Tetrahydrofolate_Riboswitch

The tetrahydrofolate (THF) riboswitch (PDB ID: 3SD1) is a synthetically constructed ribonucleic acid molecule from Streptococcus mutans (Smu) UA159. Riboswitches are structured mRNA segments, found principally in bacteria, which often regulate important metabolic pathways (1). Riboswitches bind to specific metabolites, which affects an expression platform that regulates at either the transcriptional or translational level (2).

The crystallized aptamer domain of the Smu THF riboswitch controls the FolT gene, which regulates folate transport. The aptamer domain has two sites that can bind to the reduced forms of folate, THF and DHF, as well as their derivatives. This binding of ligands results in transcriptional termination, so folate transport stops. The aptamer domain of the Smu THF riboswitch was crystallized using the hanging drop vapor diffusion method. It was crystallized with folinic acid, a THF derivative, but in the crystallized structure folinic acid appears as the derivative FOZ (1). Two FOZ ligands are bound to the crystallized THF riboswitch.

The THF riboswitch is present in many Firmicutes bacteria in addition to Smu, and controls the synthesis and transport of folate. Folate is an essential nutrient that participates in one carbon metabolism. Folic acid deficiency can result in health issues, including cardiovascular disease and birth defects (3). Effective regulation of folate synthesis could be used to decrease folic acid deficiency via biofortification. In this method, bacteria could be engineered to increase the production of folate by altering the THF riboswitch (1). The THF riboswitch might also be a useful antimicrobial drug target, since folate analogs could cause bacteria to misregulate the production of folate (4).

The Smu THF riboswitch has a molecular weight of 28,832.8 Da, calculated by the Chang Bioscience Calculator from the entered nucleotide sequence (5). The riboswitch contains a single subunit, which includes the single aptamer domain with two ligand binding sites, of 89 residues. The overall structure consists of two sets of coaxially stacked helices that are positioned next to each other via the formation of a pseudoknot. The side-by-side arrangement of the helices is also enforced by a three-way junction, which is an element of nucleic acid secondary structure (1).

The pseudoknot is formed by a series of five Watson-Crick pairs, in which residues 37-41 form hydrogen bonds with residues 79-83. The psuedoknot structure is characterized as a helix-loop type pseudoknot, since the two principle helices are coaxially stacked. The regulatory helix P1 is formed by a series of five Watson-Crick pairs, in which residues 1-5 form hydrogen bonds with residues 85-89. The regulatory helix is stabilized by ligand binding, which results in structural changes for the downstream expression platform. Unlike most riboswitches, the regulatory helix is adjacent to the pseudoknot rather than the three-way junction (2).

One of the ligand binding sites of the Smu THF riboswitch is located in a minor groove, adjacent to the pseudoknot. The other site is located in a minor groove, adjacent to the three-way junction. Both of the sites bind to ligand mainly through interacting with the reduced pterin group present in THF and its derivatives (1).

In the binding site adjacent to the pseudoknot, U-7, U-35, and U-42 participate in the recognition of folinic acid via hydrogen bonding. A-8 stacks with the first ring of the pterin group of folinic acid. A-80 interacts with the para-aminobenzoate group via stacking interactions, which increases the binding affinity. In the binding site adjacent to the three-way junction, U-25 and C-53 participate in the recognition of folinic acid via hydrogen bonding. U-25 forms three hydrogen bonds with the pterin group of folinic acid, and C-53 forms a reverse Watson-Crick type pair with the pterin group. G-26 stacks with the first ring of the pterin group. G-68 interacts with the para-aminobenzoate group of folinic acid via stacking interactions, which increases the binding affinity (1).

The THF riboswitch specifically binds to the reduced forms of folate and derivatives, but does not bind to oxidized folate. This is partly because folate cannot form a hydrogen bond to U-7 or U-25 with the nitrogen that loses a hydrogen atom. In addition, the para-aminobenzoate group of folate can no longer stack upon G-68 in the junction site or A-80 in the pseudoknot. This discrimination between folate and folinic acid suggests that the THF riboswitch only monitors the active portion of the folate pool in cells (1). The binding of two folinic acid ligands in the THF riboswitch is shown to be cooperative under physiological conditions. However, the binding of a single ligand can still result in a weakened regulatory response. Although both sites bind folinic acid with about the same affinity, the pseudoknot site is much more effective in regulating gene expression than the three-way junction site (1).

