The divergent V2 protein encoded by parsley yellow leaf curl virus is a suppressor of transcriptional and post-transcriptional gene silencing and a potential symptom determinant

Abstract

The Geminiviridae family encompasses over 500 species of economically important plant viruses. A new geminivirus species named parsley yellow leaf curl virus (PYLCV) was first identified in 2020 in Southeastern Iran. The PYLCV genome, comprising 2,779 nucleotides, exhibits a pairwise sequence identity of less than 66% with other geminiviruses and encodes a divergent V2 protein that lacks sequence similarity to known proteins in GenBank. This study explores the functions of this V2 through various analyses. An assessment of the V2 amino acid composition indicates a higher prevalence of acidic residues. By expressing N-terminal and C-terminal GFP fusions (GFP-V2 and V2-GFP) in Nicotiana benthamiana, we demonstrate that V2 localizes to the cytoplasm, nucleus, endoplasmic reticulum, and the Cajal body. Notably, PYLCV V2 suppressed local PTGS and delayed systemic PTGS in GFP-based silencing assays and restored GFP transcript accumulation in a systemic 16c-TGS assay, indicating suppression of transcriptional silencing. Co-immunoprecipitation assays further showed that PYLCV V2 interacts with NbAGO4-1 and with itself, but not detectably with NbSGS3 or NbHDA6, supporting a potential role for PYLCV V2 in targeting the AGO4-mediated transcriptional silencing pathway. Ectopic expression of V2 via a PVX vector induced necrotic lesions with HR-like features, together with systemic mosaic patterns and severe leaf curling in upper leaves. Importantly, PYLCV V2 retains core V2-like biological activities despite extensive primary-sequence divergence, suggesting that functional conservation may rely on structural or biophysical determinants rather than linear sequence similarity. Our findings demonstrate that the divergent V2 protein from PYLCV has the capacity to function as a viral suppressor of RNA silencing (VSR) and as a potential symptom determinant, suggesting relevant roles in viral pathogenicity analogous to those of other geminiviral V2 proteins.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-09617-0.

Introduction

The Geminiviridae family, which includes more than 500 species, is one of the biggest and most agronomically relevant groups of plant viruses. These viruses are characterized by small circular single-stranded DNA genomes ranging from 2.5 to 5.2 kb, encapsidated in unique twinned icosahedral particles measuring approximately 22 × 38 nm [1–4]. Geminiviruses represent a significant threat to global food security, infecting a plethora of economically important crops [5]. These viruses are divided into 15 genera based on their host range, insect vectors, and genome organization - which may be either monopartite or bipartite [6, 7]. Geminiviruses encode 4 to 8 canonical proteins; however, it is now clear that additional geminiviral proteins exist [8–14]. Geminivirus-encoded proteins play critical roles in viral processes, including replication, assembly, cell-to-cell movement, suppression of anti-viral defence, and induction of viral symptoms [15–17].

Recently, a new geminivirus species named parsley yellow leaf curl virus (PYLCV) was identified in Southeastern Iran [18]. This virus is responsible for a range of symptoms in parsley, including leaf curling, marginal yellowing, stunted growth, and reduced leaf size. The full-length genome of PYLCV consists of 2,779 nucleotides and its genome-wide pairwise sequence identity with all other known geminiviruses is less than 66%. The genome contains a single genome component with six open reading frames (ORFs) that are positional homologues to those canonical ones present in other geminiviruses, two virion-sense (V1 and V2) and four complementary-sense (C1, C2, C3 and C4) (Fig. 1a). Notably, the proteins encoded by the virion-sense ORFs are most closely related to those found in becurtoviruses and curtoviruses, while the proteins encoded by the complementary-sense ORFs show greater similarity to those of begomoviruses [18]. It has also been found that the virus is transmitted by the leafhopper Austroagallia sinuata, marking the first time a plant virus has been reported to be spread by a member of the Austroagallia genus [19]. An intriguing aspect of this virus is that the V2 protein exhibits no sequence similarity to any entries in GenBank, as confirmed by BLASTp and CD searches, which found no analogs or conserved domains using the amino acid sequence of PYLCV V2. Consequently, it is unclear what the functions of this V2 protein might be.

Fig. 1.

Structural and comparative analysis of parsley yellow leaf curl virus genome and V2 protein sequences. a Genomic organization of parsley yellow leaf curl virus (PYLCV), illustrating open reading frames (ORFs) as directional arrows. The intergenic region (IR) is annotated. b Pairwise identity matrix of amino acid sequences for V2 proteins of geminivirus, generated using SDT v1.3. c Alignment of the aminoacid sequences of the V2 proteins from parsley yellow leaf curl virus (PYLCV; QGV11343), beet curly top Iran virus (BCTIV; Becurtovirus betae, AFK14083), beet curly top virus (BCTV; Curtovirus betae, NP040558), tomato yellow leaf curl virus (TYLCV; Begomovirus coheni, CAA33687), tomato leaf curl Palampur virus (ToLCPalV; Begomovirus solanumpalampurense, CAP03292), and East African cassava mosaic Cameron virus (EACMCV; Begomovirus manihotiscameroonense, AF112354). The positions of the predicted putative phosphorylation motifs P1 (protein kinase CK2/protein kinase C), P2 (protein kinase CK2), P3 (protein kinase C), and the hydrophobic regions (H1 and H2) are shadowed in gray. Dashes (-) indicate alignment gaps

In general, the geminivirus-encoded V2 protein is a multifunctional component that plays a critical role in the viral life cycle. V2 significantly contributes to pathogenicity by acting as a viral suppressor of RNA silencing (VSR), including both transcriptional gene silencing (TGS) and post-transcriptional gene silencing (PTGS) [20–28], facilitates viral movement [29–32], and in some cases has been shown to induce symptoms [33–36].

RNA silencing is a critical antiviral defense mechanism in plants. Geminiviruses, however, like most plant viruses, have evolved sophisticated countermeasures, including viral suppressors of RNA silencing (VSRs), to overcome this defense. These VSRs target multiple stages of the RNA-silencing pathway, impairing both post-transcriptional gene silencing (PTGS) and transcriptional gene silencing (TGS), the latter mediating the effective methylation-mediated suppression of viral gene expression [37–41]. Geminiviruses encode multiple proteins, such as V2, V3, Rep, C2, C4, C5, and βC1, that inhibit TGS and/or PTGS by reducing the activity or accumulation of components of the RNA silencing pathway [10, 42–48] with one virus typically encoding several VSRs.

