The yeast Pol II structure provides insights into transcriptional mechanisms
RNA polymerase II has been the most thoroughly studied because it transcribes protein-coding genes to produce mRNA molecules; coding for proteins was long thought to be the major function of the genome. We now know, however, that there are many long noncoding RNA molecules produced in the nucleus, and Pol II appears to be responsible for these as well. Thus, Pol II is a very important enzyme.
Yeast Pol II is the only eukaryotic RNA polymerase whose structure has been resolved by crystallography. Lessons about how the enzyme functions can be learned from this structure, and these provide a general model, as the basic mechanism of Pol II functioning seems to be shared by the other eukaryotic polymerases. The structure shown in Figure 1A is that of the core enzyme lacking subunits Rpb4 and Rpb7. This structure reveals that the two largest subunits Rpb1 and Rpb2, numbered 1 and 2 in Figure 1B, form opposite sides of a positively charged active center cleft, with the smaller subunits arranged around the periphery of the complex. At the bottom of the cleft lies a magnesium ion known as metal A that participates in the catalytic addition of each ribonucleoside monophosphate residue. Figure 1C shows the structure of the core polymerase during elongation. The mobile clamp, a domain of Rpb1, in the free enzyme has an open conformation that allows DNA binding during initiation but closes down to almost fully embrace the DNA and the RNA transcript, thus ensuring processivity of transcription. The structure also reveals the position of the bridge helix that connects the two large subunits; this helix is implicated in translocation motions of the enzyme with respect to the DNA. The con formation of the DNA–RNA hybrid at the catalytic center is intermediate between the A- and B-forms of DNA, as seen in Figure 1D.
Fig1. Yeast RNA Pol II structure and the enzyme in the act of transcription. (A) Structure of enzyme at 2.8 Å resolution; note that subunits 4 and 7 are missing from the crystal. The protein is shown in a ribbon representation, with the Mg2+ ion at the active center as a pink sphere. Light blue spheres are Zn2+ ions used in crystallization. The structure contains 3500 amino acid residues, with more than 28,000 non-hydrogen atoms. (B) This schematic provides the color code for individual subunits. The thickness of connecting bars between them corresponds to the surface area buried in the corresponding subunit interface: the larger the buried surface, the stronger the interaction. (C) Polypeptide chains are shown in white, with the mobile clamp, a domain in Rpb1, shown in orange. The clamp closes on DNA and RNA in the elongation complex but has a different orientation in the free enzyme to allow DNA binding at initiation. The bridge helix, shown in green, connects the two largest subunits and undergoes conformational transitions during the translocation steps. The DNA template strand is in blue, the nontemplate strand is in green, and the nascent RNA transcript is in red. DNA is surrounded by protein over ~270°. This tight grip of the polymerase on the DNA ensures high processivity of the enzyme. (D) This schematic represents the nucleic acids in the transcribing complex: the DNA template strand is in blue, the nontemplate strand is in green, and the nascent RNA chain is in red. Solid ribbons are from the crystal structure; dashed lines indicate possible paths of the nucleic acids not revealed in the structure. Note that the entering and exiting DNA duplexes are at an angle of nearly 90° with respect to each other. The conformation of the DNA–RNA hybrid is intermediate between canonical A- and B-forms of DNA. The schematic also reveals where the nascent transcript leaves the polymerase and where it ends up if backtracking occurs. [A, B, adapted from Cramer P, Bushnell DA & Kornberg RD (2001) Science 292: 1863–1876. With permission from American Association for the Advancement of Science. C, D, adapted from Gnatt AL, Cramer P, Fu J et al. (2001) Science 292: 1876–1882. With permission from American Association for the Advancement of Science.]
The largest subunit of Pol II, Rpb1, contains a long C-terminal domain or CTD that is not present in the homologous subunits of Pol I and Pol III. CTD is characterized by the presence of numerous repeats—26 in yeast, 52 in humans—of a heptapeptide sequence, YSPTSPS. The CTD does not have a regular structure and so is not seen in crystallographic studies. The last ordered residue of Rpb1 seen in the crystal structure lies at the beginning of a linker that connects the structured body of Rpb1 to the unstructured CTD (Figure 2A). The unstructured CTD is probably not extended but compacts into a disordered state, close to the body of the enzyme. Individual amino acid residues within the repeat can undergo phosphorylation, glycosylation, and proline cis–trans isomerization (Figure 2B). The phosphorylation of serine residues and possibly also other modifications change during the transcription process (Figure 3). It is believed that these modifications, separately or in combination, determine the binding of various protein factors to the CTD, in turn regulating transcription elongation. This ability to interact with numerous partners is typical of disordered protein domains. The CTD is also believed to play a role in RNA processing.
