Autophagy plays a paramount role in mammalian antiviral immunity including direct targeting of viruses and their individual components, and many viruses have evolved measures to antagonize or even exploit autophagy mechanisms for the benefit of infection. In plants, however, the functions of autophagy in host immunity and viral pathogenesis are poorly understood. In this study, we have identified both anti-and proviral roles of autophagy in the compatible interaction of cauliflower mosaic virus (CaMV), a double-stranded DNA pararetrovirus, with the model plant Arabidopsis thaliana. We show that the autophagy cargo receptor NEIGHBOR OF BRCA1 (NBR1) targets nonassembled and virus particle-forming capsid proteins to mediate their autophagy-dependent degradation, thereby restricting the establishment of CaMV infection. Intriguingly, the CaMV-induced virus factory inclusions seem to protect against autophagic destruction by sequestering capsid proteins and coordinating particle assembly and storage. In addition, we found that virus-triggered autophagy prevents extensive senescence and tissue death of infected plants in a largely NBR1-independent manner. This survival function significantly extends the timespan of virus production, thereby increasing the chances for virus particle acquisition by aphid vectors and CaMV transmission. Together, our results provide evidence for the integration of selective autophagy into plant immunity against viruses and reveal potential viral strategies to evade and adapt autophagic processes for successful pathogenesis. A utophagy is a conserved intracellular pathway that engages specialized double-membrane vesicles, called "autophagosomes," to enclose and transport cytoplasmic content to lytic compartments for degradation and subsequent recycling (1). Autophagosome formation relies on extensive membrane rearrangements and is mediated by the concerted action of a core set of autophagy-related proteins (ATGs) (2, 3). At basal levels, autophagy serves mainly housekeeping functions in cellular homeostasis, whereas stimulated autophagy activity facilitates adaptation to developmental and environmental stress conditions including starvation, aging, and pathogen infection (1, 4). Ample evidence now indicates that autophagy, initially recognized as a mainly bulk catabolic process, is able specifically to target and degrade a multitude of cellular structures ranging from individual and aggregated proteins to entire organelles and invading microbes (5, 6). Selectivity is provided by a growing number of autophagic adaptor or receptor proteins identified in eukaryotic organisms that recruit the cargo to the developing autophagosome through interaction with membrane-associated ATG8/LC3 proteins (7, 8). Several mammalian autophagy receptors have been implicated in the targeting of intracellular bacterial and viral pathogens in a process called "xenophagy" (8-10). For instance, the cargo receptor p62 (SQSTM1) was shown to bind directly to and mediate autophagic clearance of different viral capsid pr...

Turnip mosaic virus (TuMV, family Potyviridae ) and cauliflower mosaic virus (CaMV, family Caulimoviridae ) are transmitted by aphid vectors. They are the only viruses shown so far to undergo transmission activation (TA) immediately preceding plant-to-plant propagation. TA is a recently described phenomenon where viruses respond to the presence of vectors on the host by rapidly and transiently forming transmissible complexes that are efficiently acquired and transmitted. Very little is known about the mechanisms of TA and on whether such mechanisms are alike or distinct in different viral species. We use here a pharmacological approach to initiate the comparison of TA of TuMV and CaMV. Our results show that both viruses rely on calcium signaling and reactive oxygen species (ROS) for TA. However, whereas application of the thiol-reactive compound N-ethylmaleimide (NEM) inhibited, as previously shown, TuMV transmission it did not alter CaMV transmission. On the other hand, sodium azide, which boosts CaMV transmission, strongly inhibited TuMV transmission. Finally, wounding stress inhibited CaMV transmission and increased TuMV transmission. Taken together, the results suggest that transmission activation of TuMV and CaMV depends on initial calcium and ROS signaling that are generated during the plant’s immediate responses to aphid manifestation. Interestingly, downstream events in TA of each virus appear to diverge, as shown by the differential effects of NEM, azide and wounding on TuMV and CaMV transmission, suggesting that these two viruses have evolved analogous TA mechanisms.

Host-to-host transmission—a key step in plant virus infection cycles—is ensured predominantly by vectors, especially aphids and related insects. A deeper understanding of the mechanisms of virus acquisition, which is critical to vector-transmission, might help to design future virus control strategies, because any newly discovered molecular or cellular process is a potential target for hampering viral spread within host populations. With this aim in mind, an aphid membrane-feeding assay was developed where aphids transmitted two non-circulative viruses [cauliflower mosaic virus (CaMV) and turnip mosaic virus] from infected protoplasts. In this assay, virus acquisition occurs exclusively from living cells. Most interestingly, we also show that CaMV is less efficiently transmitted by aphids in the presence of oryzalin—a microtubule-depolymerising drug. The example presented here demonstrates that our technically simple “virus-acquisition phenotyping assay” (VAPA) provides a first opportunity to implement correlative studies relating the physiological state of infected plant cells to vector-transmission efficiency.

The split GFP technique is based on the auto-assembly of GFP when two polypeptides–GFP1-10 (residues 1–214; the detector) and GFP11 (residues 215–230; the tag)–both non-fluorescing on their own, associate spontaneously to form a fluorescent molecule. We evaluated this technique for its efficacy in contributing to the characterization of Cauliflower mosaic virus (CaMV) infection. A recombinant CaMV with GFP11 fused to the viral protein P6 (a key player in CaMV infection and major constituent of viral factory inclusions that arise during infection) was constructed and used to inoculate transgenic Arabidopsis thaliana expressing GFP1-10. The mutant virus (CaMV 11P6 ) was infectious, aphid-transmissible and the insertion was stable over many passages. Symptoms on infected plants were delayed and milder. Viral protein accumulation, especially of recombinant 11P6, was greatly decreased, impeding its detection early in infection. Nonetheless, spread of infection from the inoculated leaf to other leaves was followed by whole plant imaging. Infected cells displayed in real time confocal laser scanning microscopy fluorescence in wild type-looking virus factories. Thus, it allowed for the first time to track a CaMV protein in vivo in the context of an authentic infection. 11P6 was immunoprecipitated with anti-GFP nanobodies, presenting a new application for the split GFP system in protein-protein interaction assays and proteomics. Taken together, split GFP can be an attractive alternative to using the entire GFP for protein tagging.