Schematic representation of the minireplicon replication system and primer extension analysis. (A) (a) A549 cells are infected with vTF7-3 and transfected with plasmids expressing the SeV minireplicon RNA and the N, P, and L proteins as described in Materials and Methods. The sequence of the minireplicon 3′ end is given along with that of the adjacent ribozyme (Rbz) on the plasmid to set the expected size of the primer extension product. (b) The T7 RNA polymerase transcript representing the positive sense minireplicon RNA is produced with the encapsidation nucleation site at the 5′ end (Enc). (c) Encapsidation of the T7 transcript by the N proteins produces a positive-strand minigenome containing AGP^R, the promoter first recognized by the SeV RNAp (P/L), at its 3′ end. (d) SeV RNAp copies the plus-strand minigenome to yield the minus-strand minigenome, which can in turn serve as a template for minigenome amplification (data not shown). The replicated minireplicon RNAs are then purified as nucleocapsids. (e) The encapsidated minus-strand RNAs (specific for active replication) are finally detected with a plus-strand ribo- or oligoprobe in Northern blots or primer extensions, respectively. (B) Primer extension (PE) of the replicated RNA. The IRD800 primer used for the extension and the sequencing is indicated. A digitalized image of primer extension is presented. Lane +P/N/L, complete replication reaction. The strong signal at the AGP 3′ end, corresponds to the extension of a product with the expected size, as indicated by the plasmid sequence shown (see the expected sequence in panel A). The parasite bands correspond to reverse transcription early stops caused by the arrest at hybrid structures resulting from 3′–5′ end complementarity of the RNA. These bands can be minimized by RNA fragmentation (see Materials and Methods). Moreover, they are not observed when 3′–5′ end self-complementation is not possible, as in constructs with AGP at the 5′ end and GP at the 3′ end (see for example constructs #19, #22, and #23 [Fig. 7]). Lane −L, negative control, primer extension run from a replication assay where the pGem-L plasmid was omitted. Primer extensions were performed with RNA purified from transfection-replication assays involving 0.5 × 10⁷ cells.

Testing the value of the 3′-OH congruence. (A) (Left) Schematic representation of plasmid constructs harboring an extended sequence past the AGP^R down to the ribozyme (Rbz) sequence (dark shaded rectangle). T7p, T7 RNA polymerase promoter. (Right) Expected 3′-OH ends of the T7 RNA transcript made by these plasmids. AGP^R is portrayed by three hexamers, as in Fig. 3A, to highlight the 3′-OH nature of the RNA. Constructs #6 and #8 (of 6n + 0 nucleotides) exhibit perfect 3′-OH congruence, while constructs #7 and #9 (6n + 4) exhibit dangling nucleotides. (B) Digitalized images of primer extensions (PE) performed with RNAs from replication of constructs #6, #7, #8, and #9, with sequences (Seq) obtained with the same primer on the corresponding plasmids. Arrowheads point to the first nucleotide of the ribozyme sequence. Primer extensions were performed with RNA purified from transfection-replication assays involving 4 × 10⁷ cells except for construct #1 (0.5 × 10⁷ cells).

Introducing constructs with two promoters. (A) Schematic representation of plasmid constructs expressing minireplicons harboring two promoters in tandem (AGP and GP) at the 3′-OH end of the T7 RNA transcript, as in construct #15. In constructs #16, #17, and #18, the activity of either AGP or GP is obliterated by a base substitution (X) made explicit by the name of the construct [for example, AGP-GP (C₉₁G) indicates that G in position 91 of GP has been substituted for C]. Construct #19 has a 6-nucleotide insertion between the two promoters. All the constructs led to the production of T7 RNA transcripts containing 6n + 0 nucleotides. (B) Digitalized images of primer extensions (PE) performed with RNAs from replication of constructs shown in panel A and of sequences (Seq) obtained with the same primer on the indicated plasmids. Primer extensions were performed with RNA purified from transfection-replication assays involving 4 × 10⁷ cells.

Structural features of SeV genome, promoters, and minireplicon. (A) Schematic representation of the SeV genome (−) with the plus (le⁺) and minus (le⁻) leaders, and the six transcription units (N, P, M, F, HN, and L) portrayed. Above it are shown the products of the transcription initiating at the GP only. Below it is shown the product of replication, the antigenome (+), which in turn serves as a template for genome amplification, starting at the AGP. (B) Detailed representation of GP and AGP primary structure and features. The sequence, portrayed as RNA of the sense template, is written in groups of 6 nucleotides (hexamers, numbered 1 through 16) to account for the fact that the N (small rectangle) contacts 6 nucleotides. Definitions of regions I and II are given in the text. Nucleotides 56 on GP and 58 on AGP, representing, respectively, the start of the N gene and the end of the L gene, are indicated. (C) Schematic representation of the parent minireplicon DNA sequence flanked by the T7 promoter (T7p) and the ribozyme (Rbz) as in plasmid pSP65 (see Materials and Methods). Restriction sites relevant for construction of derivatives, some harboring an insertion (CelII) (⧫), a deletion (BstX1) (◊), or an introduction of the Swap editing site (XbaI) (∗), are shown. The position of the IRD800 oligonucleotide used in the primer extensions is indicated.

