HOME :: CHAPTER 5 :: PROMOTER STRUCTURE AND THE MECHANISMS OF TRANSCRIPTION COMPLEX ASSEMBLY :: PROMOTER STRUCTURE

Promoter Structure

Promoters of genes that transcribe relatively large amounts of mRNA have similar structures. They have a TATA sequence (sometimes called the TATA box or Goldberg-Hogness box) about 30 base pairs upstream from the site where transcription begins, as well as one or more promoter elements further upstream (Figure 1; Grosschedl and Birnstiel 1980; McKnight and Tjian 1986). The "functional anatomy" of a promoter region can be analyzed by determining which of its bases are necessary for efficient transcription. Cloned genes can be accurately transcribed when they are placed into the nuclei of frog oocytes or fibroblasts or when they are incubated with RNA polymerase in the presence of nucleotides and nuclear extracts (Wasylyk et al. 1980). After the transcription of a gene is confirmed, one uses restriction enzymes to make specific deletions in the gene or in the regions surrounding it. One can then see whether such a modified gene will still be accurately transcribed. Such studies on β–globin genes (Grosveld et al. 1982; Dierks et al. 1983) showed that the first 109 base pairs preceding the cap site were sufficient for the correct initiation of β–globin gene transcription by RNA polymerase.

Figure 1 Typical promoter region for a protein-coding eukaryotic gene. The gene diagrammed here contains a TATA box and three upstream promoter elements. Examples of such upstream elements are shown beneath the diagram. (After Maniatis et al. 1987.)

Myers and co-workers (1986) refined this analysis by cloning the region of a mouse β–globin gene from 106 base pairs upstream from the start of transcription (–106) through the first 475 base pairs (+475) of the first exon. These clones were subjected to in vitro mutagenesis (wherein specific mutations can be placed into a cloned gene). In this way, 130 different single-base substitutions were introduced into the promoter region of the globin gene. These cloned genes were placed into plasmids containing an enhancer from a gene normally expressed in all tissues. The recombinant plasmids were then introduced by transfection into cultured cells that do not usually produce globin. Would the cells transcribe a truncated globin message (475 bases) from the clones? Figure 2 shows the results. In most cases, mutating a base in the 5′ flanking region did not affect the efficiency of globin gene transcription. But there were three clusters of nucleotides in which mutations drastically reduced transcription. One cluster was in the TATA box, another was in the CAAT upstream promoter element, and a third was in the CACCC region, about 95 to 87 base pairs upstream from the cap site.

Figure 2 The effect of specific point mutations in the mouse β–globin promoter on the ability of that promoter to initiate transcription. Each line represents the transcription level of a mutant promoter relative to the transcription level of a wild-type globin promoter assayed concurrently. The black dots represent nucleotides for which no mutation had been generated. Below the histogram is a diagram showing the position of the TATA box and the two upstream promoter elements of the mouse β-globin gene. (After Myers et al. 1986.)

The CAAT and TATA boxes have been found to be critical elements in numerous eukaryotic promoters (Efstratiadis et al. 1980), but the CACCC sequence is seldom seen except in the β–globin gene promoters in several species. In humans, this sequence appears to be critical. A naturally occurring mutation in this sequence causes a total loss of β–globin gene transcription (Orkin and Kazazian 1984), and this sequence is recognized by an erythroid-specific transcription factor (Mantovani et al. 1988). Two mutations, at positions –78 and –79, have actually increased transcription to three times the wild-type level. It is thought that these changes facilitate the interaction of the promoter with the trans-regulatory proteins.

Promoter Function

Promoters can function not only to bind RNA polymerase, but also to specify the places and times that transcription can occur from that gene. This function of promoters can be vividly demonstrated in certain transgenic animals. Here, a new gene is constructed wherein the normal promoter of a particular gene is replaced with the promoter of some other gene, and the fused gene is placed into a pronucleus of a mammalian zygote. Palmiter and co-workers (1982) isolated the rat growth hormone gene and deleted its 5′ promoter region. Into this space, they substituted the promoter sequence of another gene—Mt-1 for mouse metallothionein 1, a small protein involved in regulating serum zinc levels. This hybrid gene is shown in Figure 3A. The Mt-1 gene can be induced by the presence of heavy metals such as zinc or cadmium, and the sequences responsible for this induction are in the promoter of this gene. By fusing this metallothionein promoter region to the rat growth hormone gene (rGH), the rat growth hormone gene is placed under the control of the metallothionein promoter. In this case, rat growth hormone message should be made when the Mt-1 promoter is activated by the presence of zinc or cadmium.