Fig1. C-terminal repeat domain of Rpb1, the largest subunit of Pol II. Although this subunit is highly homologous in all three polymerases, only Rpb1 of Pol II contains a C-terminal repeat domain or CTD. (A) Size and location of CTD. Dashed lines represent the possible length of a fully extended linker and CTD. The absence of electron density corresponding to the linker and CTD provides evidence of disorder or motion, but even if disordered, these regions are unlikely to be in an extended conformation. The existence of free space in the crystal lattice for CTDs from four neighboring polymerases of Pol II suggests that the CTD may actually be compacted. (B) CTD contains the repeated heptapeptide sequence YSPTSPS, which can be phosphorylated, shown here as P, on all three serines, individually or in combination. These same residues can alternatively be glycosylated, shown here as G, with multiple glycosylated forms possible; Thr4 can also be glycosylated. Finally, both prolines can undergo a cis-trans isomerization reaction as illustrated. R1 and R2 represent continuations of the polypeptide chain toward the N- and C-terminus, respectively. Subsets of these modifications influence the function of Pol II by serving as binding platforms for specific proteins. The unit of recognition is believed to be a pair of heptapeptides. The number of repeats differs in different organisms, ranging from 26 in yeast to 52 in humans. [A, adapted from Cramer P, Bushnell DA & Kornberg RD (2001) Science 292: 1863–1876. With permission from American Association for the Advancement of Science.]
Fig3. C-terminal repeat domain phosphorylation pattern. The phosphorylation pattern of CTD changes during the transcription process and determines which protein factors bind to it. The serines in positions 2 and 5, as well as 7, of the YSPTSPS repeat in CTD can undergo phosphorylation in a pattern that reflects the stage of transcription. In the pre-initiation complex, CTD is not modified. The transition to elongation is accompanied by phosphorylation on Ser5 residues, represented as blue spheres. Later in the elongation process the CTD becomes doubly phosphorylated, carrying modifications on both Ser5 and Ser2. Ser2 phosphorylation is depicted as red spheres. Finally, as the enzyme approaches the end of the gene, phosphorylation is exclusively on Ser2 residues. After termination of transcription, Pol II falls off the template and undergoes a dephosphorylation reaction, which allows it to re-initiate. The enzymes involved in these phosphorylation/dephosphorylation events are indicated above and below the arrows connecting the different stage polymerases. Exactly what controls these transitions, that is, the activities of the respective kinases, is still unclear. It is clear that the phosphorylation pattern of CTD determines what protein partners bind to the polymerase and control its activity. There are more than 100 different proteins that interact with CTD; some of them are depicted above the respective complex and are discussed throughout the book. [Adapted from Phatnani HP & Greenleaf AL (2006) Genes Dev 20: 2922–2936, and Egloff S & Murphy S (2008) Trends Genet 24: 280–288. With permission from Cold Spring Harbor Laboratory Press and Elsevier.]
The structure of Pol II is more evolutionarily conserved than its sequence
Pol II structure is highly conserved across eukaryotic species: yeast and human Pol II sequences exhibit 53% overall identity, with the conserved residues distributed over the entire structure. As expected, sequence conservation between yeast and bacterial polymerases is less extensive, with the conserved regions clustering around the active center (Figure 4A). Despite the relatively weak sequence conservation, the conservation of structure is remarkable (Figure 4B). The eukaryotic enzymes share a core structure, and thus a conserved catalytic mechanism, with the bacterial enzyme. The differences in structure are at the enzymes’ periphery and surface, where interactions with other proteins occur. The detailed structure of the Pol II enzyme complex is especially revealing with respect to its mechanism of action during RNA elongation.
Fig4. Comparison between the two largest subunits of bacterial and eukaryotic RNA polymerases. (A) Bacterial RNA polymerase from Thermus aquaticus; (B) eukaryotic RNA polymerase from Saccharomyces cerevisiae. The two structures are overlaid: blocks of sequence homology are shown in red, and regions of structural homology are shown in green. The magenta sphere is the Mg2+ ion in the active site. The structures show consider able conservation around the active center, suggesting a conserved catalytic mechanism. Note that structural homology is much higher than sequence homology. The differences are mainly located in the peripheral and surface features, where the polymerases interact with other proteins. [From Cramer P, Bushnell DA & Kornberg RD (2001) Science 292: 1863–1876. With permission from American Association for the Advancement of Science.]
