A big role for small RNAs in influenza virus replication

Viruses have one mission: to make more viruses. To achieve this, they use diverse replication strategies, all of which rely on the careful orchestration of viral and host cell components. Although the outlines of these replication schemes have historically formed the basis for viral classification, the precise mechanisms by which viruses propagate their genomes are still being unraveled. In PNAS, the study by Perez et al. (1) identifies virus-encoded, short RNAs (svRNA) as a component of the influenza A viral replicase and an essential prerequisite for viral RNA synthesis. Their exciting discovery adds an additional dimension to the current paradigm regarding how influenza A viruses mediate the switch between transcription and replication.

The influenza A virus genome is comprised of eight negative-sense, single-stranded RNA segments (vRNA), which serve as templates for the transcription of viral mRNAs and full-length cRNAs; cRNAs are then replicated to produce more vRNA (Fig. 1). Notably, influenza virus mRNA and cRNA/vRNA synthesis differ mechanistically. mRNA transcription is primed by capped RNA segments snatched from host cell pre-mRNAs by the viral polymerase, whereas cRNA/vRNA synthesis is primer-independent. mRNA termination occurs at a stretch of uridines upstream of the 5′ noncoding region (NCR), whereas cRNA/vRNA synthesis terminates without a poly(A) tail (2). Paradoxically, although the interaction of the 3′ and 5′ ends of the template to create a panhandle structure is essential for both mRNA and cRNA/vRNA synthesis, the steric hindrance of this secondary structure is incompatible with transcription of a full-length cRNA copy (3). Finally, the timing of mRNA and cRNA/vRNA accumulation seems to differ, as mRNA production and de novo protein synthesis precede genome replication (4). These findings pose a question that has been the topic of intense investigation for decades: what mediates the switch?

Not surprisingly, a number of hypotheses have emerged regarding the factors that control viral transcription and replication in vivo. Several studies have implicated the nucleoprotein (NP), which encapsidates the viral RNA, as the responsible party (5, 6). Based on the RNA-binding capacity of NP, some have proposed that the protein promotes cRNA/vRNA synthesis by altering the panhandle structure (7–9). In addition, biochemical studies have suggested that NP can associate directly with the polymerase, possibly preventing its cap snatching activity and promoting unprimed transcription (10–12). Other models suggest that the polymerase itself dictates the switch, perhaps through the actions of individual subunits, including polymerase acidic protein (PA) and polymerase basic 2 protein (PB2) (13, 14), or by operating in cis to transcribe mRNA and in trans to produce cRNA (15). Binding of NP and the polymerase has also been proposed to regulate the stability of cRNA, leading to its accumulation only later in infection (16). More recently, the viral nuclear export protein (NEP/NS2) was shown to promote accumulation of cRNA/vRNA, indicating that multiple viral components are likely involved in replication control (17). Finally, the host cell environment has been speculated to contribute, because intracellular levels of nucleotides have been shown to impact vRNA synthesis (18).

The report by Perez et al. (1) poses an exciting alternative hypothesis: that an RNA molecule could reconcile the panhandle problem and regulate the transition between transcription and replication. Short, noncoding RNAs (18–30 nucleotides in length) are most well known for their role in fine tuning the host cell transcriptome through posttranscriptional gene silencing, although recent work suggests that they may be a fairly ubiquitous feature of viral infection—especially prevalent among the herpes viruses and other DNA viruses that replicate in the nucleus (19, 20). Previous studies have postulated a role for viral or host short RNAs in the regulation of viral transcription and replication (21, 22). Perez et al. (1) took advantage of deep sequencing technology to search for candidate RNA molecules in influenza virus-infected cells. Sure enough, the authors identified influenza virus-specific short RNAs that aligned predominantly to the 5′ end of each of the eight segments of the viral genome (1). The enrichment of these svRNAs at certain locations suggested they were not simply genomic breakdown products but likely served a specific function. Could influenza virus-encoded svRNAs be the elusive “switch”?

Several lines of evidence put svRNAs at the center of the action. Kinetic studies showed that viral protein expression occurs before the generation of detectable levels of svRNAs. Intriguingly, these short transcripts emerge concurrent with the onset of viral replication. Furthermore, coimmunoprecipitation of an svRNA mimic and the complete polymerase complex from cotransfected cells gave evidence for a physical interaction. These results support a role for svRNAs in regulating the activity of the replicase. More detailed analysis revealed that levels of vRNA, but not cRNA, were significantly reduced when the HA segment-specific svRNA was inhibited by a locked nucleic acid (LNA) antisense analog. Synthesizing these data with published work, the authors propose a model in which cRNA synthesis from input genomes occurs in the absence of svRNA, potentially facilitated by NP-mediated blocking of the panhandle secondary structure. svRNAs, which are likely synthesized from the cRNA transcript, then interact directly with the viral polymerase complex to orchestrate a second switch that promotes the replication of cRNA to vRNA (Fig. 1). Interestingly, the specificity of svRNAs for the 5′ end of a particular genomic segment suggests the existence of eight custom replicases.

Loss of replication and progeny virion production after depletion of svRNAs underscores the indispensable nature of these small transcripts for productive influenza virus infection. However, the synthesis of influenza virus-derived short RNAs and their precise mechanism of action remain a mystery. Does their association with the polymerase complex effect a change on its conformation or functional requirements? Could svRNAs, being complementary to the 3′ end of cRNA, negate the requirement for the panhandle structure in vRNA synthesis? Could these svRNAs, which are of similar size to classic cellular microRNAs, act to modulate host gene expression to facilitate viral replication or inhibit an antiviral response? Conversely, could a portion of svRNAs be generated by the host RNAi machinery and/or incorporated into host RNA-induced silencing complexes (RISC) that then target the virus as an antiviral defense mechanism (23)? Future work will no doubt address these questions as well as ask if small RNAs play a big role in the replication strategies of other viral pathogens.