It is suggested that cooperative binding promotes fast and efficient folding of the aptamer, which is important for co-transcription. In this model, binding of the ligand to the three-way junction site causes structural changes that stabilize the three-way junction and form the pseudoknot with an adjacent binding site. This is different from most riboswitches, in which only one ligand is necessary for the effective folding of RNA (1).

The THF riboswitch has also been crystallized from a synthetically constructed RNA sequence of Eubacterium siraeum (PDB ID: 3SUX), with P1 destabilized (2). The crystallized riboswitch only contains one ligand in the aptamer domain, located adjacent to the three-way junction, because of a domain swap in the crystal lattice. This structure prevents productive ligand binding at the pseudoknot, and so reflects the state of the aptamer domain when the antiterminator has formed and the regulatory helix P1 is destabilized (6).

N-blast is a program used to find nucleic acids with primary structures that are similar to a nucleic acid query. An E-value is assigned to these subject nucleic acids, which is calculated based on gaps of nucleotides that the query does not share with subjects. An E-value of less than 0.05 indicates significant similarity in primary structure. The THF riboswitch n-blast results only include sections of genomic DNA from a variety of Streptococcus bacteria species. For example, the Streptococcus uberis 0140J genome (accession number: AM946015.1) has an E-value of 4 x 10-20 and the Streptococcus suis D12 genome (accession number: CP002644.1) has an E-value of 5 x 10-19 (7). The n-blast results are not used as comparison structures, however, since none of the sequences are in the Protein Data Bank and there is no structural information available for any RNA segments.

Another program used for bioinformatics searches is the Dali Server, but this program only works for proteins. The purpose of the Dali Server is to find proteins with tertiary structures that are similar to a query protein. The Dali Server produces a measure of similarity, the Z-score, between protein tertiary structures by comparing intramolecular distances with a sum-of-pairs method. A Z-score greater than two indicates tertiary structure similarity. The locations of the atoms of a protein structure are used as the input for the comparison. In order for the program to function, the backbone atoms have to be N, Cα, C, and O, which are the atoms of a protein backbone and not a nucleic acid (8). Instead of using the Dali Server or n-blast to find comparison nucleic acids, comparison structures had to be found manually, since the programs do not yield enough information.

The B. subtilis xpt guanine riboswitch (PDB ID: 3FO6) is one nucleic acid that has several notable similarities to the Smu THF riboswitch. Both of these riboswitches terminate transcription in the presence of specific ligands. The binding of guanine or hypoxanthine to the guanine riboswitch leads to the formation of a downstream terminator that stops transcription for purine metabolism and synthesis genes (9). Guanine recognition in the guanine riboswitch is achieved by a base triple formed between the guanine ligand and two pyrimidine residues, U-51 and C-74, located in the binding site. This base triple is similar to the base triples formed in both binding sites of the THF riboswitch, which also involve pyrimidine residues (6).

Three base paired helices form the guanine riboswitch, while four helices form the THF riboswitch. The aptamer domain of the guanine riboswitch is also shorter, with 68 residues instead of 89. In both the THF and guanine riboswitches, the overall structure consists of a three-way junction stabilized by a distal tertiary interaction. The guanine tertiary interaction is a loop-loop structure of two terminal hairpin loops (9). A notable difference is that the regulatory helix P1 in the guanine riboswitch is adjacent to the three-way junction rather than the tertiary interaction, as in the THF riboswitch. Also, only one ligand binding site is necessary to produce an effective regulatory response, unlike the THF riboswitch. This regulation of bacterial gene expression has been studied in a variety of riboswitch classes in addition to the THF and guanine riboswitches. The study of riboswitches is also of biological importance because these structures appear to have great potential for applications in bioengineering and antimicrobial therapeutics (6).