Illuminating viral replication in living plants: A conditional probe system for tracking replicase dynamics

Plant virus diseases pose a major threat to global agriculture, significantly reducing crop yield and quality across a wide range of plant species. Among plant viruses, positive-sense single-stranded RNA ((+)ssRNA) viruses are particularly problematic due to their rapid evolution driven by high replication rates. These viruses replicate via virus-encoded RNA-dependent RNA polymerase (also called replicase). The replicase, along with other viral proteins, is known to co-opt host proteins and intracellular membranes to make virus replication compartments (VRCs). These VRCs also protect viral double-stranded RNA (dsRNA) intermediates from cellular defenses, such as RNA interference (Miller and Krijnse-Locker 2008). Understanding viral replicase and VRC architecture is therefore crucial for uncovering how viruses manipulate host cells and for developing environmentally friendly virus control strategies.

Directly labeling viral replicase during infection remains technically challenging. Fluorescent proteins are large, and their fusion interferes with replicase localization and activity. Additionally, engineering a viral genome to encode a replicase-fluorescent protein fusion often reduces replication efficiency. Existing approaches, such as immunostaining or the use of dsRNA-binding proteins, offer only snapshots or indirect localization and rarely function effectively in living tissues (Bamunusinghe et al. 2009; Monsion et al. 2018).

In this issue of Plant Physiology, Ishihara et al. (2025) describe an elegant solution to circumvent the limitations of direct labeling. Using plantago asiatica mosaic virus (PlAMV) as model (+)ssRNA virus that forms VRCs in Nicotiana benthamiana, the authors devised a strategy to label viral replicase using a genetically encoded probe known as a frankenbody (FB). Originally developed in mammalian systems, FB is a single-chain variable fragment, the minimal binding unit of an antibody, that specifically targets a widely used 9-amino acid tag derived from influenza hemagglutinin (called an HA epitope). Fusing FB to a fluorescent protein (FB-mCherry) offered a promising tool for visualizing viral replicase dynamics in live plant tissues.

To verify that the system accurately reported replicase localization without interfering with viral function, Ishihara et al. (2025) tested FB-mCherry in cells with and without HA-tagged proteins targeted to the ER. While some nonspecific signal was observed, FB-mCherry localized to the ER in the presence of the HA epitope and did not disturb ER morphology. However, expressing FB-mCherry under a strong constitutive promoter produced diffuse background fluorescence. To solve this, the authors designed a virus-dependent expression system: they inserted the synthetic transcription factor XV into the PlAMV genome and placed the FB-mCherry gene under the control of an XV-responsive LexA promoter. This design ensured that FB-mCherry expression, and thus replicase labeling, occurred only in virus-infected cells, allowing live-cell imaging of replicase dynamics without background interference (Figure).

Figure.

Schematic representation of the inducible FB system for live-cell imaging of viral replicase in Nicotiana benthamiana. A 2-component system was delivered via Agrobacterium infiltration into N. benthamiana leaves. The first construct encodes an HA-tagged PlAMV replicase under the 35S promoter, along with movement proteins (TGBp1–3), the synthetic transcription factor XV, and coat protein (CP), separated by self-cleaving 2A peptides. The second construct expresses FB-mCherry under the control of the XV-responsive LexA promoter (XVpro). Upon infection, the virus replicates and expresses both HA-tagged replicase and XV. XV activates the XVpro promoter, inducing FB-mCherry expression only in infected cells. FB-mCherry binds the HA-tag fused to the replicase, enabling its visualization within VRCs. This system allows live-cell imaging of functional viral replicase with minimal perturbation to replication or cellular organization. This figure was created with BioRender.

Building on this validation, the authors used the reporter system to investigate replicase dynamics in the context of viral replication. The system revealed that the PlAMV replicase localizes to discrete granules associated with the ER and plasmodesmata during infection, patterns consistent with previously described sites of viral replication (Tilsner et al. 2013). These granules observed at the cell periphery and along the transverse cell wall likely represent active viral replication sites, as suggested by their colocalization with B2:GFP-labeled dsRNA clusters, a hallmark of active viral replication (Monsion et al. 2018). These granule-specific signals were absent when a replicase lacking the HA tag was used, confirming system specificity. Virus accumulation was unaffected by FB-mCherry expression, indicating that the system faithfully reports replicase localization without impairing viral replication.

Like any system, this approach has limitations. Not all viral genomes will tolerate the insertion of the XV and HA coding sequences, especially those that depend on specific RNA structures for replication and translation, such as poleroviruses and luteoviruses. Additionally, while the viral replicon can move systemically within the plant, the engineered FB-mCherry construct remains confined to the infiltrated area, limiting tracking of replication in systemic tissues.

This study addresses a major technical challenge in plant virology: visualizing viral replicase dynamics in living tissues without disrupting replication. Unlike earlier GFP-binding systems, which altered target protein behavior (Schornack et al. 2009), the small size and specificity of the HA tag enabled minimal perturbation. While signal intensity may be improved by brighter fluorophores or multiple HA tags (Zhao et al. 2019), this system marks a significant step forward. With further refinement, this approach has the potential to advance our understanding of virus replication, movement, and host protein dynamics in plants.

Funding

S.S. is supported by funding from the HORIZON EUROPE Marie Skłodowska-Curie Actions (MSCA) from the European Commission.

Data availability

The data underlying this article are available in the article.