Poly(A) Polymerase
Created by Heidi Fredstrom
Poly(A) polymerase, as its name suggests, adds a tail of adenosines onto the 3' end of RNA1. The process is unique because it is a template-independent addition of RNA subunits to an RNA chain. With this function, poly(A) polymerase is described as being a member of the nucleotidyltransferase super-family, which includes CCA-adding enzyme, terminal deoxynucleotidyltransferase, and DNA polymerase β.1 Poly(A) polymerase is often abbreviated as PAP. This protein, as expressed by Escherichia coli is identified as 3AQL in the Protein Data Bank (PDB). It has been noted to being made of 454 amino acids with an average molecular weight of 52,548 D and a calculated theoretical isoelectric pH of 9.572.
PAP uses ATP as an energy source and targets almost any stretch of RNA, fragment or complete strand.3 This activity of PAP functions as a source of RNA destabilization as the strands it labels with the poly(A) tail are then marked for degradation. This addition is in fact reversible and other proteins, such as RNase II function to remove these tails while PAP works to put them back on. This pathway ensures that the rate of degradation of RNA occurs at an appropriate pace. Poly(A) polymerase also targets fragmented and “misfolded” RNA and therefore plays a role in RNA quality control.3 Interestingly, eukaryotic and prokaryotic PAP catalyze the same reaction, but they lack homology in amino acid structure as well as overall function. Eukaryotic poly(A) polymerase actually stabilizes mRNA and is an important part of the transcription termination complex.3
The nucleotidyltransferase family includes CCA-adding protein which is another enzyme that catalyzes a post-transcriptional modification of RNA by adding -CCA onto the 3' end of immature tRNA. These proteins exhibits several homologous traits between poly(A) polymerase both in function, amino acid sequence, and structure. One such protein is the Bacillus stearothermophilus CCA-adding enzyme, a 404 amino acid-long structure, which is identified by PDB as 1MIY. E. coli expressed PAP and this protein share 28% homology in their primary sequence with 91.7% of that homology coming in the first approximately 225 residues of the N-terminus of each.4 Such a high number of conserved sequence suggests that the related domain function of both- that is, the interaction with nucleotides, is found near the head of the protein. Likewise, both proteins are described as having a “head, neck, body, and tail.” 5, 2 They also share similarities in that both use ATP, although B. stearothermophilus has the option of also binding with CTP.5 Differences in amino acid almost definitely have arisen in this disparity as PAP can only utilize ATP. 2 The body and tail regions of these enzymes are described as being the RNA-interacting domains on both proteins. The disparity in amino acid structure then is possibly a result in the difference in targets. PAP targets any RNA fragment or strand and CCA-adding enzyme interacts solely with tRNA.5 Since there is a distinct cut-off in where the homology between these two proteins ends, it is very clear where the conserved domains are located and how their function is both similar and different.
In determining the structure of E. coli poly(A) polymerase, the researchers crystallized a couple of different forms. One of these was the Arg234His mutation which was used because the result led to an increase in the expression of the protein. Another form was the wildtype Arg234.1 Using these results, the researchers were able to conclude with a crystallized structure of PAP. Unfortunately, there were certain sections of the protein which were disordered. There was a section in the head domain between amino acids 115 and 137 which were disordered and also a group of 23 residues in the C-terminal region.1 The inconclusive results of these findings led to the protein being listed as being 415 residues long while the isoelectric pH of 9.57 was calculated using the 454 residues.2 One thing noticed by researchers is that this protein adopts more of a “sea otter” configuration1 than the “sea horse” structures found in the class II eubacterial and eukaryotic CCA-adding enzymes.7
Poly(A) polymerase is comprised of four different domains: head, neck, body, and leg. Functionally, the head and neck are the catalytic and nucleotide-binding domains while the body and leg are the RNA-binding domains.1
The head domain of PAP is defined as being between amino acids 17 and 185. It has nine anti-parallel β sheets which are supported by five α helices.1 Found within the beta sheets is a conserved set of three catalytic carboxylates: Asp69, Asp 71, and Glu108.6 The catalytic pocket created by these three residues is the binding site for one magnesium ion which possibly functions in a general base reaction with the 3'-OH of an RNA primer.1 One important secondary structure feature is the presence of a β turn in the catalytic domain which is used in nucleotide addition.1 There are several interactions found between the head and neck domain in PAP. For example, Asp151 in the head and Arg200 in the neck are found in their extended conformations, creating an intra-molecular hydrogen bond network.
The neck domain of PAP is found between residues 186 and 263 and is comprised of five α helices. It features two conserved amino acids: Asp194 and Arg197.1 These two residues are very important in the binding of ATP. Asp194 forms a hydrogen bond with Arg200, rotating it by approximately 30 degrees so that ATP may fit in the resulting pocket. This network is further stabilized by the presence of an RNA primer through the interaction of the 3' terminal nucleoside of RNA with the adenine of ATP.1 Two other very important residues in this domain of PAP are Arg203 and Arg234. They recognize the phosphate backbone of the RNA primer and aid in moving the 3' end of the primer to the catalytic site in order for RNA polymerization to occur.
The body region of PAP is located between amino acids 264–374 and 383–417. It is made up of a group of α helices. The C terminus of α12 loops up and interacts with the loop between α9 and α10 in the neck domain and this resulting structure is called the “hand region.”1 The body domain does not have any conserved or key amino acids, but is involved in RNA binding.1
Finally, the leg domain of PAP is defined as being between residues 375–382 and 418–427 and is also comprised of a group of α helices.1 Interestingly, this region is largely disordered and remains unresolved.1 However, it is an area of concentrated positive charges which is believed to take on a random coil structure.7 This group of charges is very likely involved in RNA recognition.1
Poly(A) polymerase exhibits most of its observed function in the head and neck regions and this is demonstrated by this N-terminal region of the protein being highly conserved among other enzymes in the nucleotidyltransferase family. The primary structure of PAP was crucial in the formation of important anti-parallel β sheets and α helices that are instrumental in creating the overall shape that allows for the substrates to enter, interact with key residues, and therefore catalyze the reaction of RNA polymerization. By knowing the structure of this protein, researchers can observe the interactions between the different domains of the protein which allow for it to function.