Read an Excerpt
Chapter 1: Regulatory transcription factors and cis-regulatory regions
Albert J. Courey
1. Introduction
In eukaryotes, the transcription of protein-encoding genes is directed by two classes of factors, the basal transcriptional apparatus and regulatory factors. The basal apparatus (reviewed in Chapter 1) is operationally defined as that set of factors required for accurately initiated basal transcription in a cell-free transcription system. The basal apparatus, which includes pol II and a number of general transcription factors, is directed to the 5' end of a transcription unit by the core promoter. Although core promoters can direct accurate initiation in vitro, they do not generally direct significant levels of transcription in vivo, especially in the context of an intact eukaryotic chromosome. Measurable transcription in vivo generally requires the action of regulatory factors, which interact either directly or indirectly with cis-regulatory elements and modulate the efficiency of transcription from linked core promoters.
Regulatory transcription factors can be grouped into two broad categories, the sequence-specific regulators and the coregulators (Berk, 1999; Mitchell and Tjian, 1989; Tjian and Maniatis, 1994; Triezenberg, 1995). As their name implies, sequence-specific regulators interact directly with cis-regulatory elements in a sequence-specific manner. In eukaryotic cells, these factors are generally modular in nature. In other words, they consist of independent structural domains that are separately responsible for the various biochemical functions of the factors. Sequencespecific regulators can be further subdivided into activators, which stimulate core promoter utilization, and repressors, which inhibit core promoter utilization. It is not always straightforward to classify a sequence-specific regulator as an activator or a repressor, since many such factors can mediate either activation or repression depending upon the context of their binding sites, or upon the array of interacting ligands available to modulate their conformation and/or activity. In contrast to sequence-specific regulators, coregulators do not interact with DNA directly. Rather, they are directed to regulatory targets by protein-protein interactions with sequence-specific regulators. Like sequence-specific regulators, coregulators can be further categorized according to whether their role is in gene activation, in which case they are termed coactivators, or in gene repression, in which case they are termed corepressors.
The cis-regulatory regions with which regulatory factors interact can be extremely complex in nature. Regulatory transcription factor binding sites are usually grouped together into regulatory modules. Such modules are referred to as enhancers, if their major role is in transcriptional activation, or as silencers, if their major role is in transcriptional repression. Many genes contain multiple enhancers and/or silencers, which can be spread out over extremely large regions (100 kilobases or more). In any given complex locus, different regulatory modules can be responsible for controlling transcription at different times during development, at different locations in a complex organism, or in response to different environmental stimuli.
2. Regulatory transcription factors
2.1 The modular nature of sequence-specific regulators
The modular nature of eukaryotic sequence-specific regulators was first recognized through studies of yeast activators such as Gcn4 (Hope and Struhl, 1986) and Ga14 (Ma and Ptashne, 1987a), and mammalian activators such as glucocorticoid receptor (Miesfeld et al., 1987) and Spl (Courey and Tjian, 1988). These studies showed that the factors contained separable DNA-binding and activation domains, which could be mixed and matched in various domain-swapping experiments to generate functional chimeric transcription factors. Although we now take the modularity of these factors for granted, the initial discovery of this phenomenon came as a surprise. This was perhaps because of the impression left from studies of enzyme structure suggesting that proteins generally consisted of single cooperatively folding units that were stabilized by long-distance tertiary interactions. In addition, the modularity of eukaryotic factors appeared to stand in contrast to what had been learned from studies of certain paradigmatic prokaryotic transcriptional regulators, such as X-repressor, in which residues responsible for transcriptional activation were found to be intimately linked to residues responsible for sequence-specific DNA recognition (Ptashne, 1986). The modularity of eukaryotic regulatory factors is likely a manifestation of the phenomenon of exon shuffling, whereby complex multifunctional proteins are thought to have evolved by the joining together of genes encoding more primitive monofunctional proteins (de Souza et al., 1996).
As the foregoing discussion implies, most eukaryotic sequence-specific regulators consist of a single, independently folding DNA-binding domain as well as one or more independently folding regulatory domains. Regulatory domains can be further categorized as activation domains or repression domains...