A plasmid containing this fused gene was grown in bacteria (see Chapter 2), the Mt-1/rGH piece was isolated, and about 600 copies of this fragment were injected into pronuclei of recently fertilized mouse eggs. DNA hybridization showed that many of these newborn mice had incorporated numerous copies of the rat growth hormone gene into their chromosomes. These transgenic mice were then fed a diet supplemented with zinc. The zinc induced large amounts of rat growth hormone to be secreted by the livers of these mice. (The liver is where metallothionein is usually made. Growth hormone is usually secreted from the pituitary gland.) The amount of growth hormone secreted correlated with the size of these mice. The transgenic mice became enormous, up to 80 percent larger than their normal littermates (Figure 3B).* The metallothionein promoter regulated the synthesis of growth hormone in these transgenic mice.

Figure 3 Promoter function seen in transgenic mice. (A) Recombinant plasmid containing rat growth hormone structural gene, mouse metallothionein regulatory region, and bacterial plasmid pBR322. The plasmid, pMGH, was injected into the mouse oocytes. The dark boxes on the injected plasmid correspond to the exons of the GH gene. The direction of transcription is indicated by an arrow. (B) A mouse derived from the eggs injected with pMGH (left) and a normal littermate (right). (From Palmiter et al. 1982, photograph courtesy of R. L. Brinster.)

RNA Polymerase and the trans-Regulatory Factors at the Promoter

Transcription requires the interaction of RNA polymerase with promoter DNA. In eukaryotic cells, there are three different types of RNA polymerases, each having particular functions and properties (Rutter et al. 1976). RNA polymerase I is found in the nucleolar region of the nucleus and is responsible for transcribing the large ribosomal RNAs; RNA polymerase II transcribes messenger RNA precursors; and RNA polymerase III transcribes small RNAs such as transfer RNA, 5S ribosomal RNA, and other small DNA sequences.** None of the eukaryotic RNA polymerases can bind efficiently to DNA. Rather, there are families of DNA-binding proteins that first bind to DNA and, once bound, interact with the RNA polymerase to initiate RNA synthesis.

The TATA element and RNA polymerase II

The classic diagram of transcription shows that DNA, in the presence of RNA polymerase and ribonucleoside triphosphates, transcribes RNA molecules. But this simple scheme does not account for the difficulties in (1) getting the RNA to start in the correct place, and (2) having transcription of a particular gene occur only at particular times and in particular types of cells. There have to be factors that enable RNA polymerase to bind solely to the promoters of particular genes. Here, we will discuss those proteins and DNA sequences that localize the RNA polymerase to the promoter sites. The enzyme responsible for the transcription of messenger RNAs is RNA polymerase II. However, accurate transcription of cloned genes in vitro will not occur if the genes are incubated with purified RNA polymerase II and nucleoside triphosphates. Nuclear extracts must be added if accurate transcription is to commence. What are these factors that allow transcription to be initiated? At least six nuclear proteins have been shown to be necessary for the proper initiation of transcription by RNA polymerase II (Figure 4; Buratowski et al. 1989; Sopta et al. 1989).

Figure 4 The formation of the active eukaryotic initiation complex. The diagrams represent the complexes formed on the TATA box by the transcription factors and RNA polymerase II. (A) The TFIID complex binds to the TATA box through its TBP subunit. (B) TFIID is stabilized by TFIIA. (C) TFIIB and TFIIH join the complex on the TATA box while TFIIE and TFIIF associate with RNA polymerase II. (D) RNA polymerase is positioned by TFIIB, and its carboxy-terminal domain (CTD) is bound by TFIID. (E) The CTD is phosphorylated by TFIIH and is released by TFIID. The RNA polymerase II is now competent to transcribe mRNA from the gene.