Nucleotide addition during transcription elongation is cyclic
Transcription elongation passes through repetitive cycles of adding one new nucleotide to the growing RNA chain, followed by translocation steps that move the enzyme complex one base pair along the DNA and prepare the complex for the next nucleotide addition. Recent advances in deciphering the structure of triple Pol II-DNA RNA complexes at different steps of the elongation process have made it possible to understand this nucleotide addition and translocation cycle. The entire cycle is illustrated in Figure 5. Most of the steps are highly evolutionarily conserved and seem to occur in a similar way in all three eukaryotic polymerases as well as in bacteria and archaea.
Fig5. Model of the nucleotide addition cycle. (A) Schematic represen tation of RNA polymerase elongation complex showing the DNA template strand in blue, the nascent RNA chain in red, and two functional polymerase elements: bridge helix in green, Rpb1 residues 811–845, and trigger loop in brown, Rpb1 residues 1062–1106. The two catalytic magnesium ions, desig nated A and B, are shown in pink. The vertical dotted line indicates the position of DNA template base +1. The cycle is based on a Brownian ratchet model, which assumes that the ground state of the elongation complex is an equilib rium between interconverting pre- and post-translocation states, denoted by double arrows. This oscillation is tem porarily stopped by substrate binding and resumes around the next template position after nucleotide addition. The cycle begins with the binding of an NTP substrate, shown in orange, to an open active center or open trigger loop. Folding of the trigger loop leads to closure of the active center. The growing 3′-end of the RNA chain becomes elongated by catalytic addition of the nucleotide, which results in formation of a pyrophosphate ion. The release of pyrophosphate destabilizes the closed conformation, leading to trigger loop opening. In the resulting pretranslocation state, the incorporated 3′-terminal nucleotide remains in the substrate site. Translocation possibly takes place in two steps. During step 1, the hybrid moves to the post-translocation position, the downstream DNA translocates until the next DNA template base, +2, reaches the pretemplating position above the bridge helix. During step 2, the DNA template base twists by 90° to reach its templating position in the active center, +1. Note the altered, flipped-out conformation of the bridge helix and the transition from an open to a wedged trigger loop conformation. After trans location of DNA and RNA, the elongation complex is in a post-translocation state, with a free substrate site to bind the next incoming NTP. The nucleotide addition cycle can then be repeated. (B) Structural models of various stages of the nucleotide addition cycle. Six of the seven different functional states of the elongation complex have been crystal lographically resolved, as marked by a black star; two other states were modeled. [Adapted from Brueckner F, Ortiz J & Cramer P (2009) Curr Opin Struct Biol 19: 294–299. With permission from Elsevier.]
Loading of the nucleoside triphosphate substrate occurs in two steps: pre-insertion and insertion. Recall from Chapter 9 that the DNA within the polymerase will be opened or denatured into a transcription bubble. The nucleoside triphosphate (NTP) first binds to an open active center in the pre-insertion stage, when the selection of the appropriate NTP—that is, the one which base-pairs with the template base to be copied—takes place. Additional contacts help to discriminate between NTPs and deoxy-nucleoside triphosphates (dNTPs). In the second step, the correct NTP is delivered to the insertion site; this step is followed by the catalytic incorporation of the nucleoside monophosphate or NMP into the nascent RNA chain. The release of pyrophosphate completes this portion of the cycle.
Next, a complex series of motions in the polymerase allows the necessary translocation of the DNA with respect to the catalytic center to occur. Translocation also occurs in two step. In step 1, the hybrid moves so that the downstream DNA translocates until the next DNA template base is properly located. In step 2, the DNA template base rotates by 90° to attain its correct orientation in the active center. Both of these steps are supported by conformational transitions in specific portions of Pol II. Once translocation is complete, the complex is ready for a new nucleotide addition cycle.
This one-directional translocation of the DNA through the enzyme requires an energy source. The only evident source is cleavage of the NTPs to add an NMP to the growing RNA chain and release pyrophosphate. This is an energetically highly favored reaction: the action of the polymerase is only catalytic. Thus, once NTPs are available, the process drives itself. The irreversibility of the process is further guaran teed by enzyme-catalyzed hydrolysis of the released pyrophosphate.