Effect of the C proteins on compliance with the rule of six and genome length correction. (A) Schematic partial representations of positive-polarity minigenomes showing the nature of their 3′-OH ends, where AGP^R lies, featured with three hexamers, i.e., 3 times 6 nucleotides boxed within rectangles representing the N proteins. Construct #1, of 6n + 0 nucleotides, exhibits perfect congruence of the last 3′-OH nucleotide and the N protein. Construct #2, with a 4-nucleotide deletion (◊) at the BstX1 site (see Fig. 1C), has 6n + 2 nucleotides and is portrayed with 2 nucleotides dangling at the 3′-OH end. Construct #3 is like construct #2 but contains a Swap editing site (✶) inserted at XbaI (see Fig. 1C). During replication, if editing leads to a +4G addition, the genome length is corrected and congruence at the 3′-OH end is recovered (indicated in #3 and #5 as new 3′-OH ends boxed in by dotted lines). Constructs #4 and #5 correspond to constructs #2 and #3 but contain a +3 nucleotide addition (⧫) at the CelII site (see Fig. 1C) rather than the −4 deletion. #01, 6n + 0 construct with the Swap site. (B) Northern blot analysis of the replicated RNAs originating from the constructs presented in panel A. Replication assays were performed with the addition of plasmid pGem-P/C^wt or pGem-PC^stop (see Materials and Methods) to generate conditions where the C proteins were present or absent, respectively, during replication (+C or −C). −L, replication assay where the pGem-L plasmid was omitted. (C) Limited primer extension assay (see Materials and Methods and Results) across the editing site of the replicated RNAs from constructs #5 and #3 (lanes #5 and #3) or across the editing site of the plasmid expressing construct #5 (lane #5p). A regular sequence of the same site is presented below to allow positioning of the limited primer extension stops.

Testing the value of a change in the phase context with unrelated sequence at the 3′-OH end. (A) Schematic representation of the expected 3′-OH ends of T7 RNA transcripts derived from the indicated plasmids. Constructs #2, #3, and #11 through #14 contain a 4-nucleotide deletion at the BstX1 site (◊) (see Fig. 1). This deletion is associated with the presence of the Swap editing site (✶) (see Fig. 1) in constructs #3, #12, and #14. Sequence extensions of 10 or 100 nucleotides past AGP^R are also present in constructs #11 and #12 or #13 and #14, respectively. For all these constructs, AGP^R is portrayed by three hexamers (as in Fig. 3A) to highlight the 3′-OH congruence (as in #01, #11, #12, #13, and #14) or the change in promoter phase context (as in #2 and #3). Note that the presence of the editing site in constructs #3, #12, and #14 can correct the RNA length by a 4-nucleotide addition, if replication starts at the out-of-phase AGP^R. This is indicated by the corrected new 3′-OH end, featured in #3 and #12 (dotted-line boxes). (B) Digitalized images of primer extensions (PE) performed with RNAs from replication of constructs shown in panel A and of sequences (Seq) obtained with the same primer on the corresponding plasmids. When appropriate, arrowheads point to the first nucleotide of the ribozyme sequence. Primer extensions were performed with RNA purified from transfection-replication assays involving 4 × 10⁷ cells.

Testing the value of a change in the phase context with two promoters in tandem. (A) Schematic representation (as in Fig. 6) of the expected 3′-OH ends of T7 RNA transcripts derived from plasmids carrying two promoters with a 6 (constructs #19 and #20)- or 3 (constructs #21, #22, and #23)-nucleotide insertion between them. Constructs #20, #22, and #23 also contain a 3-nucleotide addition at the CelII site (⧫) (see Fig. 1C). Finally, construct #23 harbors a Swap editing site (✶). For this construct, if initiation takes place at the out-of-phase AGP, the genome length correction generates a new 3′-OH end with a congruent AGP (dotted-line boxes) (see also Fig. 6). (B) Digitalized images of primer extensions (PE) performed with RNAs from replication of constructs shown in panel A and of sequences (Seq) obtained with the same primer on the indicated plasmids. Primer extensions were performed with RNA purified from transfection-replication assays involving 4 × 10⁷ cells.