Introduction

Eukaryotic gene expression relies on RNA Polymerase II to transcribe the genes that will later become proteins. A 10-subunit core polymerase will binds to transcription factors, initiating transcription. The now 12-subunit polymerase will methodically run along the DNA, transcribing it into mRNA until it reaches the termination signal and falls off. The concept is simple, but thanks to Dr. Roger Kornberg and his colleagues, we now know molecularly how the RNA Polymerase II performs, and the process is far from simple.

Initiation

In order for the core polymerase to initiate transcription, the Rpb4/Rpb7 heterodimer must first bind to the core polymerase. This occurs by binding to the TFIIB subunit of the core polymerase. Rpb7 will specifically bind to both Rpb1 and Rpb6 by hydrophobic interactions with Ala105 and Pro106 of Rpb6 (Rpb7 Binding Site). The combination of these bindings has shown to increase the stability of the transcription initiation complex. It has also been shown that the binding of the heterodimer causes the clamp of the RNA Polymerase II to conform into the closed state. Further analysis of the binding found that the Rpb4/Rpb7 is able to bind single-stranded DNA and RNA, confirmed by the oligonucleotide-binding fold in the Rpb7 subunit and a loop containing residues 62 and 64 which penetrate the groove formed by the polymerase clamp. This sets the transcription complex up to be able to properly bind single-stranded DNA and RNA located in the groove (Bushnell and Kornberg, 2003)

With the Rpb4/Rpb7 heterodimer bound to the core polymerase, it now has a greater affinity for binding to the TATA box, which is the start region .35bp upstream from the start site in eukaryotes. The actual binding of the transcription initiation complex is further mediated by a Mediator complex. This complex will bind to the CTD of the transcription initiation complex. The new Mediator-bound complex will then be guided to the appropriate TATA box for transcription initiation. The groove formed by the closed clamp will open up and allow DNA to enter (Bushnell and Kornberg, 2003).

The groove, or cleft, in the RNA Polymerase II is now positively charged from the binding of Rpb4/Rpb7 heterodimer, allowing DNA to enter the complex and make a right angle bend at the active center. This active center is also where the DNA strands are separated and the DNA-RNA hybrid will emerge. The binding of the DNA into the positively charged cleft requires several zinc ions. Three zinc ions are located in the clamp itself, aiding in the conformational change of the clamp to clamp around the DNA. The clamp rotates about 30°, with switches 1, 2, 4, and 5 forming the base of the clamp and Rpb2 stabilizing the clamp. Final ordering is induced by the binding to DNA downstream and within the DNA-RNA hybrid, which are all stabilized by a network of salt linkages, forming a bridge across the cleft. This bridge is partially formed by Rpb1 residues Arg839, Arg840, and Lys843 (Salt Bridge) (Gnatt et al. 2001).

The DNA will now start moving into the lobe region between Rpb2 and Rpb1. This positions the DNA so the transcription bubble is able to form. In moving through the lobe, the DNA template strand follows along the bottom of the clamp and over the salt bridge. At this point, the template nucleotides at positions +4, +3, and +2 are stacked as a right-handed DNA and the +1 nucleotide is flipped 90° of the +2 nucleotide. This is obtained by bridge residues Ala832 and Thr831 through van der Waals interactions and the binding of Tyr836 to the +2 nucleotide. The flipping of the nucleotides are the beginning formation of the transcription bubble, and it has been suggested that Rpb2 fork loop 1 may help maintain the bubble further upstream (Gnatt et al. 2001). Before elongation can officially begin, the carboxyl-terminal domain of the RNA Polymerase II must be phosphorylated in order to be fully active (Kerppola and Kane, 1991).

Elongation

At this point, the first ribonucleotide is able to enter the complex and hybridize with the newly opened DNA strand. The template DNA strand is protein-bound over the entire hybrid length, where the incoming RNA will have protein contacts only in the downstream region. The contacts between the downstream and upstream hybrid positions are made by Rpb1 and Rpb2 respectively, with the Rpb1 1, 2, and 3 switches binding the nucleic acids, thereby increasing the stability of the hybrid (Gnatt et al. 2001).

