Why does each branch of life defend itself the way it does?
Small RNA silencing is the principal antiviral defence in plants, nematodes and arthropods. Vertebrates retain most of the machinery and use it for microRNA regulation instead, defending against viruses with pattern recognition receptors, interferons and, in jawed vertebrates, recombinatorial adaptive immunity. If the vertebrate substitution was not forced by an inability of mammalian cells to support a small RNA defence, which is what Benitez 2015 argues, then some other explanation is required. The problem is that evolutionary questions of this kind admit no direct experiment, and both reviews in this theme say so.
tenOever 2013 in Nature Reviews Microbiology, a single-author review, states the case that chordate microRNAs are not antiviral and does so quantitatively, through the three constraints of copy number, silencing capacity and kinetics discussed elsewhere in this area. Its evolutionary contribution is the corollary rather than the constraints themselves. Because chordate RNA viruses experience no selective pressure to disrupt the host microRNA machinery, they leave it intact and available, which is the premise of the engineering line. The review notes that the two systems appear mutually exclusive and states that why this should be so remains a mystery, recording as supporting circumstance that small interfering RNAs can themselves induce type I interferon and that the features marking RNA as foreign are also products of the RNA-dependent RNA polymerase activity that amplifies virus-derived interfering RNAs in other phyla. A boxed speculation, that ubiquitous microRNAs might act as sequence restriction elements rather than as regulators with physiological targets, is labelled a hypothesis with a proposed test.
tenOever 2016 in Cell Host and Microbe is the fuller statement, and it is a single-author Perspective in which the great majority of the science discussed belongs to other groups, above all the comparative genomics of virus and defence system evolution developed by Koonin and colleagues. Its structural argument is that antiviral defence has repeatedly been assembled from a short parts list, a guide that recognises foreign sequence combined with an effector, extended in multicellular organisms by a transcriptional response and then by a secreted signal that protects cells not yet infected. Antisense RNA, restriction and modification, prokaryotic Argonaute, CRISPR-Cas, the piRNA pathway and RNA interference are read as successive combinations of the same two elements, with piRNA and CRISPR described as analogous rather than homologous. The framework is evolutionary parsimony, which the article states is probabilistic and not universally accepted, and it calls the proposed trajectory for each system necessarily speculative.
The article's own proposal, which it labels an attractive hypothesis, is that chordates lost RNA interference through incompatibility with interferon rather than through simple redundancy. Three premises are linked. Systemic small RNA defence in a large organism would require amplification and circulation of small interfering RNAs by an RNA-dependent RNA polymerase. Receptor-mediated virus entry in chordates does not carry small RNAs along with it as plant cell-to-cell movement does. And expressing such a polymerase in mammalian somatic cells itself triggers innate immunity, shown independently by Painter and colleagues and by Yu and colleagues. Correlative support for mutual exclusivity is drawn from other laboratories, on the divergent handling of long double-stranded RNA by stem and somatic cells and on reciprocal inhibition between RISC and interferon signalling. Two of the laboratory's own papers carry weight in the argument, Benitez 2015 and Aguado 2015. The article states that the evolutionary cause of the transformation cannot be determined.
Two experimental papers supply cross-kingdom evidence rather than evolutionary inference. Aguado 2017 showed that RNase III proteins from bacteria, archaea, yeast and the urochordate Ciona intestinalis all confer antiviral activity against positive-strand RNA viruses when expressed in cells lacking Drosha and Dicer, which makes structured RNA recognition by an ancient nuclease fold a plausible defensive capability predating silencing pathways. The narrative in that paper's discussion, in which an ancestral RNase III clamp preceded RNA interference and later gave rise to interferon sensors, is labelled speculation by its authors.
Oishi 2023 is a different kind of cross-kingdom result. Screening codon-optimised archaeal RNA binding proteins identified the kink-turn binding protein L7Ae, which eliminates the spliced influenza M2 and NS2 products while M1 and NS1 accumulate, requires nuclear localisation, and has no effect on a recombinant virus that makes both products through 2A peptides rather than by splicing. Only archaeal orthologues of the L30 family had the activity, with the archaea-defining residues contributing most, and substitutions abolishing canonical box C and D binding left inhibition intact. The relevance here is that a protein from a domain of life with divergent RNA biology recognises a feature of a vertebrate virus that the host machinery treats differently, which is the RNase III logic approached from the other side. The proposed kink-turn-like substrate is inferred and not observed, crosslinking produced no footprint, and the paper carries a declared conflict of interest, since tenOever is a co-founder of Archean Biologics and an author on a patent covering commercialisation of L7Ae.
The framing moved from a negative to a mechanism. In 2013 the incompatibility of interferon and silencing is named a mystery. In 2016 it is given a candidate molecular obstacle, the innate immune consequence of expressing an RNA-dependent RNA polymerase, and that obstacle is addressable because engineering around the polymerase requirement restores protection. Uhl 2023, in the reconstruction theme, later supplied a second and independent incompatibility at the level of a single interferon-associated enzyme, since ADAR1 editing destroys the sequence fidelity that perfect complementarity requires, and ADAR1 orthologues in that limited phylogenetic sampling track with interferon-like defences. Neither paper claims the two arguments together, and both are correlative, but read side by side they make the vertebrate loss look overdetermined rather than explained.
tenOever 2016 is also assigned to the innate immune signalling area under interferon and cell identity. Oishi 2023 is also assigned to influenza genome regulation under splicing and temporal control of segment 8, which is where its primary subject matter sits. Aguado 2017 belongs principally to the RNase III effector theme.
This theme supplies the reason the experimental work in the area is interesting rather than merely technical. The negative result of Backes 2014 is what it sets out to explain, the possibility claim of Benitez 2015 is what constrains the explanation, and the ADAR1 result of Uhl 2023 adds a second mechanism. The comparative framing also organises the recombination argument of Aguado 2018, which asks whether host defences bias which kinds of viruses flourish.
Archaeal members of the L30 protein family, but not their orthologues from any other domain of life, block production of the spliced influenza A virus, influenza B virus and isavirus transcripts without measurably disturbing host splicing, identifying a shared and apparently noncanonical feature of orthomyxovirus splicing.
RNase III nucleases, including human Drosha and homologues from bacteria, archaea, yeast and a urochordate, restrict positive-strand RNA viruses by recognizing unbranched RNA stem loops and impairing the viral polymerase, separably from microRNA biogenesis, catalysis and interferon.
A synthesis arguing that antiviral defenses across the three domains of life reuse a small set of designs, antisense recognition joined to nuclease activity and later to transcriptional and secreted responses, and proposing that chordates lost RNA interference through incompatibility with interferon.
Argues that chordates, unlike plants, nematodes and arthropods, do not use small RNAs as antiviral defence, and that this absence of interplay between host microRNAs and RNA viruses is what leaves the microRNA machinery available for engineering viral tropism and small RNA delivery.