In this study, we experimentally demonstrate that the V2 protein encoded by PYLCV acts as a suppressor of transcriptional gene silencing (TGS) and post-transcriptional gene silencing (PTGS). Additionally, when expressed from a potato virus X (PVX; Potexvirus ecspotati)-based vector, it induces developmental alterations, inducing severe leaf curling and necrosis. Furthermore, PYLCV-V2 is localized in the cytoplasm, nucleus, endoplasmic reticulum, and Cajal body, as well as in punctate bodies distributed throughout the cytoplasm. Collectively, these results suggest that PYLCV V2 is a multifunctional viral protein that suppresses both TGS and PTGS and may act as a symptom determinant. Intriguingly, this work indicates that, despite its sequence divergence, PYLCV V2 maintains functions exhibited by its positional homologues in other viruses. Gaining insight into these functions and the underlying molecular mechanisms could contribute to devising effective strategies to manage geminivirus-caused diseases.

Materials and methods

Plant materials and growth conditions

Wild-type Nicotiana benthamiana plants, N. benthamiana line 16c containing an active 35 S-green fluorescent protein (GFP) transgene [49], and N. benthamiana line 16c-TGS, which contains a transcriptionally silenced GFP transgene [50], were grown in a controlled growth chamber under long-day conditions (LD; 16 h of light and 8 h of darkness) at temperatures of 25 °C and 22 °C, respectively. The seeds used in this study were sourced from the Center for Plant Molecular Biology (ZMBP) at Eberhard Karls University, Tübingen, Germany. Since the plant material was not collected from the wild, no specific collection permits were required.

Bacterial strains and growth conditions

Escherichia coli strain TOP10 was used for general cloning and subcloning procedures. For in planta expression, Agrobacterium tumefaciens strain GV3101, carrying the relevant binary vectors, was cultured in liquid LB medium (1% tryptone, 0.5% yeast extract, and 1% NaCl) supplemented with the appropriate antibiotics at 28 °C overnight.

Plasmids and cloning

The open reading frame (ORF) of V2 from the PYLCV (GenBank accession number QGV11343) and NbSGS3 (SolGenomics: Niben101Scf03392g08005.1; GenBank: KJ190939.1) were amplified using polymerase chain reaction (PCR) and subsequently cloned into the pDONR/Zeo vector (Invitrogen, California, USA). Cloning was performed both with and without a stop codon, and the constructs were verified through Sanger sequencing. The PYLCV V2 gene was subsequently Gateway-cloned into the binary vector pGWB502, which contains a cauliflower mosaic virus (CaMV; Caulimovirus tessellobrassicae) 35 S promoter for constitutive expression in plants. Additionally, the gene was cloned into the vectors pGWB505, pGWB506, and pGWB511 [51] to create green fluorescent protein (GFP) fusion proteins at the C-terminus and N-terminus, and a FLAG fusion protein at the C-terminus, respectively. The NbSGS3 gene was cloned into the vector pGWB505 to create a GFP fusion protein at the C-terminus.

To generate the PVX expression constructs, the PYLCV V2 and P19 genes were cloned into the BsaI sites of the PVX vector pICH31160 [52]. All recombinant PVX vectors were subsequently introduced into the A. tumefaciens strain GV3101. A. tumefaciens harboring an empty PVX vector or a recombinant PVX-P19 was used as a negative control and a positive control, respectively.

The primers utilized in this study are detailed in Table S1. All PCR-amplified fragments that were cloned during this research underwent complete sequencing.

Agroinfiltration and infection assays

For local and systemic PTGS and TGS suppression assays, and PVX infection, wild-type N. benthamiana plants, N. benthamiana line 16c and 16c-TGS leaves were agroinfiltrated as previously described [21, 49]. A long-wave UV lamp was used for the detection of GFP fluorescence (UVP Black-Ray B-100 AP, 100 W / 365 nm/ High intensity UV lamp, United States).

Agrobacterium-mediated transient gene expression in plants

Transient expression assays were conducted as previously described [53] with minor modifications. Briefly, binary vectors containing the PYLCV V2 gene fused to GFP at either the carboxyl or amino terminus of the viral protein were transformed into the A. tumefaciens strain GV3101. The resulting A. tumefaciens clones harboring the constructs of interest were cultured overnight in LB medium supplemented with the appropriate antibiotics at 28˚C. Afterward, bacterial cultures were collected by centrifugation at 4000 g for 10 min and resuspended in infiltration buffer (10 mM MgCl₂, 10 mM MES pH 5.6, 150 µM acetosyringone) to achieve an optical density (OD₆₀₀) of 0.5. The bacterial suspensions were then incubated at room temperature in the dark for 2 to 4 h prior to infiltration into the abaxial side of 4-week-old N. benthamiana leaves using a 1 mL needleless syringe. For experiments requiring co-infiltration, A. tumefaciens suspensions containing different constructs were mixed in a 1:1 ratio before infiltration.

Visualization of protein subcellular localization

To determine the subcellular localization of proteins, plant tissues expressing GFP-fused proteins were imaged using a Leica TCS SP8 point scanning confocal microscope (Leica Microsystems) with preset settings for GFP (Ex: 488 nm, Em: 500–540 nm). For the co-localization analysis of GFP, V2-GFP, GFP-V2, RFP-HDEL [54], and CFP-Fibrillarin [55], Confocal imaging was conducted using preset sequential scanning settings for GFP (Ex: 488 nm, Em: 500–540 nm), RFP (Ex: 561 nm, Em: 650–700 nm), and CFP (Ex: 405 nm, Em: 430–480 nm).