TFIID and TFIIA.*** In the first step of mRNA transcription, the TFIID complex binds to the TATA box. This was shown by DNase protection experiments in which TFIID was added to cloned genes and the DNA was then digested with DNase. The only way to "rescue" the DNA from digestion was for the TFIID to bind to the DNA, thereby preventing the DNase from reaching it. In this way, Sawadogo and Roeder (1984) showed that TFIID specifically bound to the TATA region of the genes. TFIID is a multimeric protein, one of whose components—the TATA-binding protein (TBP)—binds directly into the minor groove of the TATA sequence (Lee et al. 1991; Starr and Hawley 1991). The TFIID complex has several activities; the first is to bind the TATA sequence and to serve as the foundation for the transcriptional complex. Another role of TFIID is to prevent the stabilization of nucleosomes in the promoter region. When promoter-containing DNA is incorporated into nucleosomes, these genes are unable to be transcribed when TFIID, RNA polymerase II, and other factors are added later. However, when TFIID is added to the genes before or during nucleosome formation, the resulting chromatin is transcriptionally active (Workman and Roeder 1987). In blocking nucleosome production, TFIID appears to act antagonistically to histone H1. Histone H1 (as we will see in the next chapter) stabilizes nucleosomes and prevents transcription in the region where it binds. The addition of histone H1 prevents TFIID from finding the TATA sites and prevents transcription; however, this inhibition is overcome if TFIID is added first (Laybourn and Kadonaga 1991). The binding of TFIID is facilitated and stabilized by the transcription factor TFIIA (Buratowski et al. 1989; Maldonado et al. 1990). Many transcription factors activate transcription function by recruiting TFIID and "activating" it so that TFIID can bind the other members of the transcription complex (Chi and Carey 1996; Stargell and Struhl 1996). Thus, the decision to transcribe or not to transcribe a particular gene often depends on the balance between inhibitory factors (such as histones) and TFIID and TFIIA.

TFIIB and RNA polymerase II. The TFIID/TFIIA complex cannot form a stable complex directly with RNA polymerase II. Rather, TFIID binds factor TFIIB. The binding of TFIIB to TFIID appears to be the key rate-limiting step in the transcription of numerous genes. This rate can be dramatically increased by the proximity of certain promoter—and enhancer-binding transcription factors. These transcription factors are sequence-specific and can regulate which genes will be transcribed. The activator domain of these transcription factors binds directly to TFIIB and facilitates its assembly with TFIID (Lin and Greene 1991; Lin et al. 1991). Once TFIIB is in place, it can bind RNA polymerase II. Most of the RNA polymerase is positioned by its interaction with TFIIB, but the carboxy-terminal tail of the large subunit of RNA polymerase II interacts directly with TFIID (see Figure 4D). In this way, RNA polymerase II is placed at the promoter.

TFIIE/F and TFIIH. Either directly before or during its binding to TFIIB, RNA polymerase II becomes associated with TFIIF and TFIIE (Buratowski et al. 1991; Conaway et al. 1991). TFIIF has an enzymatic activity needed to unwind the DNA helix. TFIIE is a DNA-dependent ATPase and is probably needed for generating the energy for transcription (Bunick et al. 1982; Sawadogo and Roeder 1984). But what good is all this if the RNA polymerase remains bound to this complex on the TATA box? For transcription to occur, the RNA polymerase must be released from the promoter region. This releasing activity appears to be the function of TFIIH. The RNA polymerase is tightly bound by its carboxy-terminal domain (CTD) to TFIID. However, TFIID will only bind the unphosphorylated form of the CTD. In mammals, the CTD contains 52 repeats of the seven-amino-acid sequence YSPTSPS. When the initiation complex is formed, the completed complex activates the serine/threonine protein kinase activity of TFIIH. TFIIH then phosphorylates each of the 52 repeats (see Figure 4E; Koleske et al. 1992; Lu et al. 1992; Usheva et al. 1992). TFIID cannot bind this heavily phosphorylated region and releases the RNA polymerase. While the first phosphodiester bond can be made without the phosphorylation of the CTD, this phosphorylation appears to be essential for the further transcription of messenger RNA (Akoulitchev et al. 1995).

TAFs and the activation of basal transcription. TFIID is a multimeric protein, only one of its subunits actually binding to the TATA sequence. Some of the other subunits are called TATA-binding protein-associated factors (TAFs). Purification of the TAFs from human and Drosophila TFIID showed that they were composed of a similar set of proteins (Figure 5; Dynlacht et al. 1991). These TAFs are thought to serve two functions: (1) they can determine whether or not the TFIID remains on the promoter, and (2) they can function as co-activators, bridging the enhancer-bound proteins to the transcription complex through protein–protein interactions.

