Transcription and Replication

The first event in the HDV replication cycle is likely to be the transcription of the 0.8-kb S-HDAg-encoding mRNA species since this protein is required for initiation of HDV RNA replication (Kuo et al. 1989). Originally, the mechanism for production of this transcript was thought to be an adjunct of rolling circle replication. This proposal, put forward from John Taylor's laboratory (Hsieh and Taylor 1991), suggested that initiation of antigenomic HDV RNA synthesis from virion RNA occurs at position 1631, just upstream of the HDAg AUG codon. Transcription continues downstream until reaching the poly(A) addition signal, after which the nascent transcript is cleaved and a poly(A)+ tail attached. However, as is the case with eukaryotic mRNA transcription, the polymerase continues to transcribe downstream of the cleavage site. Normally, this downstream transcript is very unstable and is rapidly degraded. However, in the case of HDV, it was proposed that a ribozyme cleavage event occurring 34 nucleotides past the poly(A) cleavage/addition site stabilizes this downstream transcript to allow initiation of rolling circle replication of antigenomic HDV RNA (Hsieh and Taylor 1991). As the replication proceeds, the poly(A) addition signal is thereafter inhibited by a combination of HDAg and intramocular binding with nucleotides on the opposite side of the rod structure (Hsieh and Taylor 1991). Evidence for this model were based on observations using HDV cDNA constructs where transcription was under the control of foreign promoters, often in the absence of HDAg (Hsieh and

Taylor 1991; Nie et al. 2004). Indeed, under these conditions, it was possible to produce both HDAg mRNA transcripts and full-length antigenomic HDV RNA. However, there are no cDNA intermediates in natural HDV replication. Thus, not surprisingly, this model cannot easily account for several observations made during natural HDV RNA-dependent RNA replication. The most significant of these is the implication from this model that mRNA synthesis can occur only once per initiation of replication which, in turn, implies that mRNA would primarily be synthesized early in the RNA replication cycle. This is a very restrictive limitation because HDAg needs to be continually synthesized both for maintenance of replication and for packaging after the RNA editing has occurred. Work in our laboratory has suggested a different hypothesis. In particular, we demonstrated that the amount of the 0.8-kb mRNA initially showed a steady increase and was then maintained at the same level throughout the HDV replication cycle (Modahl and Lai 1998). These results suggested that the transcription of the mRNA and replication of the HDV genome are independent processes and occur concurrently. However, this model raises a different problem as it requires that mRNA transcription and RNA replication be synchronized on the same genomic sense HDV RNA template such that the poly(A) addition signal is recognized during transcription but not replication. A potential solution to this issue was suggested from recent observations in our laboratory indicating that transcription of mRNA and genomic to antigenomic RNA replication may be carried out by different polymerases and probably in different subnuclear domains (Modahl et al. 2000; Macnaughton et al. 2002; see below). Thus, these two processes maybe both physically and biochemically separated in HDV-infected cells. There is a caveat to this, however, as infection requires only a single input genomic HDV RNA species. Thus, to serve as a template for both mRNA transcription and production of full-length antigenomic HDV RNA, this original infecting RNA species must be moved between different subnuclear domains. Since mRNA transcription is likely occur first, this relocalization event probably depends on newly synthesized HDAg. If so, it might be expected that the de novo synthesized HDAg would be, in some way, different from the HDAg species present in the original infecting ribonucleoprotein complex, most likely due to different post-translational modifications (see next section).

0 0

Post a comment