Protein extraction and western blot analysis

Protein extraction was performed as previously described [53] with minor modifications. In brief, 70–100 mg of infiltrated leaf samples were harvested. After grinding the plant tissue in liquid nitrogen, total proteins were extracted by adding lysis buffer (100 mM Tris-HCl pH 8.0; 150 mM NaCl; 10% glycerol; 5 mM EDTA; 5 mM DTT, 1 mM PMSF; 1% protease inhibitor cocktail; 1% NP-40) and incubating at 4˚C for 15 min. Then the supernatant was carefully collected after the samples were centrifuged at 15,000 g at 4 ˚C for 15 min. Proteins were separated by SDS/PAGE 10% and analyzed by western blot as previously described [56]. For western blot, the following primary and secondary antibodies were used at the indicated dilutions: goat polyclonal anti-green fluorescent protein (GFP) (AB0020, SICGEN, Lisbon, Portugal) at a dilution of 1:5000, and rabbit anti-goat IgG peroxidase antibody (A8919, Sigma-Aldrich, St. Louis, MO, USA) at a dilution of 1:20000.

Nucleic acid extraction and reverse transcription quantitative PCR (RT-qPCR)

Total DNA was extracted from the infected tissue using the CTAB method [57]. The Citrate-Citric Acid RNA Isolation (CiAR) method was used for RNA extraction [58]. RNA was reverse- transcribed into cDNA using the PrimeScript™ RT reagent Kit with gDNA Eraser (Perfect Real Time) (Takara, Japan) according to the manufacturer’s instructions. Elongation factor-1 alpha (NbEF1a) was used as a reference gene [59]. Relative expression levels were calculated using the comparative Ct method (2^−∆∆Ct). Each experiment was performed in triplicate. A list of all primers used for PCR, RT-PCR and RT-qPCR is provided in Table S1.

Bioinformatic analysis

The amino acid sequence of the PYLCV V2 protein was obtained from the National Center for Biotechnology Information (NCBI) under the accession number QGV11343. To identify potential homologs and conserved domains, a BLASTp analysis was conducted using the NCBI BLAST web interface (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Furthermore, a Conserved Domain (CD) search was performed utilizing the CD-search tool available on the NCBI website (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) to explore any conserved functional regions within the sequence.

V2 homolog proteins were aligned using the ClustalW algorithm (https://www.genome.jp/tools-bin/clustalw). Subsequently, hydrophobic domains were identified using ProtScale software (https://web.expasy.org/protscale/).

Multiple amino acid sequence alignments were performed using SDT v1.3 [60] to investigate phylogenetic relationships and sequence conservation. These alignments included the V2 protein from PYLCV and 40 other V2 proteins from 15 genera of the Geminiviridae family. Details of the proteins used are provided in Table S2.

Before performing the phylogenetic analysis, the best substitution model for all of our datasets was found using the Find DNA/protein models tool implemented in the MEGA 12 software [61]. The maximum likelihood phylogenetic tree using Whelan and Goldman plus discrete Gamma plus empirical codon Frequencies (WAG + G+F) substitution model was used for amino acid sequences. Evolutionary analyses were conducted in MEGA12 utilizing up to 8 parallel computing threads. Bootstrap branch support was computed for over 1000 replicates. The resulting tree was initially exported in Newick format and subsequently visualized and edited using Interactive Tree of Life (iTOL) software, version 7.4 [62, 63].

To predict the amino acid composition of the suppressor proteins, the Peptide 2.0 server was employed (https://www.peptide2.com/N_peptide_hydrophobicity_hydrophilicity.php). Moreover, cNLS Mapper was used to predict the presence of a nuclear localization signal (NLS) within 60-N-terminal amino acids with a cut-off score of 2.0 to determine whether any of the suppressor proteins localize in the nucleus (https://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi). This analysis included V2 from PYLCV and 19 RNA silencing suppressor proteins from eight genera of the Geminiviridae family. Details of the analyzed proteins are provided in Table S3.

Structural prediction and comparison of V2 Proteins

Three-dimensional structures of V2 proteins from PYLCV, TYLCV and BCTV were predicted using AlphaFold3 [64]. Predicted models were visually inspected and subsequently subjected to pairwise structural alignment using the Matchmaker tool implemented in UCSF ChimeraX. Matchmaker aligns structures based on sequence correspondence and iteratively removes structurally divergent atom pairs to identify conserved structural regions [65, 66]. Structural similarity was evaluated by calculating the root-mean-square deviation (RMSD) of aligned atom pairs after iterative pruning. Structural alignments and RMSD calculations were performed using default Matchmaker parameters.

Statistical analysis

Statistical analyses were conducted using GraphPad Prism version 8.4.3. A one-way ANOVA was performed to assess significant differences among groups. Statistical significance was determined at a p-value threshold of less than 0.05.

Results

Sequence analysis and prediction of domains and targeting signals in PYLCV V2

BLASTp and CD searches were employed to determine the presence of homologs or conserved domains in the V2 protein from PYLCV. No significant hits were identified in either analysis. To study the conservation of V2 proteins in the Geminiviridae family, we performed a multiple amino acid sequence alignment. The result indicates that PYLCV shares low amino acid identity with other V2 proteins from geminiviruses (14.50–32.50%) (Fig. 1b).

Consistent with these sequence similarities, maximum-likelihood phylogenetic reconstruction placed PYLCV V2 within a clade containing representatives of OpV1, JGV1, GGVA, and AGV1 (Fig. S1). As noted in the introduction, the proteins encoded by the virion-sense ORFs of PYLCV are most closely related to those of becurtoviruses and curtoviruses, whereas the proteins encoded by the complementary-sense ORFs exhibit greater similarity to those of begomoviruses [18]. Despite the weak sequence homology between the PYLCV V2 protein and members of the Curtovirus, Becurtovirus, and Begomovirus genera, we undertook an analysis of its phosphorylation motifs and hydrophobic regions. A comparative analysis of the amino acid sequences of V2 from PYLCV and its positional homologs in curtoviruses, becurtoviruses, and begomoviruses revealed that the V2 protein of PYLCV contains a conserved CK2/PKC phosphorylation motif, designated as P1, which is universally present in all V2 proteins [27, 67].

