Attenuation hotspots in neurotropic human astroviruses.

During the last decade, the detection of neurotropic astroviruses has increased dramatically. The MLB genogroup of astroviruses represents a genetically distinct group of zoonotic astroviruses associated with gastroenteritis and severe neurological complications in young children, the immunocompromised, and the elderly. Using different virus evolution approaches, we identified dispensable regions in the 3' end of the capsid-coding region responsible for attenuation of MLB astroviruses in susceptible cell lines. To create recombinant viruses with identified deletions, MLB reverse genetics (RG) and replicon systems were developed. Recombinant truncated MLB viruses resulted in imbalanced RNA synthesis and strong attenuation in iPSC-derived neuronal cultures confirming the location of neurotropism determinants. This approach can be used for the development of vaccine candidates using attenuated astroviruses that infect humans, livestock animals, and poultry.

Innate Intracellular Antiviral Responses Restrict the Amplification of Defective Virus Genomes of Parainfluenza Virus 5.

During the replication of parainfluenza virus 5 (PIV5), copyback defective virus genomes (DVGs) are erroneously produced and are packaged into "infectious" virus particles. Copyback DVGs are the primary inducers of innate intracellular responses, including the interferon (IFN) response. While DVGs can interfere with the replication of nondefective (ND) virus genomes and activate the IFN-induction cascade before ND PIV5 can block the production of IFN, we demonstrate that the converse is also true, i.e., high levels of ND virus can block the ability of DVGs to activate the IFN-induction cascade. By following the replication and amplification of DVGs in A549 cells that are deficient in a variety of innate intracellular antiviral responses, we show that DVGs induce an uncharacterized IFN-independent innate response(s) that limits their replication. High-throughput sequencing was used to characterize the molecular structure of copyback DVGs. While there appears to be no sequence-specific break or rejoining points for the generation of copyback DVGs, our findings suggest there are region, size, and/or structural preferences selected for during for their amplification.IMPORTANCE Copyback defective virus genomes (DVGs) are powerful inducers of innate immune responses both in vitro and in vivo They impact the outcome of natural infections, may help drive virus-host coevolution, and promote virus persistence. Due to their potent interfering and immunostimulatory properties, DVGs may also be used therapeutically as antivirals and vaccine adjuvants. However, little is known of the host cell restrictions which limit their amplification. We show here that the generation of copyback DVGs readily occurs during parainfluenza virus 5 (PIV5) replication, but that their subsequent amplification is restricted by the induction of innate intracellular responses. Molecular characterization of PIV5 copyback DVGs suggests that while there are no genome sequence-specific breaks or rejoin points for the generation of copyback DVGs, genome region, size, and structural preferences are selected for during their evolution and amplification.

Unusual, stable replicating viruses generated from mumps virus cDNA clones.

In reverse genetic experiments we have isolated recombinant mumps viruses (rMuV) that carry large numbers of mutations clustered in small parts of their genome, which are not caused by biased hyper-mutation. In two separate experiments we obtained such recombinant viruses: one virus had 11 mutations in the V/P region of the genome; the other, which also contained an extra transcription unit encoding green fluorescent protein (EGFP), had 32 mutations in the N gene. These specific sets of mutations have not been observed in naturally occurring MuV isolates. Unusually, the vast majority of the mutations (48/51) were synonymous. On passage in Vero cells and human B-LCL cells, a B lymphocyte-like cell line, these mutations appear stable as no reversion occurred to the original consensus sequence, although mutations in other parts of the genome occurred and changed in frequency during passage. Defective interfering RNAs accumulate in passage in Vero cells but not in B-LCL cells. Interestingly, in all passaged samples the level of variation in the EGFP gene is the same as in the viral genes, though it is unlikely that this gene is under any functionality constraint. What mechanism gave rise to these viruses with clustered mutations and their stability remains an open question, which is likely of interest to a wider field than mumps reverse genetics.

Analysis of Paramyxovirus Transcription and Replication by High-Throughput Sequencing.

We have developed a high-throughput sequencing (HTS) workflow for investigating paramyxovirus transcription and replication. We show that sequencing of oligo(dT)-selected polyadenylated mRNAs, without considering the orientation of the RNAs from which they had been generated, cannot accurately be used to analyze the abundance of viral mRNAs because genomic RNA copurifies with the viral mRNAs. The best method is directional sequencing of infected cell RNA that has physically been depleted of ribosomal and mitochondrial RNA followed by bioinformatic steps to differentiate data originating from genomes from viral mRNAs and antigenomes. This approach has the advantage that the abundance of viral mRNA (and antigenomes) and genomes can be analyzed and quantified from the same data. We investigated the kinetics of viral transcription and replication during infection of A549 cells with parainfluenza virus type 2 (PIV2), PIV3, PIV5, or mumps virus and determined the abundances of individual viral mRNAs and readthrough mRNAs. We found that the mRNA abundance gradients differed significantly between all four viruses but that for each virus the pattern remained relatively stable throughout infection. We suggest that rapid degradation of non-poly(A) mRNAs may be primarily responsible for the shape of the mRNA abundance gradient in parainfluenza virus 3, whereas a combination of this factor and disengagement of RNA polymerase at intergenic sequences, particularly those at the NP:P and P:M gene boundaries, may be responsible in the other viruses.IMPORTANCE High-throughput sequencing (HTS) of virus-infected cells can be used to study in great detail the patterns of virus transcription and replication. For paramyxoviruses, and by analogy for all other negative-strand RNA viruses, we show that directional sequencing must be used to distinguish between genomic RNA and mRNA/antigenomic RNA because significant amounts of genomic RNA copurify with poly(A)-selected mRNA. We found that the best method is directional sequencing of total cell RNA, after the physical removal of rRNA (and mitochondrial RNA), because quantitative information on the abundance of both genomic RNA and mRNA/antigenomes can be simultaneously derived. Using this approach, we revealed new details of the kinetics of virus transcription and replication for parainfluenza virus (PIV) type 2, PIV3, PIV5, and mumps virus, as well as on the relative abundance of the individual viral mRNAs.

