A self-interacting carmovirus movement protein plays a role in binding of viral RNA during the cell-to-cell movement and shows an actin cytoskeleton dependent location in cell periphery

Small hydrophobic viral proteins involved in intercellular movement of diverse plant virus genomes

Most plant viruses code for movement proteins (MPs) targeting plasmodesmata to enable cell-to-cell and systemic spread in infected plants. Small membrane-embedded MPs have been first identified in two viral transport gene modules, triple gene block (TGB) coding for an RNA-binding helicase TGB1 and two small hydrophobic proteins TGB2 and TGB3 and double gene block (DGB) encoding two small polypeptides representing an RNA-binding protein and a membrane protein. These findings indicated that movement gene modules composed of two or more cistrons may encode the nucleic acid-binding protein and at least one membrane-bound movement protein. The same rule was revealed for small DNA-containing plant viruses, namely, viruses belonging to genus Mastrevirus (family Geminiviridae) and the family Nanoviridae. In multi-component transport modules the nucleic acid-binding MP can be viral capsid protein(s), as in RNA-containing viruses of the families Closteroviridae and Potyviridae. However, membrane proteins are always found among MPs of these multicomponent viral transport systems. Moreover, it was found that small membrane MPs encoded by many viruses can be involved in coupling viral replication and cell-to-cell movement. Currently, the studies of evolutionary origin and functioning of small membrane MPs is regarded as an important pre-requisite for understanding of the evolution of the existing plant virus transport systems. This paper represents the first comprehensive review which describes the whole diversity of small membrane MPs and presents the current views on their role in plant virus movement.

A conserved motif in three viral movement proteins from different genera is required for host factor recruitment and cell-to-cell movement

Due to their minimal genomes, plant viruses are forced to hijack specific cellular pathways to ensure host colonization, a condition that most frequently involves physical interaction between viral and host proteins. Among putative viral interactors are the movement proteins, responsible for plasmodesma gating and genome binding during viral transport. Two of them, DGBp1 and DGBp2, are required for alpha-, beta- and gammacarmovirus cell-to-cell movement, but the number of DGBp-host interactors identified at present is limited. By using two different approaches, yeast two-hybrid and bimolecular fluorescence complementation assays, we found three Arabidopsis factors, eIF3g1, RPP3A and WRKY36, interacting with DGBp1s from each genus mentioned above. eIF3g1 and RPP3A are mainly involved in protein translation initiation and elongation phases, respectively, while WRKY36 belongs to WRKY transcription factor family, important regulators of many defence responses. These host proteins are not expected to be associated with viral movement, but knocking out WRKY36 or silencing either RPP3A or eIF3g1 negatively affected Arabidopsis infection by Turnip crinkle virus. A highly conserved FNF motif at DGBp1 C-terminus was required for protein-protein interaction and cell-to-cell movement, suggesting an important biological role.

Key checkpoints in the movement of plant viruses through the host

Plant viruses cannot exploit any of the membrane fusion-based routes of entry described for animal viruses. In addition, one of the distinctive structures of plant cells, the cell wall, acts as the first barrier against the invasion of pathogens. To overcome the rigidity of the cell wall, plant viruses normally take advantage of the way of life of different biological vectors. Alternatively, the physical damage caused by environmental stresses can facilitate virus entry. Once inside the cell and taking advantage of the characteristic symplastic continuity of plant cells, viruses need to remodel and/or modify the restricted pore size of the plasmodesmata (channels that connect plant cells). In a successful interaction for the virus, it can reach the vascular tissue to systematically invade the plant. The connections between the different cell types in this path are not designed to allow the passage of molecules with the complexity of viruses. During this process, viruses face different cell barriers that must be overcome to reach the distal parts of the plant. In this review, we highlight the current knowledge about how plant RNA viruses enter plant cells, move between them to reach vascular cells and overcome the different physical and cellular barriers that the phloem imposes. Finally, we update the current research on cellular organelles as key regulator checkpoints in the long-distance movement of plant viruses.