Two putative phosphorylation motifs, CK2 and PKC, were predicted in the V2 proteins of beet curly top Iran virus (BCTIV; Becurtovirus betae) and beet curly top virus (BCTV; Curtovirus betae), designated as P2 and P3, respectively. However, these motifs were not identified in the V2 protein of PYLCV or other begomoviral proteins (Fig. 1c). Furthermore, the hydrophobic regions of these V2 proteins exhibit a consistent pattern, featuring two hydrophobic regions, H1 and H2, located at the N-terminus, followed by an extended hydrophilic region at the C-terminus (Fig. 1c). Experimental evidence confirms that both the hydrophobic regions and P1 motif are indispensable for two key functions in Curtovirus and Begomovirus species: suppression of PTGS and mediation of viral pathogenicity [33, 67]. Analysis of the amino acid composition of selected geminivirus-encoded VSRs using the Peptide 2.0 server showed that suppressor proteins can exhibit either acidic or basic properties (Table 1). Notably, the PYLCV V2 protein was found to have a higher number of acidic amino acid residues, as shown in Table 1. Utilizing the cNLS Mapper software, we found that all selected VSRs possess nuclear localization signals (NLS), while none was identified in the PYLCV V2 protein (Table 1). Taken together, these results highlight the divergence of the PYLCV V2 protein compared to other positional homologues in the family.

Table 1.

Amino acid composition analysis and predicted features of viral suppressors of RNA silencing (VSRs) from geminiviruses

VSRs protein	Hydrophobic (%)	Acidic (%)	Basic (%)	Neutral (%)	NLS
PYLCV-V2	40.7	19.77	9.3	30.23	No
BCTIV-V2	34.71	20.66	15.7	28.93	Yes
CCDaV-V2	35.25	15.11	17.27	32.37	Yes
TbCSV-V2	37.29	11.86	17.8	33.05	Yes
TYLCV-V2	42.24	11.21	18.97	27.59	Yes
TGMV-AC2	31.78	10.08	18.6	39.53	Yes
CabLCV-AC2	37.21	10.85	19.38	32.56	Yes
AGV1-V2	41.84	17.35	10.2	30.61	Yes
BCTV-V2	36.51	18.25	17.46	27.78	Yes
GRBV-V2	36.26	18.13	14.04	31.58	Yes
GRBV-C2	40.97	10.42	14.58	34.03	Yes
MCLV-V3	34.23	12.08	18.12	35.57	Yes
MYMIV-AC2	33.09	10.29	16.91	39.71	Yes
MYMV-AC4	35.35	8.08	17.17	39.39	Yes
CLCuMuV-AC4	29	6	12	53	Yes
TYLCV-C5	50.75	7.46	17.91	23.88	Yes
EuLCuV-C5	45.54	4.46	16.34	33.66	Yes
TYLCV-βC1	43.22	17.8	11.86	27.12	Yes
SLCCNV-AC5	40.7	5.23	16.86	37.21	Yes
WDV-Rep	38.75	12.54	13.39	35.33	Yes

Conserved structural elements among geminivirus V2 proteins

Despite their limited primary sequence conservation, predicted structures of V2 proteins from PYLCV, TYLCV, and BCTV revealed a conserved structural core. Pairwise alignment of PYLCV V2 with TYLCV V2 identified six structurally equivalent atom pairs with an RMSD of 1.37 Å after iterative pruning, whereas comparison with BCTV V2 revealed a larger conserved core comprising 20 atom pairs with a lower RMSD of 0.41 Å (Fig. S2) (Table S4). These results suggest that, although V2 proteins have diverged considerably at the sequence level, specific structural elements have been maintained during evolution. Such conserved structural features may represent regions important for V2 functionality, including interactions with host factors involved in antiviral RNA silencing suppression.

The V2 Protein from PYLCV Localizes in the Cytoplasm, Endoplasmic Reticulum, Nucleoplasm, and Cajal Body

As an initial step in characterizing the V2 protein from PYLCV, we investigated its subcellular localization through transient expression of GFP-fused versions of the protein. Specifically, we transiently expressed both N-terminal and C-terminal fusions to GFP, GFP-V2 and V2-GFP, respectively, alongside free GFP as a control in Nicotiana benthamiana leaves. Tissues were collected thirty-six hours post-agroinfiltration and subsequently visualized using a Leica TCS SP8 confocal microscope. Our results show that GFP-V2 is predominantly localized within the cytoplasm, as well as in the nucleus and perinuclear regions (Fig. 2).

Fig. 2.

Subcellular localization of GFP-fused V2 protein encoded by parsley yellow leaf curl virus in epidermal cells of wild-type N. benthamiana. Leaves were agroinfiltrated with a construct expressing GFP and the GFP-V2 fusion protein. The samples were examined using a confocal microscope 36 h post agroinfiltration (hpi). Close up confocal images of the observed areas and maximum intensity projection of the Z-stack image of GFP-V2 are shown. Scale bar: 20 μm. GFP fluorescence (GFP), autofluorescence (AF), bright field channel (BF) and merge results. This experiment was conducted three times, yielding consistent results; representative images are shown

In contrast, V2-GFP was observed to localize within the cytoplasm and the nucleus; interestingly, a closer examination of the nucleus revealed a distinct dot that corresponds to the Cajal body (Fig. 3). Notably, the maximum intensity projection views derived from Z-stack images of V2-GFP revealed the presence of small punctate structures throughout the cell (Fig. 3a). Western blotting demonstrated the integrity of the fusion proteins, as evidenced by the presence of a protein band with a higher molecular mass corresponding to the size of V2 fusion proteins in these tissues (Fig. 3b).

Fig. 3.

Subcellular localization of V2 from parsley yellow leaf curl virus fused to GFP in epidermal cells of wild-type N. benthamiana. a Leaves were agroinfiltrated with constructs expressing GFP and the V2-GFP fusion protein. The samples were examined using a confocal microscope 36 h post agroinfiltration (hpi). Close-up confocal images of the observed areas and maximum intensity projection of the Z-stack image of V2-GFP are shown. Scale bar: 20 μm. The images display GFP fluorescence (GFP), autofluorescence (AF), bright field channel (BF) and merge results. b Western blot analysis of total proteins extracted from mock (untreated plants), GFP (27 kDa), GFP-V2 (37.54 kDa) and V2-GFP (38.13 kDa) fusion proteins, was performed using anti-GFP antibodies. The large subunit of Rubisco is utilized as a loading control via Ponceau S staining. This experiment was conducted three times, yielding consistent results; representative images are shown

The observed perinuclear localization of V2 prompted us to test whether PYLCV V2 associates with the endoplasmic reticulum (ER), for which V2-GFP was co-expressed with the ER marker RFP-HDEL, and co-localization was evaluated. The results indicate that PYLCV V2 indeed associates with the ER network, extending from the perinuclear region to the cell periphery, while also accumulating in the nucleus (Fig. 4a). Importantly, the localization of V2-GFP in the Cajal body was corroborated by co-localization with fibrillarin (Fig. 4b).

