Double Rolling Circle Model

For some time, attempts have been made to create models for the replication of HDV RNAs (Flint et al. 2004; Gerin et al. 2001; Macnaughton et al. 2002; Taylor 1990). Most of these attempts have borrowed from the concept of a 'rolling-circle model of replication.' This idea was applied previously to the replication of RNAs of the plant viroids (Branch and Robertson 1984), with which HDV has numerous similarities, as discussed in detail elsewhere (Taylor 1999).

The viroid RNAs differ from HDV RNAs in that they are several times smaller and are noncoding. This, together with the fact that plant viroid RNAs are never assembled into virus-like particles, means that the concept of a viroid 'genome' has to be different for HDV. For some viroids RNA-directed RNA replication leads to the accumulation of unit-length RNA circles of both polarities. In contrast, for most viroids, circles of only one polarity can be found. For the complementary strand the RNA template is a linear RNA multimer of unit-length. It is considered that these linear RNAs act as templates for multimeric transcripts that can be processed to unit-length circular RNAs. Thus, for these two classes of viroids the replication models are described as double- and single-rolling circle models, respectively.

For HDV, since both the genome and antigenome exist as unit-length circles, it was quickly extrapolated from the viroids, that the replication would

Table 3 Some revisions to rolling-circle model of HDV genome replication

1. Nascent antigenomic transcripts can be ribozyme-cleaved independent of polyadenylation (Nie et al. 2004)

2. Nascent antigenomic transcripts can be polyadenylated independent of ribozyme cleavage (Nie et al. 2004)

3. Nascent antigenomic transcripts, undergo polyadenylation or ribozyme-cleavage as largely alternative processing events (Nie et al. 2004)

4. While circular forms of genomic and antigenomic RNA are preferred templates for RNA-directed transcription, linear forms can also act (Gudima et al. 2004)

be a double-rolling circle model. Furthermore, this basic model had to be adapted, to allow for the fact that a polyadenylated mRNA was also being produced. This model, with time, has required a number of additional modifications, some of which are listed in Table 3.

Without resorting to a diagram, Table 4 attempts to describe the series of events that could be considered as essential steps in HDV genome replication. At the same time it should be clear from this review, and maybe from the other reviews in this volume, that for many of these steps we have yet to obtain actual experimental evidence.

0 0

Post a comment