4 RNA structure and function
4.1 The varied structures of RNA
RNA is a versatile cellular molecule with the ability to adopt a number of complex structural conformations. Although RNA is often thought of as a single-stranded molecule it is actually highly structured.
Why does RNA adopt a higher-order structure?
Remembering that the bases in nucleic acids are hydrophobic, base interactions that minimise exposure of the bases to water will drive RNA secondary structure formation, just as they do with DNA.
In effect, this means folding to maximise base pairing, since hydrophobic interactions arise from stacking of bases aligned by complementary hydrogen bonding.
What type of helix is formed by duplex RNA?
A-form helices are formed by duplex RNAs (Section 3.1).
The level of structural complexity in RNAs is much greater than that in DNA and many RNAs have a defined tertiary structure analogous to that observed in proteins. As we discussed earlier, it has been more difficult to study RNA structures than DNA structures by X-ray diffraction. However, for many RNAs, secondary structures can be predicted from theoretical folding patterns based upon RNA base sequence and by assuming maximal base pairing. By comparing similar predicted structures from different organisms (which will have slightly different base sequences), the most likely structure can often be determined. An example of this approach is the original ‘evidence’ for the cloverleaf structure of tRNA shown in Figure 18, which was derived by predicting the secondary structure of many different tRNAs. Only later were these interactions and the basic structure of tRNA confirmed by X-ray diffraction.
Figure 18: (a) Cloverleaf diagram of the secondary structure of tRNA as predicted from the sequence, according to the rule of maximising base pairing. Note how base pairing produces four arms and a small extra arm. The dihydrouridine and TΨC arms are so named because of their characteristic modified bases. The anticodon arm contains bases 34 to 36, which are complementary to the mRNA codon, and the acceptor arm is the site of amino acid attachment, a reaction catalysed by a tRNA synthase specific for both the amino acid and the tRNA. (b) Tertiary structure of a typical tRNA shown in surface view. The charged ribose–phosphate backbone (shaded red) can be traced (pdb file 1ehz). (c) Diagram illustrating tRNA tertiary structure. The ribose–phosphate backbone is shown as a heavy continuous orange line. The longer black lines represent hydrogen bonds in double-helical regions; short black lines represent single bases in unpaired regions. Red lines represent hydrogen bonds stabilising the tertiary structure.
However, many RNA molecules are much larger than the small tRNA molecules, making it more difficult to choose between the many alternative theoretical conformations that could potentially form. Some RNAs may adopt different folding patterns under different circumstances, and thus modify their biological activity. Even so, although RNA structure prediction has its limits, some general structural motifs are apparent. It is clear that larger RNA molecules can be visualised as separate domains, each with a complex folding pattern made up of hairpin loops where the polyribonucleotide chain folds back on itself to allow base pairing between nearby lengths of RNA with complementary base sequences. Just as in proteins, each domain can be considered as an independently folded unit and the arrangement of these domains relative to each other defines the tertiary structure of the RNA. The stem of each hairpin loop is often a near-complementary base-paired region and, if long enough, the stem twists into a double helix, increasing stability by promoting base stacking. Two other features are important in the biological role of these hairpin loops. The unpaired bases in the tip of the loop, and those in the occasional nucleotide bulge along the stem, are susceptible to nuclease enzymes that recognise only single-stranded RNA (the so-called single-strand nucleases). Unpaired bases could also provide important opportunities for base pairing to other nucleic acid molecules and for recognition by specific proteins.
How might single-strand nucleases be used to confirm predicted RNA structures?
If a folded RNA is treated with a ribonuclease specific for single-stranded molecules, the products of the reaction can be analysed, thereby confirming unpaired regions within the structure or revealing the accessibility of these regions within a tertiary structure.