Efficient generation of recombinant RNA viruses using targeted recombination-mediated mutagenesis of bacterial artificial chromosomes containing full-length cDNA

Abstract

Background: Infectious cDNA clones are a prerequisite for directed genetic manipulation of RNA viruses. Here, a strategy to facilitate manipulation and rescue of classical swine fever viruses (CSFVs) from full-length cDNAs present within bacterial artificial chromosomes (BACs) is described. This strategy allows manipulation of viral cDNA by targeted recombination-mediated mutagenesis within bacteria.

Results: A new CSFV-BAC (pBeloR26) derived from the Riems vaccine strain has been constructed and subsequently modified in the E2 coding sequence, using the targeted recombination strategy to enable rescue of chimeric pestiviruses (vR26_E2gif and vR26_TAV) with potential as new marker vaccine candidates. Sequencing of the BACs revealed a high genetic stability during passages within bacteria. The complete genome sequences of rescued viruses, after extensive passages in mammalian cells showed that modifications in the E2 protein coding sequence were stably maintained. A single amino acid substitution (D3431G) in the RNA dependent RNA polymerase was observed in the rescued viruses vR26_E2gif and vR26, which was reversion to the parental Riems sequence.

Conclusions: These results show that targeted recombination-mediated mutagenesis provides a powerful tool for expediting the construction of novel RNA genomes and should be applicable to the manipulation of other RNA viruses.

Figure 1

Schematic representation of the CSFV genome organization and the BACs constructed and used in this study. Nucleotide (nt) and amino acid (aa) positions within R26 for the 5′ and 3′ termini together with the translational start and stop codons of the polyprotein coding region plus cleavage sites used to make the individual proteins (N^pro, C, E^rns, E1, E2, p7, NS2, NS3, NS4A, NS4B, NS5A and NS5B) are indicated. Insertion of the rpsL-neo in place of the TAV-epitope within CSFV E2 for the intermediate construct (R26_rpsLneo) and the subsequent replacement with the TTVSTSTLA sequence (R26_TAV) and the complete substitution of the E2 sequence (R26_E2gif) are shown. Names of BAC constructs begin with “pBelo” and rescued viruses with “v” (e.g. pBeloR26 and vR26). Cell culture passage no. of virus is indicated with “/P” (e.g. vR26/P-4).

Figure 2

Antibody reaction patterns of pestivirus infected cells. SFT-R cells were infected with vR26 and its two derivatives vR26_E2gif and vR26_TAV plus vGifhorn [26]. After 72 h, the cells were fixed and stained with monoclonal antibodies against the NS3 protein (WB103/105, left column), the CSFV E2 protein (WH303 and WH211, middle columns) and the BDV E2 protein (WB166, right column) as indicated and viewed using a fluorescence microscope.

Figure 3

Growth characteristics of vR26 compared to the parental vRiemser. The growth of the rescued vR26 and the parental vRiemser strains was evaluated in PK15 cells using an MOI of 0.1 pfu/cell. Virus titers were determined from harvests prepared at 2, 8, 24, 48, and 72 h post infection. Data are represented as mean + SD (n = 3).

Figure 4

Characteristics of the vR26 and the chimeric vR26_E2gif in cells. (A) Virus yield assays were performed during passage (from passage 1 to 12) of vR26 and vR26_E2gif in PK15 and SFT-R cells as indicated. For each passage, cells were harvested after 3 days and the virus titres determined. Data are presented as mean + SD (n = 2). (B) Cells (SFT-R) infected with the vR26/P-12 or with the vR26_E2gif /P-12 viruses were stained using a polyclonal anti-pestivirus serum (which recognizes both BDV and CSFV proteins) and with specific mAbs directed against the CSFV E2 (WH211) and BDV E2 (WB166) proteins as indicated.

Figure 5

Growth characteristics and quasispecies distribution of vR26/P-4, vR26/P-12, vR26_E2gif/P-4 and vR26_E2gif/P-12. (A) Time courses of virus yields after infection of SFT-R cells with vR26/P-4, vR26/P-12, vR26_E2gif/P-4 and vR26_E2gif/P-12. Cells were infected with the viruses (MOI of 0.1 plaque-forming units per cell) and harvested after 3, 24, 48 and 72 h, as indicated, and the virus titres determined (n = 1). (B) Single nucleotide variant (SNV) data obtained from cDNA generated from the rescued viruses vR26/P-4, vR26/P-12, vR26_E2Gif/P-4 and vR26_E2Gif/P-12. The bar diagram depicts the quasispecies distribution (as % variant) in the data set from the 454 FLX by Lofreq SNV-caller above 2% in frequency. SNVs (nt position in the cDNA) leading to silent mutations (-) or non-silent mutations (e.g. D3431G) are illustrated.