Rotaviruses are major causes of severe gastroenteritis in infants and young children, as well as the young of many other mammalian and avian species1. As members of the Reoviridae family, rotaviruses have a segmented double-stranded RNA (dsRNA) genome. The genome segments are contained within a nonenveloped icosahedral virion formed from three concentric layers of protein2. Based on sequencing and phylogenetic analysis of the genome segments, nine species of rotavirus (A−D, F−J) have been defined3. Those strains comprising rotavirus species A are responsible for the vast majority of human disease4. The introduction of rotavirus vaccines into childhood immunization programs beginning in the last decade is correlated with significant reductions in rotavirus mortality and morbidity. Most notably, the number of rotavirus-associated childhood deaths has decreased from approximately 528,000 in 2000 to 128,500 in 20164,5. Rotavirus vaccines are formulated from live attenuated strains of the virus, with 2 to 3 doses administered to children by 6 months of age. The large number of genetically diverse rotavirus strains circulating in humans and other mammalians species, combined with their ability to rapidly evolve through mutagenesis and reassortment, may lead to antigenic changes in the types of rotaviruses infecting children6,7,8. Such changes may undermine the efficacy of existing vaccines, requiring their replacement or modification.
The development of fully plasmid-based reverse genetics systems enabling manipulation of any of the 11 rotavirus genome segments was only recently achieved9. With the availability of these systems, it has become possible to unravel molecular details of rotavirus replication and pathogenesis, to develop improved high-throughput screening methods for anti-rotavirus compounds, and to create new potentially more effective classes of rotavirus vaccines. During rotavirus replication, capped viral (+)RNAs not only guide the synthesis of viral proteins, but also serve as templates for the synthesis of progeny dsRNA genome segments10,11. All rotavirus reverse genetics systems described to date rely on the transfection of T7 transcription vectors into mammalian cell lines as a source of cDNA-derived (+)RNAs used in recovering recombinant viruses9,12,13. Within the transcription vectors, full-length viral cDNAs are positioned between an upstream T7 promoter and downstream hepatitis delta virus (HDV) ribozyme such that viral (+)RNAs are synthesized by T7 RNA polymerase that contain authentic 5’ and 3’-termini (Figure 1A). In the first-generation reverse genetics system, recombinant viruses were made by transfecting baby hamster kidney cells expressing T7 RNA polymerase (BHK-T7) with 11 T7 (pT7) transcription vectors, each directing synthesis of a unique (+)RNA of the simian SA11 virus strain, and three CMV promoter-drive expression plasmids, one encoding the avian reovirus p10FAST fusion protein and two encoding subunits of the vaccinia virus D1R-D12L capping enzyme complex9. Recombinant SA11 viruses generated in transfected BHK-T7 cells were amplified by overseeding with MA104 cells, a cell line permissive for rotavirus growth. A modified version of the first-generation reverse genetics system has been described that no longer uses support plasmids12. Instead, the modified system successfully generates recombinant rotaviruses simply by transfecting BHK-T7 cells with the 11 SA11 T7 transcription vectors, with the caveat that vectors for the viral factory (viroplasm) building blocks (nonstructural proteins NSP2 and NSP5) are added at levels 3-fold higher than the other vectors14,15. Modified versions of the reverse genetics system have also been developed that support the recovery of the human KU and Odelia strains of rotavirus16,17. The rotavirus genome is remarkably amenable to manipulation by reverse genetics, with recombinant viruses generated to date with mutations introduced into VP418, NSP19, NSP219, NSP320,21, and NSP522,23. Among the most useful viruses generated so far are those that have been engineered to express fluorescent reporter proteins (FPs)9,12,21,24,25.
In this publication, we provide the protocol for the reverse genetics system that we use in our laboratory to generate recombinant strains of SA11 rotavirus. The key feature of our protocol is co-transfection of BHK-T7 cells with the 11 pT7 transcription vectors (modified to include 3x levels of the pT7/NSP2SA11 and pT7/NSP5SA11 vectors) and a CMV expression vector encoding the African swine fever virus (ASFV) NP868R capping enzyme21 (Figure 2). In our hands, presence of the NP868R plasmid leads to the production of higher titers of recombinant viruses by transfected BHK-T7 cells. In this publication, we also provide a protocol for modifying the pT7/NSP3SA11 plasmid such that recombinant viruses can be generated that express not only the segment 7 protein product NSP3 but also a separate FP. This is accomplished by re-engineering the NSP3 open reading frame (ORF) in the pT7/NSP3SA11 plasmid to contain a downstream 2A translational stop-restart element followed by an FP ORF (Figure 1B)24,26. Through this approach, we have generated recombinant rotaviruses expressing various FPs: UnaG (green), mKate (far-red), mRuby (red), TagBFP (blue), CFP (cyan), and YFP (yellow)24,27,28. These FP-expressing rotaviruses are made without deleting the NSP3 ORF, thus yielding viruses that are expected to encode a full complement of functioning viral proteins.