Fidelity of transcription, that is, avoiding incorporation of the wrong nucleotide or even a deoxyribonucleotide into the growing chain, is assured by the two-step nature of the elongation process: the incoming NTP is tested before the second step can occur. This testing is favored by the slow rate of transcriptional elongation in eukaryotes, which is about a tenth of the rate in bacteria.
Transcription initiation depends on multisubunit protein complexes that assemble at core promoters
The Pol II core promoter is defined as the DNA sequence(s) around the transcription start site, TSS, that binds the polymerase and directs the initiation of transcription. For many years, it was believed that a single motif upstream of TSS served as a universal promoter. This is known as the TATA box, similar to the −10 element in bacterial promoters. Now we know that promoter recognition in eukaryotes is much more varied and more complex.
There are two broad types of core promoters: focused or sharp-type promoters and dispersed or broad-type promoters (Figure 6). In vertebrates, less than a third of all promoters are focused. The motifs constituting focused promoters are relatively well-known (Figure 7), but less is known about the characteristic sequence motifs and factors involved in PIC formation at dispersed promoters.
Fig6. Focused versus dispersed core promoters. Focused or sharp type promoters contain single or tightly clustered TSSs. They are more ancient and widespread in nature. Dispersed or broad-type promoters, on the other hand, contain multiple weak TSSs over a region of 50–100 nucleotides. They are usually found in CpG islands (see Chapter 12) and are more common than focused promoters in vertebrates. [Adapted from Juven-Gershon T, Hsu JY, Theisen JW et al. (2008) Curr Opin Cell Biol 20: 253–259. With permission from Elsevier.]
Fig7. Known core promoter motifs for Pol II transcription. (A) Schematic representation of some individual promoters and the protein complexes that bind to them. Each indi vidual element occurs in only a fraction of all promoters, with the Initiator element, Inr, being the most frequent. Double headed arrows in some of the promoters indicate that the spacing between the two elements is critical. Often, promoter elements controlling specific genes work in combination. Elements that may or may not be present in a specific promoter are circled; for example, the TATA box is a secondary promoter element that is present in some Inr promoters. The synergy between motif ten element, MTE, and other promoter elements led to the creation of the super core promoter, SCP, which contains optimized versions of TATA, Inr, MTE, and downstream promoter element DPE. These elements are marked by asterisks in part B. SCP is the strongest existing promoter that functions both in in vitro transcription reactions and when introduced into cells in culture. SCP shows an unusually high affinity for TFIID. The green ovals represent TFIID that recognizes at least five of the seven promoters. That TFIID is a complex of TATA-binding protein (TBP), core subunits, and some of the TBP-associated factors or TAFs; thus the TFIID complexes that recognize specific promoter elements may be different. (B) To-scale diagram showing the mutual disposition of various promoter elements. The lower schematic illustrates elements that overlap the elements presented in the upper schematic. There is no promoter element that is universal, that is, present in all promoters. [A, adapted from Sandelin A, Carninci P, Lenhard B et al. (2007) Nat Rev Genet 8: 424–436. B, adapted from Juven-Gershon T, Hsu JY, Theisen JW et al. (2008) Curr Opin Cell Biol 20: 253–259. With permission from Elsevier.]
Transcription initiation in eukaryotes is exceedingly complex and depends upon assembly on the promoter of a huge multiprotein complex. The nuclear polymerases themselves are not even capable of recognizing promoter sequences; thus, they need the help of basal transcription factors, also called general transcription fac tors (GTFs), to bind to promoters and initiate the process. Basal TFs are absolutely required for transcription, independent of the frequency at which a particular gene is transcribed. The level of transcription is determined mainly by the frequency of transcription initiation and the stability of the minimal preinitiation complex (PIC), which in turn depend on signals from regulatory proteins such as activators or repressors. These bind specific nucleotide sequences in regions distal to the basal promoter.
A schematic of the minimal PIC assembled on a TATA-box promoter is presented in Figure 8A, and more detailed information about each of the PIC factors is given in Table 1. The first complex to bind to the promoter in vitro is transcription fac tor IID (TFIID), which contains the TATA-binding protein (TBP), six core subunits, and a variable set of additional subunits, termed TATA-binding protein associated factors (TAFs) (Figure 8B). TAFs discriminate among the alternative promoters recently recognized (see Figure 7). TBP binding to the TATA box induces a kink in the DNA and opens the minor groove of the double helix, wherein most of its contacts occur (Figure 8C). Remember that most proteins recognize nucleotide sequences in duplex DNA through the major groove. If a protein interacts with DNA through the minor groove, it usually must deform the double helix to ensure access to the base edges for sequence-specific recognition.