Dissecting the Subcellular Localization, Intracellular Trafficking, Interactions, Membrane Association, and Topology of Citrus Leprosis Virus C Proteins

Citrus leprosis (CL) is a re-emergent viral disease affecting citrus crops in the Americas, and citrus leprosis virus C (CiLV-C), belonging to the genus Cilevirus, is the main pathogen responsible for the disease. Despite the economic importance of CL to the citrus industry, very little is known about the performance of viral proteins. Here, we present a robust in vivo study around functionality of p29, p15, p61, MP, and p24 CiLV-C proteins in the host cells. The intracellular sub-localization of all those viral proteins in plant cells are shown, and their co-localization with the endoplasmic reticulum (ER), Golgi complex (GC) (p15, MP, p61 and p24), actin filaments (p29, p15 and p24), nucleus (p15), and plasmodesmata (MP) are described. Several features are disclosed, including i) ER remodeling and redistribution of GC apparatus, ii) trafficking of the p29 and MP along the ER network system, iii) self-interaction of the p29, p15, and p24 and hetero-association between p29-p15, p29-MP, p29-p24, and p15-MP proteins in vivo. We also showed that all proteins are associated with biological membranes; whilst p15 is peripherally associated, p29, p24, and MP are integrally bound to cell membranes. Furthermore, while p24 exposes an N-cytoplasm-C-lumen topology, p29, and p15 are oriented toward the cytoplasmic face of the biological membrane. Based on our findings, we discuss the possible performance of each protein in the context of infection and a hypothetical model encompassing the virus spread and sites for replication and particle assembly is suggested.

Hexanoic Acid Treatment Prevents Systemic MNSV Movement in Cucumis melo Plants by Priming Callose Deposition Correlating SA and OPDA Accumulation

Unlike fungal and bacterial diseases, no direct method is available to control viral diseases. The use of resistance-inducing compounds can be an alternative strategy for plant viruses. Here we studied the basal response of melon to Melon necrotic spot virus (MNSV) and demonstrated the efficacy of hexanoic acid (Hx) priming, which prevents the virus from systemically spreading. We analysed callose deposition and the hormonal profile and gene expression at the whole plant level. This allowed us to determine hormonal homeostasis in the melon roots, cotyledons, hypocotyls, stems and leaves involved in basal and hexanoic acid-induced resistance (Hx-IR) to MNSV. Our data indicate important roles of salicylic acid (SA), 12-oxo-phytodienoic acid (OPDA), jasmonic-isoleucine, and ferulic acid in both responses to MNSV. The hormonal and metabolites balance, depending on the time and location associated with basal and Hx-IR, demonstrated the reprogramming of plant metabolism in MNSV-inoculated plants. The treatment with both SA and OPDA prior to virus infection significantly reduced MNSV systemic movement by inducing callose deposition. This demonstrates their relevance in Hx-IR against MNSV and a high correlation with callose deposition. Our data also provide valuable evidence to unravel priming mechanisms by natural compounds.

Self-interaction of Citrus tristeza virus p33 protein via N-terminal helix

Citrus tristeza virus (CTV), the most economically important viral pathogen of citrus, encodes a unique protein, p33. CTV p33 shows no similarity with other known proteins, yet plays an important role in viral pathogenesis: it extends the virus host range and mediates virus ability to exclude superinfection by other variants of the virus. Previously we demonstrated that p33 is an integral membrane protein and appears to share characteristics of viral movement proteins. In this study, we show that the p33 protein self-interacts in vitro and in vivo using co-immunoprecipitation, yeast two hybrid, and bimolecular fluorescence complementation assays. Furthermore, a helix located at the N-terminus of the protein is required and sufficient for the protein self-interaction.