Fig. 4.

Subcellular localization of V2 from parsley yellow leaf curl virus fused to GFP in epidermal cells of wild type N. benthamiana. a Co-localization of V2-GFP and the ER marker RFP-HDEL. Leaves were agroinfiltrated with a construct expressing the V2-GFP fusion protein along with the ER marker RFP-HDEL and observed under a confocal microscope at 36 h post-agroinfiltration (hpi). To analyze the co-localization signals in greater detail, intensity plots were generated from randomly chosen lines (in red) across the ER, as shown in the diagrams to the right. b Co-localization of V2-GFP and the nucleolus and Cajal body marker CFP-fibrillarin. Leaves were agroinfiltrated with a construct expressing the V2-GFP fusion protein together with the Cajal body marker CFP-fibrillarin and observed under the confocal microscope at 36 hpi. The accumulation of V2-GFP in the Cajal body is indicated by white arrowheads. GFP fluorescence (GFP), autofluorescence (AF), bright field channel (BF), RFP fluorescence (RFP), CFP fluorescence (CFP), and merge are shown. Scale bar: 20 μm. This experiment was conducted three times, yielding consistent results; representative images are shown

PYLCV V2 Is a Suppressor of Post-Transcriptional Gene Silencing (PTGS)

V2 was previously reported as a VSR in both TGS and PTGS, in multiple geminiviruses [21, 22, 27, 28, 68]. To assess the VSR activity of PYLCV V2, the viral protein was first transiently co-expressed with GFP in leaves of N. benthamiana wild-type or 16c transgenic plants (which constitutively express GFP) [49] to evaluate post-transcriptional gene silencing (PTGS). An empty vector was used as a negative control, while expression of the well-characterized VSR P19 from tomato bushy stunt virus (TBSV; Tombusvirus lycopersici) was used as a positive control. At 4 days post-infiltration (dpi), the leaves co-expressing GFP and PYLCV V2 displayed a strong green fluorescence under UV light, similar in intensity to that seen in leaves co-expressing GFP and P19. In contrast, the areas (co-)infiltrated with the empty vector displayed a complete loss of fluorescence (Fig. 5a, d). Semi-quantitative RT-PCR confirmed the expression of viral genes in the infiltrated tissues (Fig. 5c, f). The differences in GFP fluorescence correlated with changes in GFP protein accumulation, as shown by Western blot analysis using an anti-GFP antibody, and with changes in GFP transcript accumulation, as quantified by RT-qPCR.

Fig. 5.

Local and systemic post-transcriptional gene silencing (PTGS) suppression assays of PYLCV V2 protein in wild-type and 16c Nicotiana benthamiana. a Local PTGS suppression assay in wild-type N. benthamiana. Leaves co-infiltrated by a mixture of two A. tumefaciens cultures expressing GFP and V2 from PYLCV, under UV light at 4 dpi. P19 from TBSV and the empty vector (EV) were used as positive and negative controls respectively, and mock (untreated plants). b Western blot was performed by anti-GFP antibody. The large subunit of Rubisco is utilized as a loading control via Ponceau S staining (Densitometric quantification of the Western blot bands is shown in Fig. S3a). Transcription of the V2 viral protein was validated using RT-PCR, with Elongation factor alpha (NbEF1α) serving as an internal control (Densitometric quantification of the RT-PCR products is shown in Fig. S3c). c RT-qPCR analysis of relative GFP expression levels in the agroinfiltrated leaf patches. NbEF1α was used as internal reference. Significant differences among samples are indicated by different letters (one-way ANOVA, p < 0.05). Six plants were agroinfiltrated per treatment (n = 6). This experiment was conducted three times, yielding consistent results; representative results are shown. d Local and systemic PTGS suppression assay in 16c N. benthamiana plants. Leaves co-infiltrated by a mixture of two A. tumefaciens cultures expressing GFP and V2 from PYLCV, under UV light at 4 dpi and 16 dpi. P19 from TBSV and the empty vector (EV) were used as positive and negative controls respectively. GFP expression in the cells surrounding the agroinfiltrated area at 4 dpi. Agroinfiltrated 16c plants were observed under UV light at 16 dpi. e Western blot was performed by anti-GFP antibody. The large subunit of Rubisco is utilized as a loading control via Ponceau S staining (Densitometric quantification of the Western blot bands is shown in Fig. S3b). Transcription of the V2 viral protein was validated using RT-PCR, with Elongation factor alpha (NbEF1α) serving as an internal control (Densitometric quantification of the RT-PCR products is shown in Fig. S3d). f RT-qPCR analysis of relative GFP expression levels in the agroinfiltrated leaf patches. NbEF1α was used as internal reference. Significant differences among samples are indicated by different letters (one-way ANOVA, p < 0.05). Six plants were agroinfiltrated per treatment (n = 6). This experiment was conducted three times, yielding consistent results; representative results are shown

While both V2 and P19 enhanced GFP accumulation, the effect of V2 was consistently less pronounced than that of P19 (Fig. 5b, e). These results suggest that PYLCV V2 functions as a PTGS suppressor.

PYLCV V2 does not suppress the cell-to-cell movement of RNA silencing in leaf patches in Nicotiana benthamiana

Agroinfiltrated 16c plants were utilized to investigate the impact of V2 on the short-range movement of RNA silencing in leaf patches of N. benthamiana. To achieve this, we monitored GFP expression under UV light in the cells surrounding the infiltrated tissues at 4 dpi to determine whether the silencing signal induced by GFP overexpression spread beyond the infiltrated area, resulting in a red ring around the patch. In 16c plants infiltrated with the empty vector, a red ring formed around the infiltrated area due to reduced GFP expression, resulting from local cell-to-cell movement of the silencing signal. This ring was also observed in leaves infiltrated with V2; however, leaves expressing P19 did not exhibit this phenotype (Fig. 5d) in accordance with previous results [21, 69]. Our findings indicate that V2 does not inhibit the short-range (cell-to-cell) spread of gene silencing in leaf patches of N. benthamiana.