Figure 5 Outline of experiments suggesting that different TAFs interacted with different transcription factors to activate transcription. The full complement of TAFs is shown in (D). The "activator" in (D) is a protein bound to a DNA sequence that has been stabilized by the other interactions. (After Chen et al. 1994.)

Whether the TATA-binding protein stays on the promoter is of great concern to a gene. If it comes off, the gene will not be transcribed. Verrijzer and colleagues (1995) have shown that the 250- and 150-kDa TAFs are critical in determining whether the TBP remains bound at the TATA box. These TAFs recognize upstream elements of the promoter, which, if present, stabilize or destabilize the TBP at the promoter. This means that some promoters are intrinsically more "difficult" to transcribe from and that certain factors might have to be present to make these promoters transcribable. As we will see later, some promoters (such as the one for human interferon–β), are transcribed only after a great amount of effort has been put into bending them, contorting them, and bracing up the fragile transcription complex.

The association of TBP with different TAFs enables the transcription complex to be activated by proteins bound at the enhancer and upstream promoter sites. Moreover, different TAFs are able to co-activate with different trans factors. For instance, one of the most common transcription factors is Sp1. This non-TAF protein binds to the promoter or enhancer GGGCGG sequences through its carboxyl end but regulates transcriptional activity via its amino terminus (Dynan and Tjian 1985; Kadonaga et al. 1988). This factor is probably found in all cells and so cannot regulate differential gene expression. However, it appears to be involved with the interactions between the promoter region and the enhancer region in ways that do result in the differential transcription of particular genes in particular cells. Sp1 needs to bind to the 110-kDa TAF for its activation of the transcription complex. Thus, this TAF bridges Sp1 to the TBP, forming a loop in the DNA (Hoey et al. 1993; Chen et al. 1994). TAFs would enable the looping of DNA so that Sp1 elements in the enhancer would meet the TFIID protein at the promoter (see Figure 5). The Bicoid transcription factor binds to the 110-kDa and 60-kDa TAFs, and mutations in either of these TAFs reduce Bicoid-dependent transcription in Drosophila embryos (Sauer et al. 1996). Similarly, the Drosophila NTF-1 transcription factor binds to both the 60-kDa and 150-kDa TAFs, and in humans, the transcriptional activation by the estrogen receptor is accomplished by its binding to a 30-kDa TAF (Jacq et al. 1994). Indeed, Jacq and colleagues found that not all TBPs had the 30-kDa TAF. It appears that there are some TAFs that are found in all TFIIDs, while other TAFs may be more specific.

Promoters lacking TATA elements

There are many genes (mostly those encoding general metabolic proteins and not cell-specific proteins) that use RNA polymerase II, but whose promoters lack the TATA sequence. In these cases, some other protein binds to the promoter region. These are usually general promoter-binding proteins such as Sp1. The Sp1 protein on the GC-rich promoter element then binds TFIID either directly or through a TAF. The TFIID is now able to begin the cascade of factors that will form the transcription initiation complex and bind an RNA polymerase II protein to the promoter region (Figure 6; Pugh and Tjian 1991; Rigby 1993). So even though these promoters lack a TATA box sequence, TFIID is still the deciding factor in regulating whether transcription occurs.

Figure 6 Possible configuration for transcription factors mediating RNA polymerase II binding to a TATA-less promoter containing an Sp1-binding site. (After Pugh and Tjian 1991; Comai et al. 1992.)

*. Two other things grew with this experiment. The first is our potential ability to cure genetic disease by fertilizing eggs in vitro and injecting a normal gene into a pronucleus. These eggs can begin their development and then be returned to the woman's uterus. The second thing that grew was our responsibility (which usually is proportional to our power, whether we like it or not).

**. In most cells, ribosomal and transfer RNAs are constitutively synthesized. However, animals have evolved remarkable mechanisms for upregulating rRNA synthesis in their oocytes. We will therefore postpone our discussion of RNA polymerases I and III until we detail the events of oogenesis in Chapter 22.

***. TF stands for transcription factor; II indicates that the factor was first found to be needed for RNA polymerase II; and the letter designations refer to fractions from phosphocellulose columns that had the activity.

© All the material on this website is protected by copyright. It may not be reproduced in any form without permission from the copyright holder.

HOME :: CHAPTER 5 :: PROMOTER STRUCTURE AND THE MECHANISMS OF TRANSCRIPTION COMPLEX ASSEMBLY :: PROMOTER STRUCTURE

PREVIOUS :: NEXT