We have recovered infectious Sendai virus (SeV) from full‐length cDNA (FL‐3) by transfecting this cDNA and pGEM plasmids expressing the nucleocapsid protein (NP), phosphoprotein and large proteins into cells infected with a vaccinia virus which expresses T7 RNA polymerase. These cells were then injected into chicken eggs, in which SeV grows to very high titers. FL‐3 was marked with a BglII site in the leader region and an NsiI site (ATGCAT) in the 5′ nontranslated region of the NP gene, creating a new, out‐of‐frame, 5′ proximal AUG. All the virus stocks generated eventually removed this impediment to NP expression, by either point mutation or recombination between FL‐3 and pGEM‐NP. The recovery system was found to be highly recombinogenic. Even in the absence of selective pressure, one in 20 of the recombinant SeV generated had exchanged the NP gene of FL‐3 with that of pGEM‐NP. When a fifth plasmid containing a new genomic 3′ end without the presumably deleterious BglII site was included as another target for recombination, the new genomic 3′ end was found in the recombinant SeV in 12 out of 12 recoveries. Using this approach, a novel copy‐back nondefective virus was generated which interferes with wild‐type virus replication.

The template for paramyxovirus RNA synthesis is not naked RNA but the helical nucleocapsid core of the virus, in which each nucleocapsid protein (N protein) is predicted to be associated with precisely 6 nucleotides (nt) (11). Presumably as a consequence of this association, paramyxovirus genomes are replicated efficiently only when they are a multiple of 6 nt in length, and this has been dubbed the "rule of six" (4). The structure of paramyxovirus nucleocapsids is thus central to understanding how this rule might operate.

No abstract

The P gene of bovine parainfluenza virus 3 (bPIV3) contains two downstream overlapping ORFs, called V and D. By comparison with the mRNA editing sites of other paramyxoviruses, two editing sites were predicted for bPIV3; site a to express the D protein, and site b to express the V protein. Examination of the bPIV3 mRNAs, however, indicates that site b is non‐functional whereas site a operates frequently. Insertions at site a give rise to both V and D protein mRNAs, because a very broad distribution of Gs is added when insertions occur. This broad distribution is very different from the editing sites of Sendai virus or SV5, where predominantly one form of edited mRNA containing either a one or two G insertion respectively is created, to access the single overlapping ORF of these viruses. A model is proposed to explain how paramyxoviruses control the range of G insertions on that fraction of the mRNAs where insertions occur. The bPIV3 P gene is unique as far as we know, in that a sizeable portion of the gene expresses all 3 reading frames as protein. bPIV3 apparently does this from a single editing site by removing the constraints which control the number of slippage rounds which take place.

We have used a cDNA copy of a natural, internally deleted, Sendai virus defective interfering genome to study the effect of insertions and deletions (which maintain the hexamer genome length) on the ability of viral genomes to be amplified in a transfected cell system. The insertion of 18 nt at nt72 (In the 5' untranslated region of the N gene, just downstream of the le+ region) was found to be lethal, whereas similar insertions further from the genome ends were well tolerated. Curiously, the insertion of 6 nt on either side of the le+/N junction (at nt47 and nt87) was well tolerated, but the insertion of 12 nt at either site, or of 6 nt at both sites, largely ablated genome amplification. These results suggest that an element of this replication promoter is located downstream of nt87, in the 5' untranslated region of the first gene. Remarkably, the addition of 6 nt by the insertion of 2, 3, or 4 nt at nt47 plus the insertion of 4, 3, or 2 nt, respectively, at nt87 was poorly tolerated, presumably because the hexamer phase of the intervening sequence was altered with respect to the N subunits of the template. These results suggest that the rule of six operates, at least in part, at the level of the initiation of antigenome synthesis.