Fig8. Assembly of minimal functional pre-initiation complex (PIC) on the TATA box during transcription initiation. (A) Schematic of the partial PIC. TBP and the other components of TFIID bind to the TATA box and recruit all other general transcription fac tors, GTFs, and the polymerase itself. Additional general transcription factors E, H, and J are not shown. The numbers show the succession of binding in vitro. (B) TFIID: the core complex of six subunits, 4–6 and 8–10, depicted in light green, associates with combinations of other TAFs and TBP to form a large variety of different complexes. These different complexes bind to different types of promoters to initiate transcription. (C) Structure of TBP, shown in blue, bound to a promoter TATA box, shown in red. TBP induces a kink in the DNA strands and forces open the minor groove of the DNA double helix, where most of its contacts with DNA occur. The other GTFs, as well as the RNA polymerase II complex, assemble around it. The TBP-DNA complex is slightly asymmetrical, ensuring the directionality of transcription, that is, that transcription occurs on the correct DNA strand.
Table1. Protein complexes involved in pre-initiation complex (PIC) assembly. TFIIA, previously thought of as a general transcription factor, is actually a coactivator that interacts with activators and components of the basal initiation machinery to enhance transcription. [Adapted from Sikorski TW & Buratowski S (2009) Curr Opin Cell Biol 21: 344–351.]
The absolute requirement for GTFs for transcription initiation motivated structural biologists to use a combination of sophisticated methods, mainly cryo-EM, mass spectrometry and computational analysis to resolve the structure of the two multi subunit factors TFIID and TFIIH.
The studies of TFIID revealed that its initial binding to the TATA box of the downstream promoter facilitated scanning of TBP for the TATA box and the subsequent engagement of the promoter. Interestingly, there are conformational changes occur ring in TFIID from its free unbound form to the so-called “engaged” form in the PIC; these involve the position of lobe A with respect to the rest of the molecule (see Figure 9 for the structure of the complex and its three lobes). Lobes B and C have a fairly rigid connection between them, but lobe A’s attachment to the BC core is flexible. It rotates away from lobe C toward lobe B, bringing to the formation, through several intermediate conformational steps, of the PIC, where the TBP forms a stable complex with bent DNA to open the minor groove. There are two highly instructional videos in Patel AB et al. [Science (2018) 362: 1376–1312], which will help to visualize the process in its entirety.
Fig9. High-resolution structures of basal transcription complex TFIID. (A) Transparent cryo-EM map in the free state, with fitted cryo-EM maps from focused refinements of the BC core in solid blue and lobe A in yellow. (B) TFIID structural model viewed from the same perspective.
The second multisubunit GTF TFIIH has also been structurally well characterized (Figure 10). The complex is also involved in nucleotide excision repair. The factor is composed of ten subunits, amounting to 500 kDa. Seven of the subunits, XPB (xeroderma pigmentosum type B), XPD (xeroderma pigmentosum type D), p62, p52, p44, p34, and p8, form a core complex competent for DNA repair; three more subunits, the kinase CDK, cyclin H, and Mat1, are required for transcription initiation. The importance of the complex is evident from the fact that mutations in XPB, XPD, and p8 lead to severe premature aging and cancers in the genetic dis eases xeroderma pigmentosum, Cockayne syndrome, and trichothiodystrophy. XPB and XPD are ATPase/helicases; the helicase activity of XPD is upregulated by inter actions with p44, which are also required for the recruitment of the helicase to the complex.
Fig10. High-resolution structures of basal transcription complex TFIIH. (A) Cryo-EM map of human TFIIH (including the TFIIH core complex and MAT1) colored according to its protein subunits. (B) Atomic model of human TFIIH (PDB ID 6NMI) colored according to its protein subunits (p62 shown as sur face, other subunits as spheres). [Figure provided by Basil Greber & Eva Nogales, Lawrence Berkeley National Laboratory, Berkeley, California, USA.]
Conformational changes in molecular complexes are often associated with functional transitions (see above the case of TFIID). In the free TFIIH complex, the inter action between XPB and XPD brings together the two ends of the core complex in a horseshoe shape. In the preinitiation complex containing Pol II, TFIIH undergoes a conformational change that breaks the interactions between XPB and XPD, increasing the distance between them. Other bound proteins are also relocated by this motion.