PYLCV V2 delays systemic RNA silencing in N. benthamiana 16c plants

Infiltrated 16c plants were also monitored for the initiation of systemic silencing in the newly emerging leaves. After 16 dpi, plants infiltrated with the empty vector exhibited near-complete systemic silencing, evidenced by the red color under UV light. In contrast, GFP expression persisted in all newly emerging leaves of plants infiltrated with P19. In plants co-infiltrated with V2, apical leaves maintained GFP expression in some areas, although this expression was weaker than that observed in P19-infiltrated plants (Fig. 5d). At 30 dpi, plants infiltrated with V2 showed complete silencing of GFP expression. On the contrary, plants infiltrated with P19 displayed a sustained expression of GFP across all examined tissues at the same time point (data not shown). These results demonstrate that PYLCV V2 delays the long-distance spread of RNA silencing in 16c line but does not block the process entirely.

PYLCV V2 reverses the transcriptional silencing of GFP at systemic level

To determine whether PYLCV V2 could suppress methylation-mediated TGS, 16c-TGS transgenic N. benthamiana plants (in which the GFP gene is transcriptionally silenced) [50] were used. The initiation of systemic silencing in newly emerging leaves was monitored in 16c-TGS plants upon systemic expression of V2 (PVX-PYLCV V2) or P19 (PVX-TBSV P19, used as positive control), both delivered through a PVX vector. At 6 dpi, both V2- and P19-expressing plants displayed a clear restoration of GFP fluorescence in systemic tissues under UV illumination, whereas plants inoculated with PVX alone or the empty vector (EV) remained silenced and showed no detectable fluorescence (Fig. 6a). These observations indicate that PYLCV V2 is able to suppress established GFP transcriptional silencing and restore transgene expression in upper, non-inoculated leaves.

Fig. 6.

V2 protein from PYLCV functions as a suppressor of TGS. a 16c-TGS N. benthamiana plants were systemically inoculated with PVX, PVX-PYLCV V2, or PVX-TBSV P19 (positive control), or inoculated with empty vector (EV). Naïve 16c N. benthamiana plants were included as an additional control. Plants were imaged at 6 dpi under white light (upper panel) and UV light (lower panel). Each independent replicate (R1, R2) consists of a pool of three individual plants. b Systemic leaves from (a) were harvested at 6 dpi and analyzed by semiquantitative RT-PCR. Transcription of PVX, P19, and V2 was validated using specific primers. Elongation factor 1 alpha (NbEF1α) was used as internal reference. c RT-qPCR analysis of relative GFP mRNA levels in the systemic leaves from (a). NbEF1α was used as internal reference. d Immunoblot assay of GFP protein using anti-GFP antibody in systemic leaves from (a). Molecular weights (kDa) are indicated on the left. The GFP-specific band corresponding to free GFP at approximately 27 kDa is indicated with an asterisk. PS: Ponceau S staining. This experiment was conducted two times, yielding consistent results

To verify successful systemic infection and transgene expression, semiquantitative RT-PCR was performed using RNA extracted from systemic leaves. PVX transcripts were detected in all virus-inoculated samples, while V2- and P19-specific amplicons were exclusively detected in their corresponding treatments, confirming systemic accumulation of the recombinant vectors (Fig. 6b).

Consistent with the fluorescence phenotype, RT-qPCR analysis revealed a significant increase in GFP transcript accumulation in plants infected with PVX-PYLCV V2 or PVX-TBSV P19 compared with PVX or EV controls (Fig. 6c). GFP mRNA levels in these treatments indicate effective reversal of transcriptional repression.

Immunoblot analysis further confirmed that the transcriptional reactivation of GFP resulted in protein accumulation. A clear GFP-specific band corresponding to free GFP (27 kDa) was detected in systemic leaves expressing V2 or P19, whereas GFP protein was absent or barely detectable in PVX- and EV-treated 16c-TGS plants (Fig. 6d).

Together, these results demonstrate that PYLCV V2 is capable of suppressing TGS, as evidenced primarily by the restoration of GFP transcript accumulation. The increased GFP protein abundance is therefore interpreted as a downstream consequence of transcriptional reactivation.

PYLCV V2 Alters Plant Development When Expressed from a Viral Vector

To assess the pathogenicity of parsley yellow leaf curl virus (PYLCV), N. benthamiana plants were agroinoculated with an infectious clone of the virus. A control group was inoculated with A. tumefaciens strain C58 harboring the empty pGreen0029 vector (mock). At six weeks post-inoculation, plants infected with PYLCV exhibited severe upward curling of the leaves, whereas mock-inoculated plants remained asymptomatic (Fig. S4). Viral infection was confirmed by PCR amplification of an 830 bp fragment from total DNA extracted from newly emerged leaves, using the specific primer pair PYLCV-F-446/PYLCV-R-1275. No symptoms or PCR product was observed in the mock-inoculated control plants (see Fig. S4).

Plus-strand RNA viruses are often employed as expression vectors in plants due to their ability to accommodate additional transgene insertion and expression without interfering with normal viral functions [70]. For instance, a vector based on potato virus X (PVX) is effective for the systemic expression of any gene of interest in host plants susceptible to PVX. Some studies have demonstrated that ectopic expression of V2 proteins encoded by certain geminiviruses using a PVX-derived vector induces localized cell death or a hypersensitive response-like (HR-like) phenotype in infiltrated tissues, which can ultimately expand to systemic necrosis in N. benthamiana [21, 71–74].