The ribonucleotides will enter the transcription elongation complex in what is called the A site. The ribonucleotide is stabilized by a manganese ion that it carries in with it, and a second manganese ion located in the complex itself. The two Mn ions are coordinated by Rpb1 residues Asp481, Asp483, and Asp485 for the first Mn ion, and the second Mn ion is stabilized by the same residues plus Rpb2 residue Asp836. This two-metal ion mechanism is what catalyzes the elongation of the RNA. Magnesium ions have also been able to function equally as well for the metal ion catalysis mechanism (Westover et al. 2004).

The incorrect binding, or mismatch binding of a ribonucleotide causes the nucleotide to be flipped and positioned in the E site of the elongation complex. This ribonucleotide will be positioned with the sugar and base projecting down into the pore, which inhibits the binding of it to the DNA template strand. The phosphates and sugar will interact with Rpb2 residue Lys752 and Rpb2 residues Arg766, Tyr769, Lys987, Ser1019, and Arg1020. The base is subsequently projected into the solution with no protein contact. The rotated mismatched nucleotide in the E site will then stimulate correct binding and entry of the next ribonucleotide to the A site (Westover et al. 2004).

The correct ribonucleotide binding will be stabilized by the Rpb1 residues Tyr1349, Tyr1353, Trp954, and Phe947 and Rpb5 residues Arg200, Lys201, Ser202, and Glu203 which form the binding pocket, earlier named site A. The aromatic residues are thought to interact with the RNA bases and properly position them, with the aid of the metal ion, for hybridization. This pocket is also thought to prevent backtracked state stabilization, forcing the elongation complex to remain transcriptionally active (Westover et al 2004). This site extends from the floor of the cleft through the back of the enzyme, providing additional support for backtracking prevention. The Rpb1 residue Asn479 is responsible for ensuring a ribonucleic acid entering this binding pocket rather than a diribonucleic acid. In the advent of a diribonucleic acid being bound in the site, the complex will become destabilized, hindering the elongation process. In the end, van der Waals interactions from Rpb1 residues Thr831 and Ala832 are key for interacting with the nucleotide base at the end of the hybrid region (Gnatt et al. 2001).

It has been suggested that the bridge previously mentioned oscillates between a straight and bent conformation, which is thought to aid in the translocation of the elongation complex along the DNA strand. This is supported because the ribonucleotides must be held at positions +1 and -1 for phosphodiester bond synthesis. With the +1 nucleotide being flipped 90°, this allows for proper complementation binding between the DNA strand and the growing RNA strand. After the bond formation, the dinucleotide product must be held until -1 and -2 positions by protein-RNA contacts because base-pairing along is insufficient to retain inside the complex properly (Gnatt et al. 2001).

The growing RNA strand will exit through the exit channel in the elongation complex. This protects it from nucleases in the area. This channel has three components to it: the rudder, lid, and zipper, which play a role in DNA-RNA hybrid dissociation, RNA exit, and the maintenance of the upstream end of the transcription bubble. The RNA is first guided into the rudder, which leads it beneath the lid where the separation of the RNA and DNA is performed and maintained. The zipper is also thought to play a role in the separation of the DNA-RNA hybrid (Gnatt et al. 2001).

Termination

Transcription termination in eukaryotes requires a number of factors. The most common and effective termination signal is a series of sequences on the DNA template strand that are around 500-1,000 nucleotides downstream of the poly(A) site. This region has been termed the termination region or gF in mice. There are four termination elements within this gF sequence, the smallest functional one being only 69bp long, 70% purine rich, and able to form a very weak stem-loop configuration commonly seen in prokaryotic transcription termination. However, not a single termination element is able to effectively terminate transcription. The largest percent efficiency was around 37% effective in transcription termination (Tantravahi et al 1993).

In order for transcription termination, the elongation complex will read through the poly(A) site and the first termination element that is around 500bp downstream. The first termination element destabilizes the elongation complex, allowing the complex to loosen the clamp around the DNA strand. As the elongation complex hits the second, third, and fourth termination elements, the elongation complex is now no longer able to bind the DNA or keep the transcription bubble open and stable. The RNA Polymerase II then closes the transcription bubble, opens the clamp, and dissociates from the DNA strand (Tantravahi et al. 1993).