Three Habits of Highly Effective Signaling Pathways: Principles of Transcriptional Control by Developmental Cell Signaling
Total Page:16
File Type:pdf, Size:1020Kb
Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press REVIEW Three habits of highly effective signaling pathways: principles of transcriptional control by developmental cell signaling Scott Barolo and James W. Posakony1 Division of Biology/CDB, University of California San Diego, La Jolla, California 92093-0349, USA Seven major cell–cell signaling pathways—Wnt, TGF-, naling ligand to the pathway’s receptor. The transcrip- Hedgehog (Hh), receptor tyrosine kinase (RTK), nuclear tional control function of each pathway is carried out by receptor, Jak/STAT, and Notch—control the vast major- one or more signal-regulated transcription factors, which ity of cell fate decisions during the development of bila- bind to specific signaling pathway response elements terian animals (Gerhart 1999). Each pathway is used re- (SPREs) in the promoters or enhancers of target genes. peatedly during the development of a given organism, Over the last few years, an unexpected theme has activating different subsets of target genes in different emerged: Several of these developmental control path- developmental contexts. These seven pathways are ways have been found to use transcriptional switch strikingly diverse in both their complexity and the bio- mechanisms, whereby target genes are activated in the chemical mechanism of signal transduction, ranging presence of signaling, but repressed in its absence (Fig. 1). from direct transcriptional regulation by the nuclear re- We use the term default repression to refer to the tran- ceptor proteins to the extended protein phosphorylation scriptional repression of target genes in the absence of cascades characteristic of RTK pathways. Nevertheless, signaling. A particularly exciting finding in the case of the primary consequence of signaling is the same: acti- the Notch, Wnt, Hh, and nuclear receptor pathways is vation of specific target genes by signal-regulated tran- that default repression and signal-dependent activation scription factors. of target genes are both mediated not only by the same Recent work has revealed several surprising and fun- SPREs but, in fact, by the same transcription factors— damental commonalities in the transcriptional mecha- Su(H)/CBF1 in the Notch pathway, Lef/Tcf proteins in nisms by which these pathways control the expression of the Wnt pathway, Ci/Gli in the Hh pathway, and nuclear their target genes. In this review, we discuss transcrip- receptors themselves. The fundamentally different tional regulation by developmental cell signaling path- mechanisms by which these four pathways convert ways, and suggest that three functional properties—ac- SPRE-binding transcription factors from repressors to ac- tivator insufficiency, cooperative activation, and default tivators (diagrammed in simplified form in Fig. 1) suggest repression—are shared among the major pathways. To- that each signaling pathway has independently evolved a gether, these three “habits” allow signaling pathways to transcriptional switch. Others of the major signaling sys- strongly activate target genes in their proper context, tems, including the TGF- and RTK-type pathways, ap- while preventing their expression in all other cells. Such pear to extend the known prevalence of this switch strict control over target gene expression explains an ex- mechanism by making use of yet another strategy for traordinary feature of developmental cell signaling: the repression via their SPREs, to be discussed below in this capacity of a single pathway to elicit a large variety of review. gene expression patterns, and hence to control the speci- Thus, a remarkable generalization emerges: Of the ma- fication of a large variety of cell fates. jor signaling pathways that control cell fate in bilaterians (and possibly in all metazoans), most, and perhaps all, act by switching their target genes from a state of transcrip- Signaling flips a switch tional repression to one of activation, often using the very same signal-regulated transcription factor and the All major developmental signaling pathways function at same binding sites (SPREs) to do so. Although the term least in part by regulating the transcription of specific switch implies that target genes in an individual signal- nuclear target genes in response to the binding of a sig- responsive cell are converted from repression to activa- tion by a signaling event, this has not been directly 1Corresponding author. shown in vivo. However, the existence of the switch is E-MAIL [email protected]; FAX (858) 534-0566. Article and publication are at http://www.genesdev.org/cgi/doi/10.1101/ strongly supported by a substantial body of evidence (de- gad.976502. scribed below) that (1) the same SPREs can mediate re- GENES & DEVELOPMENT 16:1167–1181 © 2002 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/02 $5.00; www.genesdev.org 1167 Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press Barolo and Posakony Figure 1. Major developmental signaling pathways use different mechanisms to switch from transcriptional repression to activation of target genes. Signal-regulated transcription factors for the Notch, Wnt, Hh, and nuclear receptor signaling pathways bind signaling pathway response elements (SPREs) and function as transcriptional repressors in the absence of signaling (default repression), but act as activators in response to ligand. Signal transduction mechanisms and mechanisms of transcriptional switching are depicted in simplified form. See text for discussion. pression and activation in different cells of the same tis- pathways’ ability to maintain strict control of target sue, depending on the state of signaling in those cells; gene expression. and (2) a transcriptional switch can be thrown experi- mentally in normally nonresponsive cells by ectopically activating the signaling pathway in those cells. Habit #1: activator insufficiency Default transcriptional repression via SPREs is clearly effective for keeping target genes turned off where there “If the prolactin, growth hormone, and possibly TSH is no signal, but it raises the question of why it has genes require the same factor . for their activation, evolved independently to become part of the regulatory then restrictive mechanisms, whether repression or a re- mechanism underlying the operation of diverse signaling quirement for additional interactive factors, must ac- pathways. In other words, what is the selective advan- count for their differential expression. .” tage of default repression? Why isn’t the absence of path- Nelson et al. (1988) way activation sufficient to keep target genes turned off? Recent advances in understanding the specificity of gene “How might the siamois promoter become unresponsive activation by signaling pathways offer, we think, impor- to Wnt signals, even when these are presented in the tant insight into the answers to these questions. There is form of activated Tcf? One can imagine two classes of growing evidence that signal-regulated transcription fac- models. On one hand, promoter activation might require tors share two other functional characteristics, activator additional inputs. .Alternatively, loss of Wnt respon- insufficiency and cooperative activation, which, to- siveness could result from repression. .” gether with default repression, may explain signaling Darken and Wilson (2001) 1168 GENES & DEVELOPMENT Downloaded from genesdev.cshlp.org on September 27, 2021 - Published by Cold Spring Harbor Laboratory Press Transcriptional control by developmental signaling pathways Signaling pathway activity is generally insufficient Because in general each target gene’s response to sig- to activate target genes naling pathway activity is unique, the mechanisms that have evolved to limit the activation of targets to their A critical, but perhaps not fully appreciated, property of proper developmental contexts must operate at the level developmental signaling systems is the selective tran- of individual promoters. The possible mechanisms fall scriptional responsiveness of target genes to pathway ac- into two classes: active repression of target promoters/ tivity. The Notch system, which is used to control a enhancers in all inappropriate signaling contexts, and in- diverse array of cell fate specification events during the ability of signal-regulated transcription factors to acti- development of a single organism, offers an instructive vate transcription in any but the correct contexts. Both example. No single Notch signaling event activates ev- models require the presence of specific cis-regulatory el- ery Notch target gene; conversely, no single known tar- ements in signal-regulated enhancers, binding either get gene is activated by all Notch signaling events. In transcriptional repressors to restrict gene expression, or Drosophila, a substantial number of genes have been additional activators to facilitate expression, respec- identified that are directly activated by Suppressor of tively. A direct experimental approach to discriminating Hairless, Su(H), the key signal-regulated transcription these possibilities is to delete specific sequences in sig- factor at the terminus of the Notch pathway, in response nal-regulated enhancers and promoters, and observe the to Notch signaling activity. These include at least 10 effects on reporter gene expression in vivo. The active genes in the Enhancer of split complex, as well as vesti- repression model predicts that signal-regulated enhanc-