To evaluate the capacity of PYLCV V2 to induce developmental alterations (a proxy for symptoms during infection), we constructed a recombinant PVX vector containing the V2 gene (designated as PVX-V2) to facilitate ectopic overexpression of this viral sequence in four-week-old wild-type N. benthamiana plants. Agroinfiltration with an empty PVX vector served as negative control, while infiltration with PVX expressing the P19 protein (PVX-P19) acted as positive control. At 4 dpi, tissues infiltrated with the empty vector exhibited local yellowing characteristic of PVX infection. In contrast, tissues infiltrated with PVX-V2 displayed pronounced local necrosis, similar to those infiltrated with the positive control PVX-P19 (Fig. 7a). These results suggest that, similar to the V2 proteins of begomoviruses and curtoviruses, the V2 protein from PYLCV induces necrotic cell death with HR-like features in N. benthamiana when expressed from the PVX vector.

Fig. 7.

Local and systemic symptoms induced by potato virus X (PVX; Potexvirus ecspotati) in wild-type Nicotiana benthamiana plants. Plants were inoculated with recombinant PVX viruses expressing V2 from PYLCV (PVX-V2) through agroinfiltration. P19 from TBSV and the empty vector (EV) were used as positive and negative controls respectively. a Local symptoms induced in the infiltrated area at 4 dpi. Representative plants showing systemic symptoms at 6 dpi. Scale bar: 5 cm. b Systemic leaf tissues were harvested at 6 dpi and analyzed using Semi-quantitative RT-PCR. Each experiment involved agroinfiltration of six plants. This experiment was repeated three times with similar results; representative results are shown

By 6 dpi, N. benthamiana plants inoculated with PVX expressing the V2 viral protein exhibited systemic mosaic patterns, severe leaf curling, and necrosis. Similarly, plants inoculated with PVX-P19 showed mosaic symptoms, leaf curling, and significant necrosis (Fig. 7a). Neither PVX-V2- nor PVX-P19-inoculated plants recovered from viral infection; symptoms persisted until the death of the plants at 10 dpi. In contrast, plants inoculated with the empty PVX vector displayed mosaic symptoms and yellowing but survived beyond 10 dpi (Fig. 7a).

To rule out the possibility that the observed symptoms were due to varying accumulation levels of the PVX vector, we quantified RNA levels of the PVX coat protein (cp) gene in all agroinfiltrated N. benthamiana plants using semi-quantitative RT-PCR at 6 dpi. In parallel, we assessed the expression level of V2/P19. The results indicate that the PVX cp RNA accumulated similarly across all infiltrated plants (Fig. 7a), suggesting that the severe symptoms observed in N. benthamiana plants infiltrated with PVX-V2 were not a result of increased PVX virus accumulation, but of the action of V2. Furthermore, semi-quantitative RT-PCR results confirmed the expression of each construct, revealing similar mRNA accumulation levels across individually infiltrated N. benthamiana plants (Fig. 7b). These findings indicate that PVX-mediated expression of PYLCV V2 enhances PVX-associated phenotypic alterations in N. benthamiana, including local necrosis, systemic mosaic, leaf curling, and plant decline, without a detectable increase in PVX accumulation at the sampling time tested. However, because this assay relies on heterologous expression from a PVX vector and does not fully recapitulate wild-type PYLCV infection, these data should be interpreted as evidence that V2 can modulate host phenotypic alterations rather than as definitive proof that V2 is the symptom determinant during PYLCV infection.

PYLCV V2 Interacts with Itself and NbAGO4, but Not with NbSGS3 or NbHDA6

To gain insight into the molecular mechanisms underlying the silencing suppressor activity of PYLCV V2, we investigated its ability to associate with host factors involved in RNA silencing and epigenetic regulation by co-immunoprecipitation (co-IP) assays following transient expression in N. benthamiana leaves. PYLCV V2-FLAG was co-expressed with GFP-tagged NbAGO4-1 [26], NbSGS3, NbHDA6 [26], or PYLCV V2-GFP, and GFP immunoprecipitates were analyzed using anti-FLAG antibody. Free GFP was included as negative control, whereas TYLCV V2-FLAG [75] co-expressed with GFP-NbAGO4-1 served as positive control [26].

As expected, TYLCV V2-FLAG specifically co-immunoprecipitated with GFP-NbAGO4-1, validating the experimental system (Fig. S5). Under the same conditions, PYLCV V2-FLAG was also consistently detected in GFP-NbAGO4-1 immunoprecipitates, demonstrating that PYLCV V2 associates with NbAGO4-1 in planta. In addition, PYLCV V2-FLAG co-immunoprecipitated with GFP-tagged PYLCV V2, indicating that the protein is able to self-interact and likely forms homomeric complexes.

By contrast, no specific co-immunoprecipitation signal was detected when PYLCV V2-FLAG was co-expressed with GFP-NbSGS3 or GFP-NbHDA6, despite clear accumulation of the corresponding GFP fusion proteins in input samples, suggesting that PYLCV V2 does not stably associate with these host proteins under the conditions tested (Fig. S5). No FLAG signal was detected in the free GFP negative control.

Taken together, these results identify NbAGO4-1 as a host interactor of PYLCV V2 and support a potential role of V2 in targeting the AGO4-mediated transcriptional gene silencing pathway. The observed self-interaction further suggests that oligomerization may contribute to PYLCV V2 function.

Discussion

This study demonstrates that the V2 protein encoded by PYLCV, despite its sequence divergence from other geminiviral V2 proteins, can act as a suppressor of both TGS and PTGS and as a potential symptom determinant, as established for many of its positional homologues. Additionally, it triggers a necrotic phenotype when expressed in N. benthamiana from a PVX vector. We show that PYLCV V2 is localized within the cytoplasm, ER, nucleoplasm, and Cajal body, which can offer hints on the molecular mechanisms underlying the aforementioned activities. Notably, this is the first report identifying a PYLCV-encoded protein that acts as a viral suppressor of RNA silencing (VSR) and as a potential symptom determinant.

Bioinformatic analysis of the V2 protein of PYLCV revealed no significant homology between PYLCV V2 and other proteins in GenBank, initially complicating the prediction of its role in infection. It was reported that the analysis of the amino acid composition of VSRs using the Peptide 2.0 server indicated that all VSRs possess a higher proportion of basic amino acid residues compared to acidic amino acids [76]. However, our results indicated that VSRs within the Geminiviridae family can exhibit either more acidic or more basic characteristics, indicating that is not a universal property (Table 1).