An additional protein complex is needed to connect Pol II to regulatory proteins
It was long believed that the six basal transcription factors TFIIB, D, E, F, H, and J were sufficient for transcription initiation. More recently, another large protein com plex, Mediator, was discovered in yeast and subsequently also observed in mammalian cells (Figure 11A). As stated by Roger Kornberg, “Mediator is no less essential for transcription than Pol II itself.” In addition to its role as a necessary basal factor, Mediator forms the basis for regulated transcription: it serves as a bridge between Pol II and the sequence-specific proteins that recognize regulatory elements sur rounding the gene. Contacts between Mediator and Pol II are extensive and direct and involve Pol II subunits Rpb4 and Rpb7 (Figure 11B and Figure 11C). The binding between the two protein complexes stabilizes the PIC for efficient initiation. Thus, according to Kornberg, “mediator is a co-activator, a co-repressor, and a general transcription factor, all in one.”
Fig11. Mediator, a multiprotein complex of >20 subunits. Mediator is a co activator, co-repressor, and general transcription factor all in one. (A) Schematic showing Mediator serving as a bridge between proteins that activate or repress transcription and the transcriptional machinery assembled at the promoter. Note the direct contacts with both the activator protein and Pol II. (B) Model of the Mediator-RNAP II complex, based on cryo-EM imaging, reveals contacts between the Head module of Mediator and the RNAP II subunits Rpb4/Rpb7. Interaction of RNAP II with Mediator is very extensive, with Mediator surrounding the face of the polymerase where TBP, TFIIB, and TFIIF presumably bind. All of these contacts are consistent with reported stabilization of the pre-initiation complex by Mediator. (C) Interaction between the Head module of Mediator and Rpb4/Rpb7 could affect the con formation of the polymerase clamp domain and possibly facilitate opening, shown by a white open arrow, of the RNA polymerase II active-site cleft, outlined in black, to allow access of double-stranded promoter DNA to the polymerase active site. [A, adapted from Kornberg RD (2007) Proc Natl Acad Sci USA 104: 12955–12961. With permission from National Academy of Sciences, USA. Copyright (2007) National Academy of Sciences, USA. B, from Cai G, Imasaki T, Takagi Y et al. (2009) Structure 17: 559–567. With permission from Elsevier. C, from Cai G, Imasaki T, Yamada K et al. (2010) Nat Struct Mol Biol 17: 273–279. With permission from Macmillan Publishers, Ltd.]
The current model of how all these factors work together to promote initiation is presented in Figure 12. Importantly, once the polymerase clears the promoter to enter the processive elongation phase of transcription, a scaffold complex comprising Mediator, most of the basal TFs, and an activator protein stays behind; this scaffold can now be quickly repopulated by a free Pol II/TFIIF complex to form the closed promoter complex. This speeds up the next round of transcription initiation.
Fig12. Structural model for Mediator initiation and re-initiation. The steps of the process are as follows: (Step 1) Mediator is recruited to a promoter through the action of activator A1. (Step 2) The action of a second activator, A2, leads to a conformational change of Mediator that now provides a binding site for components of the basal transcriptional machinery. (Step 3) Transcription factors TFIIB and TFIID position themselves at the promoter and form a landing platform for the Pol II/TFIIF complex. (Step 4) Pol II/ TFIIF complex binds, or Pol II and TFIIF concomitantly bind, to the pre-formed Mediator/DNA/TBP/TFIIB complex. (Step 5) Through the action of TFIIE and TFIIH, the promoter opens and the template DNA strand reaches the polymerase active site. (Step 6) Following the abortive transcription phase, Pol II, or the Pol II/TFIIF complex, escapes the promoter, leaving behind a platform that facilitates rapid assembly of a new pre-initiation complex and re-initiation by new incoming Pol II/TFIIF complex. [Adapted from Asturias FJ (2004) Curr Opin Struct Biol 14: 121–129. With permission from Elsevier.]
Termination of eukaryotic transcription is coupled to polyadenylation of the RNA transcript
Termination of eukaryotic transcription is strongly coupled to post-transcriptional processing of the RNA transcript (see Chapter 14). Each mRNA primary transcript, with a few exceptions, is polyadenylated at its 3′-end: a string of A residues is added by a special polymerase. The site of polyadenylation is defined by a particular sequence called the polyadenylation signal; the RNA transcript is cleaved at some point 3′ to the signal, and the transcribing polymerase is released. The CTD may be involved here. One function of polyadenylation seems to be to aid in exporting RNA from the nucleus to the cytoplasm. Thus, it is not surprising that polyadenylation is not found in bacteria.
