Comparative analysis of Reoviridae reverse genetics methods

Abstract

Effective methods to engineer the segmented, double-stranded RNA genomes of Reoviridae viruses have only recently been developed. Mammalian orthoreoviruses (MRV) and bluetongue virus (BTV) can be recovered from entirely recombinant reagents, significantly improving the capacity to study the replication, pathogenesis, and transmission of these viruses. Conversely, rotaviruses (RVs), which are the major etiological agent of severe gastroenteritis in infants and children, have thus far only been modified using single-segment replacement methods. Reoviridae reverse genetics techniques universally rely on site-specific initiation of transcription by T7 RNA polymerase to generate the authentic 5' end of recombinant RNA segments, but they vary in how the RNAs are introduced into cells: recombinant BTV is recovered by transfection of in vitro transcribed RNAs, whereas recombinant MRV and RV RNAs are transcribed intracellularly from transfected plasmid cDNAs. Additionally, several parameters have been identified in each system that are essential for recombinant virus recovery. Generating recombinant BTV requires the use of 5' capped RNAs and is enhanced by multiple rounds of RNA transfection, suggesting that translation of viral proteins is likely the rate-limiting step. For RV, the efficiency of recovery is almost entirely dependent on the strength of the selection mechanism used to isolate the single-segment recombinant RV from the unmodified helper virus. The reverse genetics methods for BTV and RV are presented and compared to the previously described MRV methods. Analysis and comparison of each method suggest several key lines of research that might lead to a reverse genetics system for RV, analogous to those used for MRV and BTV.

Figure 1. Comparison of Reoviridae reverse genetics methods with authentic virus replication

(A) Schematic of Reoviridae virus replication. A multilayered, icosahedral infectious virus particle (IVP) binds and enters a target cell, prompting loss of the outer capsid. Once in the cytosol, the resulting subviral particle (SVP) becomes transcriptionally active and extrudes capped, [+]RNAs that serve as templates for viral protein synthesis and dsRNA genome replication. Viral inclusions are the sites of genome replication and nascent IVP assembly. (B) MRV reverse genetics. Ten plasmid cDNAs (corresponding to the ten MRV genome segments) are transfected into cells expressing T7 RNA polymerase (T7pol, green brick), wherein the cDNAs are transcribed into MRV [+]RNAs (blue squiggle). As with authentic virus, these [+]RNAs serve as templates for viral protein expression and dsRNA genome replication, resulting in formation of entirely recombinant IVPs (red). (C) BTV reverse genetics. Linearized plasmid cDNAs are transcribed in vitro using T7pol to generate a complement of the ten BTV [+]RNAs. The [+]RNAs are transfected into cells and are used for viral protein expression and virus replication to generate recombinant IVPs. (D) RV single-segment replacement. A plasmid cDNA encoding a single RV genome segment is transfected into cells expressing T7pol, followed by infection with a helper RV. Few helper virus particles incorporate the recombinant [+]RNA into their genome during replication, in place of the native species. This results in a virus population that is largely unmodified helper RV, but also contains a small population of single-segment recombinant RV (indicated by the red and grey IVP). Several selection strategies have been devised to isolate the single-segment recombinant RV from the helper virus.

Figure 2. Reoviridae cDNA construction

(A) A schematic of MRV cDNAs. A minimal T7 RNA polymerase (T7pol) promoter is used to generate the authentic 5′ end of each segment, including the 5′ consensus sequence (5′CS) present in all genome segments. The native 3′ end of each cDNA is generated by a hepatitis delta virus (HDV) ribozyme that is positioned to specifically cleave (arrow) after the 3′ consensus sequence (3′CS). A downstream T7pol terminator sequence (not shown) halts transcription after the HDV ribozyme sequence. (B) A schematic of BTV cDNAs. A minimal T7pol promoter is used to generate the authentic 5′ end and 5′CS, as with MRV. BTV plasmid cDNAs are linearized with a non-palindromic restriction enzyme positioned in the antisense orientation (BsaI is shown here, as an example) that cleaves the template cDNA strand immediately following the 3′CS. In vitro, run-off transcription by T7pol yields the correct 3′ end. (C) A schematic of RV cDNAs. RV cDNAs are constructed, and function, virtually identically to those of MRV (A).

Figure 3. Bicistronic, single-segment plasmids can function as an internal control in RV single-segment replacement experiments

(A) Diagram of two bicistronic RV plasmids. The SA11_AUK_B construct has a wild-type (wt)-like SA11 strain (yellow) genome segment 8 (encoding the NSP2 protein) in the “A” position, a small linker, and a chimeric (mutant) segment 8 derived from the UK strain (red) in the “B” position. Each cDNA has its own T7 RNA polymerase promoter (T7P; crooked arrow), ribozyme (triangle), and T7 terminator sequence (T7T). The UK_ASA11_B construct is similarly constructed, except that the UK segment 8 is in the A position, and the SA11 segment 8 in the B position. (B) Titers of recombinant virus from a reverse genetics experiment. Plasmids encoding the wt-like SA11 segment 8 (SA11g8^R) or the two bicistronic plasmids were used to generate recombinant virus. After two rounds of selection using MA104-g8D cells cultured at 39°C (see sections 4.1 and 4.6, and reference (7)), viruses were titered using MA104 cells. RT-PCR and sequence analysis of the second-passage supernatants confirmed that each plasmid successfully generated single-segment recombinant RV (not shown). (C) Analysis of plaque isolates. Individual plaques (n=8 per sample) were isolated from the plaque assays performed in (B) and amplified in individual wells of a 12-well plate of MA104 cells. RT-PCR and sequence analysis allowed typing of the segment 8 from each isolate. The identities of the isolates are colored as in (A). Note that (i) recovery of UK recombinant virus is lower than SA11 in all cases and (ii) UK_ASA11_B yields a lower titer of virus (B) but a higher frequency of UK virus (relative to SA11) (C), suggesting that placement of the mutant segment in the A position and wt-like control in the B position allows for more efficient isolation and recovery of the mutant virus.