Vesicular stomatitis virus mRNA map and UV map. The genome is shown as a light green line at the top, and the N, P, M, G, and L genes and their relative sizes are indicated. The 47-nucleotide leader RNA is encoded at the 3′ end of the genomic RNA. The leader and intergenic regions are shown in orange. The RNAs encoded at the 3′ end of the genome are made in larger quantities than the RNAs encoded at the 5′ end of the genome. UV irradiation experiments determined the size of the vesicular stomatitis virus genome (UV target size) required for synthesis of each of the viral mRNAs. The UV target size of each viral mRNA corresponded to the size of the genomic RNA sequence encoding the mRNA plus all of the genomic sequence 3′ to this coding sequence. The transition from reiterative copying and termination to initiation is not perfect, and only about 70 to 80% of the polymerase molecules accomplish this transition at each intergenic region. Such inefficiency accounts for the observation that mRNAs encoded by 3′-proximal sequences are more abundant than those from 5′-proximal sequences.

      The effects of ultraviolet (UV) irradiation provided insight into the mechanism of vesicular stomatitis virus mRNA synthesis. In these experiments, virus particles were irradiated with UV light, and the effect on the synthesis of individual mRNAs was assessed. UV light causes the formation of pyrimidine dimers that block passage of the RNA polymerase. In principle, larger genes require less UV irradiation to inactivate mRNA synthesis and have a larger target size. The dose of UV irradiation needed to inactivate synthesis of the N mRNA corresponded to the predicted size of the N gene, but this was not the case for the other viral mRNAs. The target size of each other mRNA was the sum of its size plus the size of other genes located 3′ to it. For example, the UV target size of the L mRNA is the size of the entire genome. These results indicate that these mRNAs are synthesized sequentially, in the 3′ → 5′ order in which their genes are arranged in the viral genome: N-P-M-G-L.

Vesicular stomatitis virus mRNA synthesis illustrates a second mechanism for poly(A) addition: reiterative copying of, or “stuttering” at, a short U sequence in the (−) strand template. After initiation, vesicular stomatitis virus mRNAs are elongated until the RdRP reaches a conserved stop-polyadenylation signal (3′-AUACU₇-5′) located in each intergenic region (Fig. 6.21). Poly(A) (approximately 150 nucleotides) is added by reiterative copying of the U stretch, followed by termination.

      The transition from mRNA to genome RNA synthesis in cells infected with vesicular stomatitis virus is dependent on the viral nucleocapsid (N) protein (Fig. 6.19). To produce a full-length (+) strand RNA, the stop-start reactions at intergenic regions must be suppressed, a process that depends on the synthesis of the N and P proteins. The P protein maintains the N protein in a soluble form so that it can encapsidate the newly synthesized genomic RNA. N-P assemblies bind to leader RNA and cause antitermination, signaling the polymerase to begin processive RNA synthesis. Additional N protein molecules then associate with the (+) strand RNA as it is elongated, and eventually bind to the seven A bases in the intergenic region. This interaction blocks reiterative copying of the seven U bases in the genome because the A bases cannot slip backward along the genomic RNA template. Consequently, RNA synthesis continues through the intergenic regions. The number of N-P protein assemblies in infected cells therefore regulates the relative efficiencies of mRNA synthesis and genome RNA replication. The copying of full-length (+) strand RNAs to (−) strand genomic RNAs also requires the binding of N-P proteins to elongating RNA molecules. Newly synthesized (−) strand RNAs are produced as nucleocapsids that can be packaged readily into progeny viral particles.

Figure 6.21 Poly(A) addition and termination at an intergenic region during vesicular stomatitis virus mRNA synthesis. Copying of the last seven U residues of an mRNA-encoding sequence is followed by slipping of the resulting seven A residues in the mRNA off the genomic sequence, which is then recopied. This process continues until approximately 200 A residues are added to the 3′ end of the mRNA. Termination then occurs, followed by initiation and capping of the next mRNA. The dinucleotide NA in the genomic RNA is not copied.

      The (−) strand RNA genome of paramyxoviruses is copied efficiently only when its length in nucleotides is a multiple of 6. This requirement, called the rule of six, is probably a consequence of the association of each N monomer with six nucleotides. Assembly of the nucleocapsid begins with the first nucleotide at the 5′ end of the RNA and continues until the 3′ end is reached. If the genome length is not a multiple of 6, then the 3′ end of the genome will not be precisely aligned with the last N monomer. Such misalignment reduces the efficiency of initiation of RNA synthesis at the 3′ end. Curiously, although the N protein of rhabdoviruses binds nine nucleotides of RNA, the genome length need not be a multiple of this number for efficient copying.

The segmented (−) strand RNA genome of influenza virus is expressed by the synthesis of subgenomic mRNAs in infected cells by the heterotrimeric RdRP described previously (Fig. 6.12). Individual mRNAs are initiated with a capped primer derived from host cell mRNA, and terminate 20 nucleotides short of the template 3′ end. Polyadenylation of these mRNAs is achieved by a similar mechanism to that observed during vesicular stomatitis virus mRNA synthesis, reiterative copying of a short U sequence in the (−) strand template. Such copying is thought to be a consequence of the RdRP binding specifically to the 5′ end of (−) strand RNA and remaining at this site throughout mRNA synthesis. The genomic RNAs are threaded through the polymerase in a 3′ → 5′ direction as mRNA synthesis proceeds (Fig.