An Update on the Intracellular and Intercellular Trafficking of Carmoviruses

Despite harboring the smallest genomes among plant RNA viruses, carmoviruses have emerged as an ideal model system for studying essential steps of the viral cycle including intracellular and intercellular trafficking. Two small movement proteins, formerly known as double gene block proteins (DGBp1 and DGBp2), have been involved in the movement throughout the plant of some members of carmovirus genera. DGBp1 RNA-binding capability was indispensable for cell-to-cell movement indicating that viral genomes must interact with DGBp1 to be transported. Further investigation on Melon necrotic spot virus (MNSV) DGBp1 subcellular localization and dynamics also supported this idea as this protein showed an actin-dependent movement along microfilaments and accumulated at the cellular periphery. Regarding DGBp2, subcellular localization studies showed that MNSV and Pelargonium flower break virus DGBp2s were inserted into the endoplasmic reticulum (ER) membrane but only MNSV DGBp2 trafficked to plasmodesmata (PD) via the Golgi apparatus through a COPII-dependent pathway. DGBp2 function is still unknown but its localization at PD was a requisite for an efficient cell-to-cell movement. It is also known that MNSV infection can induce a dramatic reorganization of mitochondria resulting in anomalous organelles containing viral RNAs. These putative viral factories were frequently found associated with the ER near the PD leading to the possibility that MNSV movement and replication could be spatially linked. Here, we update the current knowledge of the plant endomembrane system involvement in carmovirus intra- and intercellular movement and the tentative model proposed for MNSV transport within plant cells.

Melon necrotic spot virus Replication Occurs in Association with Altered Mitochondria

Melon necrotic spot virus (MNSV) (genus Carmovirus, family Tombusviridae) is a single-stranded, positive-sense RNA virus that has become an experimental model for the analysis of cell-to-cell virus movement and translation of uncapped viral RNAs, whereas little is known about its replication. Analysis of the cytopathology after MNSV infection showed the specific presence of modified organelles that resemble mitochondria. Immunolocalization of the glycine decarboxylase complex (GDC) P protein in these organelles confirmed their mitochondrial origin. In situ hybridization and immunolocalization experiments showed the specific localization of positive-sense viral RNA, capsid protein (CP), and double-stranded (ds)RNA in these organelles meaning that replication of the virus takes place in association with them. The three-dimensional reconstructions of the altered mitochondria showed the presence of large, interconnected, internal dilations which appeared to be linked to the outside cytoplasmic environment through pores and/or complex structures, and with lipid bodies. Transient expression of MNSV p29 revealed that its specific target is mitochondria. Our data document the extensive reorganization of host mitochondria induced by MNSV, which provides a protected environment to viral replication, and show that the MNSV p29 protein is the primary determinant of this effect in the host.

Comparative analysis among the small RNA populations of source, sink and conductive tissues in two different plant-virus pathosystems

Background

In plants, RNA silencing plays a fundamental role as defence mechanism against viruses. During last years deep-sequencing technology has allowed to analyze the sRNA profile of a large variety of virus-infected tissues. Nevertheless, the majority of these studies have been restricted to a unique tissue and no comparative analysis between phloem and source/sink tissues has been conducted. In the present work, we compared the sRNA populations of source, sink and conductive (phloem) tissues in two different plant virus pathosystems. We chose two cucurbit species infected with two viruses very different in genome organization and replication strategy; Melon necrotic spot virus (MNSV) and Prunus necrotic ringspot virus (PNRSV).

Results

Our findings showed, in both systems, an increase of the 21-nt total sRNAs together with a decrease of those with a size of 24-nt in all the infected tissues, except for the phloem where the ratio of 21/24-nt sRNA species remained constant. Comparing the vsRNAs, both PNRSV- and MNSV-infected plants share the same vsRNA size distribution in all the analyzed tissues. Similar accumulation levels of sense and antisense vsRNAs were observed in both systems except for roots that showed a prevalence of (+) vsRNAs in both pathosystems. Additionally, the presence of overrepresented discrete sites along the viral genome, hot spots, were identified and validated by stem-loop RT-PCR. Despite that in PNRSV-infected plants the presence of vsRNAs was scarce both viruses modulated the host sRNA profile.

Conclusions

We compare for the first time the sRNA profile of four different tissues, including source, sink and conductive (phloem) tissues, in two plant-virus pathosystems. Our results indicate that antiviral silencing machinery in melon and cucumber acts mainly through DCL4. Upon infection, the total sRNA pattern in phloem remains unchanged in contrast to the rest of the analyzed tissues indicating a certain tissue-tropism to this polulation. Independently of the accumulation level of the vsRNAs both viruses were able to modulate the host sRNA pattern.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-015-1327-5) contains supplementary material, which is available to authorized users.

