculate that SIP1 plays a role in specification/differentiation of neuroectoderm. Unlike deltaEF1, SIP1 can bind directly to activated receptor regulated Smads (R-Smads) and recruit them to the DNA. This indicates that Smads may modulate the activity of SIP1 as a transcriptional repressor. CONCLUSIONS Our data point to a role of SIP1 in developmental processes regulated by members of the TGFbeta family such as induction of mesoderm (mediated through activin-like signalling) and inhibition of neuroectoderm formation (mediated by bone morphogenetic proteins [BMPs]). Whereas SIP1 could act in TGFbeta signal transduction by virtue of interaction with activated R-Smads, genetic studies in the mouse indicate that deltaEF1 may act in certain TGFbeta pathways-i.e., BMPs and growth and differentiation factors (GDFs)-as well. The molecular mechanisms by which these transcriptional repressors act, as well as the function of the SIP1/Smad interaction, remain to be elucidated. CLINICAL RELEVANCE Mutations in components of the TGFbeta signalling pathways have been associated with disease and congenital malformations. We anticipate that identification of Smad interacting transcription factors including SIP1 and their targets will help us to understand the molecular basis of certain pathologies.