The switch between acute and persistent paramyxovirus infection caused by single amino acid substitutions in the RNA polymerase P subunit.

Paramyxoviruses can establish persistent infections both in vitro and in vivo, some of which lead to chronic disease. However, little is known about the molecular events that contribute to the establishment of persistent infections by RNA viruses. Using parainfluenza virus type 5 (PIV5) as a model we show that phosphorylation of the P protein, which is a key component of the viral RNA polymerase complex, determines whether or not viral transcription and replication becomes repressed at late times after infection. If the virus becomes repressed, persistence is established, but if not, the infected cells die. We found that single amino acid changes at various positions within the P protein switched the infection phenotype from lytic to persistent. Lytic variants replicated to higher titres in mice than persistent variants and caused greater infiltration of immune cells into infected lungs but were cleared more rapidly. We propose that during the acute phases of viral infection in vivo, lytic variants of PIV5 will be selected but, as the adaptive immune response develops, variants in which viral replication can be repressed will be selected, leading to the establishment of prolonged, persistent infections. We suggest that similar selection processes may operate for other RNA viruses.

Genome Sequence of the Parainfluenza Virus 5 Strain That Persistently Infects AGS Cells.

We have sequenced the parainfluenza virus 5 strain that persistently infects the commonly used AGS human cell line without causing cytopathology. This virus is most closely related to human strains, indicating that it may have originated from biopsy material or from laboratory contamination during generation of the cell line.

Spatiotemporal Reconstruction of the Introduction of Hepatitis C Virus into Scotland and Its Subsequent Regional Transmission.

UNLABELLED: A more comprehensive understanding of hepatitis C virus (HCV) transmission dynamics could facilitate public health initiatives to reduce the prevalence of HCV in people who inject drugs. We aimed to determine how HCV sequences entered and spread throughout Scotland and to identify transmission hot spots. A Scottish data set with embedded demographic data was created by sequencing the NS5B of 125 genotype 1a (Gt1a) samples and 166 Gt3a samples and analyzed alongside sequences from public databases. Applying Bayesian inference methods, we reconstructed the global origin and local spatiotemporal dissemination of HCV in Scotland. Scottish sequences mainly formed discrete clusters interspersed between sequences from the rest of the world; the most recent common ancestors of these clusters dated to 1942 to 1952 (Gt1a) and 1926 to 1942 (Gt3a), coincident with global diversification and distribution. Extant Scottish sequences originated in Edinburgh (Gt1a) and Glasgow (Gt3a) in the 1970s, but both genotypes spread from Glasgow to other regions. The dominant Gt1a strain differed between Edinburgh (cluster 2 [C2]), Glasgow (C3), and Aberdeen (C4), whereas significant Gt3a strain specificity occurred only in Aberdeen. Specific clusters initially formed separate transmission zones in Glasgow that subsequently overlapped, occasioning city-wide cocirculation. Transmission hot spots were detected with 45% of samples from patients residing in just 9 of Glasgow's 57 postcode districts. HCV was introduced into Scotland in the 1940s, concomitant with its worldwide dispersal likely arising from global-scale historical events. Cluster-specific transmission hubs were identified in Glasgow, the key Scottish city implicated in HCV dissemination. This fine-scale spatiotemporal reconstruction improves understanding of HCV transmission dynamics in Scotland. IMPORTANCE: HCV is a major health burden and the leading cause of hepatocellular carcinoma. Public health needle exchange and "treatment as prevention" strategies targeting HCV are designed to reduce prevalence of the virus in people who inject drugs (PWID), potentially mitigating the future burden of HCV-associated liver disease. Understanding HCV transmission dynamics could increase the effectiveness of such public health initiatives by identifying and targeting regions playing a central role in virus dispersal. In this study, we examined HCV transmission in Scotland by analyzing the genetic relatedness of strains from PWID alongside data inferring the year individuals became infected and residential information at a geographically finer-scale resolution than in previous studies. Clusters of Scotland-specific strains were identified with regional specificity, and mapping the spread of HCV allowed the identification of key areas central to HCV transmission in Scotland. This research provides a basis for identifying HCV transmission hot spots.