Aminopropyltransferases involved in polyamine biosynthesis localize preferentially in the nucleus of plant cells

Plant aminopropyltransferases consist of a group of enzymes that transfer aminopropyl groups derived from decarboxylated S-adenosyl-methionine (dcAdoMet or dcSAM) to propylamine acceptors to produce polyamines, ubiquitous metabolites with positive charge at physiological pH. Spermidine synthase (SPDS) uses putrescine as amino acceptor to form spermidine, whereas spermine synthase (SPMS) and thermospermine synthase (TSPMS) use spermidine as acceptor to synthesize the isomers spermine and thermospermine respectively. In previous work it was shown that both SPDS1 and SPDS2 can physically interact with SPMS although no data concerning the subcellular localization was reported. Here we study the subcellular localization of these enzymes and their protein dimer complexes with gateway-based Bimolecular Fluorescence Complementation (BiFC) binary vectors. In addition, we have characterized the molecular weight of the enzyme complexes by gel filtration chromatography with in vitro assembled recombinant enzymes and with endogenous plant protein extracts. Our data suggest that aminopropyltransferases display a dual subcellular localization both in the cytosol and nuclear enriched fractions, and they assemble preferably as dimers. The BiFC transient expression data suggest that aminopropyltransferase heterodimer complexes take place preferentially inside the nucleus.

Parallels and distinctions in the direct cell-to-cell spread of the plant and animal viruses

The paradigm that viruses can move directly, and in some cases covertly, between contacting target cells is now well established for several virus families. The underlying mechanisms of cell-to-cell spread, however, remain to be fully elucidated and may differ substantially depending on the viral exit/entry route and the cellular tropism. Here, two divergent cell-to-cell spread mechanisms are exemplified: firstly by human retroviruses, which rely upon transient adhesive structures that form between polarized immune cells termed virological synapses, and secondly by herpesviruses that depend predominantly on pre-existing stable cellular contacts, but may also form virological synapses. Plant viruses can also spread directly between contacting cells, but are obliged by the rigid host cell wall to move across pore structures termed plasmodesmata. This review will focus primarily on recent advances in our understanding of animal virus cell-to-cell spread using examples from these two virus families to highlight differences and similarities, and will conclude by comparing and contrasting the cell-to-cell spread of animal and plant viruses.

Wrapping membranes around plant virus infection

Positive strand RNA viruses cause membrane modifications which are microenvironments or larger intracellular compartments, also called 'viroplasms'. These compartments serve to concentrate virus and host factors needed to produce new genomes. Forming these replication sites often involves virus induced membrane synthesis, changes in fatty acid metabolism, and viral recruitment of cellular factors to subcellular domains. Interacting viral and host factors builds the physical scaffold for replication complexes. Such virus induced changes are a visible cytopathology that has been used by plant and mammalian virologists to describe virus disease. This article describes key examples of membrane modifications that are essential for plant virus replication and intercellular transport.

Contribution of topology determinants of a viral movement protein to its membrane association, intracellular traffic, and viral cell-to-cell movement

The p7B movement protein (MP) of Melon necrotic spot virus (MNSV) is a single-pass membrane protein associated with the endoplasmic reticulum (ER), the Golgi apparatus (GA), and plasmodesmata (Pd). Experimental data presented here revealed that the p7B transmembrane domain (TMD) was sufficient to target the green fluorescent protein (GFP) to ER membranes. In addition, the short extramembrane regions of p7B were essential for subsequent ER export and transport to the GA and Pd. Microsomal partitioning and bimolecular fluorescence assays supported a type II topology of p7B in planta. Mutations affecting conventional determinants of p7B membrane topology, such as the TMD secondary structure, the overall hydrophobicity profile, the so-called "aromatic belt," and the net charge distribution on either side of the TMD, were engineered into infectious RNAs to investigate the relationship between the MP structure and MNSV cell-to-cell movement. The results revealed that (i) the overall hydrophobic profile and the α-helix integrity of the TMD were relevant for virus movement, (ii) modification of the net charge balance of the regions flanking both TMD sides drastically reduced cell-to-cell movement, (iii) localization of p7B to the GA was necessary but not sufficient for virus movement, and (iv) membrane insertion was essential for p7B function in virus movement. Our results therefore indicate that MNSV cell-to-cell movement requires sequential transport of p7B from the ER via the GA to Pd, which is modulated by a combination of several signals with different strengths in the extramembrane regions and TMD of the MP.

