Can a microRNA be made outside the canonical nuclear pathway?
Canonical microRNA maturation is compartmentalised. A primary transcript is cropped in the nucleus by the Drosha and DGCR8 microprocessor, exported by Exportin-5, and cleaved a second time by Dicer in the cytoplasm. That arrangement was taken to explain why virus-encoded microRNAs had been described almost exclusively in nuclear DNA viruses. Two further arguments were advanced against RNA viruses making microRNAs at all. A hairpin embedded in an RNA genome would be excised by an RNase III enzyme and the genome fragmented, and the genome would present a perfect complement to the microRNA it produced and so be silenced by its own product. Varble 2010 states both arguments as the prevailing explanation the study set out to test.
Varble 2010 removed the assumed prohibition for a nuclear RNA virus. Splitting the overlapping NS1 and NEP/NS2 reading frames of influenza A virus segment 8 created an extended intron, and the murine miR-124-2 locus placed there yielded mature miR-124 within four hours of infection at levels comparable to abundant endogenous microRNAs. Production was orientation dependent, required NEP/NS2 splicing, and required Dicer, so the hairpin reached Drosha in the excised lariat rather than in the genome. Multicycle growth matched wild type. Target sites placed on the genomic strand were not silenced while identical sites in messenger RNA were, and the authors attribute that to nuclear ribonucleoprotein organisation, which the data are consistent with but do not establish.
Shapiro 2010 in RNA then moved the question into a compartment where the nuclear machinery is not available. The pri-miR-124-2 locus carried in an extra subgenomic transcript of Sindbis virus, an alphavirus that replicates only in the cytoplasm, produced abundant precursor and mature miR-124. Genetic dissection in knockout fibroblasts placed the activity downstream of Dicer with no requirement for DGCR8, for Exportin-5, or for type I interferon signalling, and the product silenced a target-bearing reporter and attenuated the virus by roughly two logs at low multiplicity in a Dicer-dependent manner. The authors named these products virtrons and stated plainly that the responsible processing step was undefined.
Langlois 2012 addressed a specific published objection, that cytoplasmic processing might be an artifact of rapidly dividing transformed cells whose nuclear envelope breakdown grants viral transcripts access to the microprocessor. Vesicular stomatitis virus, cytoplasmic and of negative polarity, produced Dicer-dependent miR-124 at an estimated 25,000 to 35,000 copies per cell, loaded it into Argonaute 2, repressed reporters by roughly 60 to 90 percent, and delivered the microRNA to lung, spleen, liver, kidney and heart of mice, with reduced induction of the target transcript Ptbp1. Star strand accumulation reached as much as 40 percent of reads and repressed its own reporter, which the paper presents as a genuine off-target liability of these vectors rather than a hypothetical one.
The most important development in this theme is the laboratory revising its own reading. Shapiro 2010 had reported cytoplasmic processing as microprocessor independent. Shapiro 2012 in RNA, using conditional deletion rather than inference, found an absolute requirement for Drosha, with Dicer needed only at the precursor to mature step and the DGCR8 phenotype confined to mature accumulation and left explicitly unresolved between two incompatible explanations. Serum starvation that fully blocked division did not change processing efficiency, and in situ hybridisation placed the viral transcript in the cytoplasm. Endogenous Drosha itself redistributed from nucleus to cytoplasm after infection with either parental or microRNA-expressing Sindbis virus, so relocalisation is a response to infection rather than to the presence of a substrate. The 2012 record describes this as revising the earlier reading, and that correction is what converts an operational observation into a defined genetic requirement. It is also the observation that opens the RNase III effector theme, since a nuclear enzyme found in the cytoplasm during infection invites the question of what else it is doing there.
Morales 2017 sits at the edge of this theme and was led by the Enjuanes and Sola group in Madrid, with the tenOever contribution recorded as reagents, conceptual advice and manuscript writing. It reports three discrete small viral RNAs from SARS-CoV in infected mouse lung, produced in cells lacking both Drosha and Dicer, which places their biogenesis outside canonical processing without identifying the route. Candidate mechanisms offered there, including an Argonaute 2-dependent route or cleavage by the viral endoribonuclease nsp15, are untested.
The mechanism of cytoplasmic pri-microRNA cleavage was never resolved in this corpus. Every paper in the theme says so. What the line delivered instead was a set of vectors and a genetic framework, and the redirection of attention onto Drosha localisation.
Langlois 2012 and Varble 2010 are also assigned to the programmable virology area under RNA virus vectors for delivery, where the same constructs are treated as delivery tools rather than as evidence about biogenesis. Morales 2017 is also assigned to influenza genome regulation under small viral RNAs.
The Drosha relocalisation reported in Shapiro 2012 is the direct antecedent of Shapiro 2014 and Aguado 2017 in the RNase III effector theme. The self-attenuation observed in Shapiro 2010, and the failure of genomically oriented target sites to silence influenza A virus in Varble 2010, both feed the question of whether small RNA activity restricts viruses in mammalian cells, taken up in the theme on whether mammalian antiviral RNA interference exists. The vectors themselves become instruments in programmable virology.
SARS-CoV generates three discrete small viral RNAs during infection of mouse lung, and blocking the one derived from the nucleocapsid gene with a locked nucleic acid antagomir reduces pulmonary inflammation, tissue damage and proinflammatory cytokine expression without measurably lowering lung virus titres.
Primary microRNA transcripts generated in the cytoplasm by a recombinant Sindbis virus are cleaved without any nuclear involvement yet still require Drosha, which relocalises from nucleus to cytoplasm on infection while the endogenous microRNA profile of the cell remains largely unchanged.
A negative-sense cytoplasmic RNA virus, vesicular stomatitis virus, can be engineered to produce mature Dicer-dependent miR-124 that loads into Argonaute 2, silences targets, reaches many tissues in mice, and persists after the vector itself is cleared.
Influenza A virus can be engineered to encode a cellular pri-microRNA inside an artificial intron of segment 8 and to produce mature, silencing-competent miR-124 during infection without measurable loss of replication or genome stability.
Insertion of a primary microRNA locus into the exclusively cytoplasmic Sindbis virus genome yields mature, functional miR-124 through a Dicer-dependent but microprocessor- and Exportin-5-independent route, defining a cytoplasmic hairpin-processing activity in vertebrate cells that the authors term a virtron.