RNA Structure

It was promptly deduced from the first full sequence of an HDV genome that this RNA could theoretically be folded into an unbranched rod-like structure (Wang et al. 1986). Further theoretical calculations predicted a high negative free energy of 805 kcal/mol, consistent with a stable structure (Kuo et al. 1988a). Experimental evidence also supports such a structure. In electron microscopic studies, the genomic RNA appears as a short double-stranded rod that upon progressive denaturation opens into a circle (Kos et al. 1986). In addition, by gel electrophoresis under nondenaturing conditions, both the genomic and antigenomic RNAs migrate as expected for double-stranded species (Lazinski and Taylor 1995). Moreover, upon prior denaturation, most of the RNAs migrate consistent with a circular conformation, the remainder behaving as linear species (Chen et al. 1986). Similar conclusions apply for the structure of the unit-length antigenome.

While the rod-like folding is generally true, the details of the exact folding remain to be determined. One study attempted to use nuclease susceptibility assays to test the folding of a segment at one end of the rod-like structure of the genomic RNA in vitro. The detected folding was very close to that predicted (Beard et al. 1996).

It is obvious that the structures of the two ribozyme domains are not compatible with the rod-like folding. Furthermore, folding of these domains into the rod-like structure should inhibit the ribozymes. Intuitively, if this inhibition were not the case the circular RNAs might undergo efficient self-cleavage to form linear RNAs. The prediction that the rod-like folding overrides the ability to fold into the active ribozyme conformation has been proven using both in vivo and in vitro studies of natural and modified HDV RNAs (Lazinski and Taylor 1993, 1994a). Modified RNAs that cannot fold the ribozyme domain into a rod-like structure do not form stable circles in vivo. Conversely, unmodified HDV RNA circles are cleaved only inefficiently by ribozymes in vitro. And yet if these RNAs are first hybridized with a separate oligonu-cleotide to stop the ribozyme domain from being inactivated by being drawn into the rod-like folding, then these RNAs are efficiently cleaved by ribozymes in vitro.

Additional alternative foldings of HDV RNA sequences have also been reported. One such pairing is considered to produce a binding site for the protein PKR (Circle et al. 2003, 1997; Robertson et al. 1996). Another alternative folding, also based on in vitro data, has been proposed to explain a specific cross-linking induced by irradiation with ultraviolet light (Branch et al. 1989).

In summary, it should not be surprising that HDV RNAs can, and do have, multiple ways in which they can fold. It is considered that most RNAs undergo a series of alternative foldings, that is, metastable states that can facilitate the molecule ultimately achieving a predominant final more stable structure (Uhlenbeck 1995). On top of this, as will be considered in Sect. 2.5, the final structure for HDV genomic and antigenomic RNAs probably involves the consequences of S-HDAg binding.

0 0

Post a comment