A membrane-associated movement protein of Pelargonium flower break virus shows RNA-binding activity and contains a biologically relevant leucine zipper-like motif

Two small viral proteins (DGBp1 and DGBp2) have been proposed to act in a concerted manner to aid intra- and intercellular trafficking of carmoviruses though the distribution of functions and mode of action of each protein partner are not yet clear. Here we have confirmed the requirement of the DGBps of Pelargonium flower break virus (PFBV), p7 and p12, for pathogen movement. Studies focused on p12 have shown that it associates to cellular membranes, which is in accordance to its hydrophobic profile and to that reported for several homologs. However, peculiarities that distinguish p12 from other DGBps2 have been found. Firstly, it contains a leucine zipper-like motif which is essential for virus infectivity in plants. Secondly, it has an unusually long and basic N-terminal region that confers RNA binding activity. The results suggest that PFBV p12 may differ mechanistically from related proteins and possible roles of PFBV DGBps are discussed.

Viral cell-to-cell movement requires formation of cortical punctate structures containing Red clover necrotic mosaic virus movement protein

Movement protein (MP) of Red clover necrotic mosaic virus (RCNMV) forms punctate structures on the cortical endoplasmic reticulum (ER) of Nicotiana benthamiana cells, which are associated with viral RNA1 replication (Kaido et al., Virology 395, 232-242. 2009). We investigated the significance of ER-targeting by MP during virus movement from cell to cell, by analyzing the function of a series of MPs with varying length deletions at their C-terminus, either fused or not fused with green fluorescent protein (GFP). The C-terminal 70 amino acids were crucial to ER-localization of MP-GFP and cell-to-cell movement of the recombinant virus encoding it. However, C-terminal deletion did not affect MP functions, such as increasing the size exclusion limit of plasmodesmata, single-stranded RNA binding in vitro, and MP interacting in vivo. We discuss the possible role of this MP region in virus movement from cell to cell.

Identification and characterization of RNA-binding activity in the ORF1-encoded replicase protein of Pelargonium flower break virus

Pelargonium flower break virus (PFBV) belongs to the genus Carmovirus (family Tombusviridae) and, as with the remaining members of the group, possesses a monopartite genome of single-stranded, positive-sense RNA that contains five ORFs. The two 5'-proximal ORFs (ORFs 1 and 2) encode two polypeptides of 27 and 86 kDa (p27 and p86), respectively, that show homology with replication proteins. The p27 does not present any motif to explain its presumed involvement in replication, while p86 has the motifs conserved in RNA-dependent RNA polymerases. In this work, we have confirmed the necessity of p27 and p86 for PFBV replication. To gain insights into the function(s) of p27, we have expressed and purified the protein from Escherichia coli and tested its ability to bind RNA in vitro. The results have shown that p27 is able to bind ssRNA with high affinity and in a cooperative fashion and that it is also capable of binding other types of nucleic acids, though to a lesser extent. Additionally, competition experiments suggest that p27 has a preference for PFBV-derived ssRNAs. Using truncated forms of p27, it can be concluded that several regions of the protein contribute to its RNA-binding properties and that this contribution is additive. This study is the first to show nucleic acid-binding ability of the ORF1 product of a carmovirus and the data obtained suggest that this product plays an essential role in selection and recruitment of viral RNA replication templates.

Cellular remodeling during plant virus infection

This review focuses on the extensive membrane and organelle rearrangements that have been observed in plant cells infected with RNA viruses. The modifications generally involve the formation of spherules, vesicles, and/or multivesicular bodies associated with various organelles such as the endoplasmic reticulum and peroxisomes. These virus-induced organelles house the viral RNA replication complex and are known as virus factories or viroplasms. Membrane and organelle alterations are attributed to the action of one or two viral proteins, which additionally act as a scaffold for the assembly of a large complex of proteins of both viral and host origin and viral RNA. Some virus factories have been shown to align with and traffic along microfilaments. In addition to viral RNA replication, the factories may be involved in other processes such as viral RNA translation and cell-to-cell virus transport. Confining the process of RNA replication to a specific location may also prevent the activation of certain host defense functions.