We investigated the potential VSR abilities of the V2 protein encoded by PYLCV using GFP-based silencing assays in N. benthamiana wild type, 16c, and 16c-TGS plants. PYLCV-V2 exhibited the ability to suppress both local and systemic PTGS (Fig. 5). However, V2 could not suppress the intercellular movement of RNA silencing to neighboring cells in this system (Fig. 5), suggesting that its mechanism of action does not involve the sequestration of small RNAs (sRNAs). Furthermore, systemic expression of V2 from the PVX vector reversed the transcriptionally silenced status of the GFP transgene in 16c-TGS plants, indicating that TGS can be suppressed by PYLCV V2 (Fig. 6). To gain insight into the molecular basis of this activity, we tested whether PYLCV V2 associates with known host factors involved in RNA silencing and epigenetic regulation. Co-immunoprecipitation assays showed that PYLCV V2 interacts with NbAGO4-1 and with itself, but not detectably with NbSGS3 or NbHDA6 under the conditions tested (Fig. S5). AGO4 is a central component of the RNA-directed DNA methylation (RdDM) pathway and has been shown to be targeted by other geminiviral V2 proteins during suppression of methylation-mediated TGS. Therefore, the interaction between PYLCV V2 and NbAGO4-1 supports a model in which PYLCV V2 suppresses TGS, at least in part, by targeting the AGO4-mediated RdDM pathway. The lack of detectable interaction with NbSGS3 suggests that PYLCV V2 may differ mechanistically from TYLCV V2, which suppresses PTGS through interference with the RDR6/SGS3 pathway. Similarly, the absence of detectable interaction with NbHDA6 suggests that PYLCV V2 may not rely on a stable association with this factor, although transient, indirect, or condition-dependent interactions cannot be excluded.

These findings align with several studies indicating that the V2 protein in other geminiviruses functions as a VSR, participating in the suppression of TGS and PTGS. For example, TYLCV V2 suppresses PTGS by interacting with suppressor of gene silencing 3 (SGS3) and impairing the RDR6 (RNA- dependent RNA polymerase 6)/suppressor of gene silencing 3 (SGS3) pathway [77], and suppresses TGS by interacting with histone deacetylase 6 (HDA6) and AGO4 [22, 26, 44]. The V2 protein from CLCuMuV also interacts with AGO4 in N. benthamiana, thereby counteracting RNA-dependent DNA methylation (RdDM)-mediated TGS to promote viral infection [25]. Additionally, CLCuMuV V2 sequesters long double-stranded RNA (dsRNA) and inhibit its dicer-mediated cleavage, and disrupts the interaction between calmodulin and CAMTA3, thereby counteracting the PTGS defense mechanism [78, 79]. The V2 protein of tomato yellow leaf curl China virus (TYLCCNV; Begomovirus solanumflavuschinaense) interferes with the silencing pathway by binding to both the 21-nt siRNA duplex and the 24-nt single-stranded siRNA, effectively disrupting the siRNAs generated against the virus [68]. BCTV V2 suppresses PTGS by interfering with the RDR6/SGS3 silencing pathway [23]. BCTIV V2 functions as a viral suppressor of PTGS; however, its PTGS suppression activity is independent of RDR6 [28].

These findings indicate that, despite its strong sequence divergence, PYLCV V2 retains functional properties characteristic of geminiviral V2 proteins, while potentially using a distinct combination of host targets. Future studies should determine whether the PYLCV V2–AGO4 interaction is required for TGS suppression by testing interaction-defective V2 mutants and by assessing changes in viral or transgene DNA methylation and siRNA accumulation.

The retention of V2-like biological activities by PYLCV V2 despite its extremely low primary-sequence similarity to known geminiviral V2 proteins is consistent with previous observations in the Geminiviridae. For example, curtoviral and begomoviral V2 proteins show very poor sequence homology but share a comparable genomic position, similar size, conserved N-terminal hydrophobic regions, and a putative CK2/PKC phosphorylation motif; these features have been linked to PTGS suppression, subcellular localization, systemic infection, and pathogenicity-related activities [27, 67]. Similarly, the V2 protein of BCTIV, a becurtovirus, has been shown to act as a silencing suppressor despite the absence of detectable sequence homology with begomoviral AV2 proteins [27, 67]. These examples indicate that, in geminiviral proteins, functional conservation may be maintained through conserved structural or biophysical properties rather than through extensive linear sequence conservation.

In the case of PYLCV V2, the presence of hydrophobic regions and a conserved P1 phosphorylation-related motif, together with nuclear/Cajal body localization, self-interaction, and association with NbAGO4-1, suggests that this divergent protein may have retained structural or surface features compatible with V2-like functions. This may explain how PYLCV V2 can suppress both PTGS and TGS despite its extensive sequence divergence from canonical begomoviral V2 proteins. Nevertheless, this remains a working hypothesis. Comparative structural modelling, identification of conserved interaction surfaces, and mutational analysis of the hydrophobic regions, the P1 motif, and the AGO4-interacting interface will be required to determine whether conserved three-dimensional or biophysical determinants underlie the observed functional conservation.

The multifunctionality of the V2 protein from geminiviruses is closely linked to its ability to localize to various subcellular compartments. Each localization allows V2 to interact with different cellular processes and host factors, ultimately contributing to viral replication, movement, and evasion of host defenses [80]. Gaining insights into the dynamics of V2 localization and function, particularly regarding the association between the subcellular localization of VSR proteins and their silencing suppression capabilities, is crucial for advancing our understanding of geminivirus biology and developing effective strategies for managing geminiviral infections in plants. Previous studies have shown that the NLS of certain VSRs, such as the V2 protein of CCDaV, the V2 protein of mulberry mosaic dwarf-associated virus (MMDaV), AC2 of MYMIV, C2 of tomato yellow leaf curl virus-China (TYLCV-C), TrAP of East African cassava mosaic Cameroon virus (EACMCV; Begomovirus manihotiscameroonense), AC5 of SLCCNV, 2b of cucumber mosaic virus (CMV; Cucumovirus CMV), P6 of strawberry vein banding virus (SVBV; Caulimovirus venafragariae), NS3 of rice stripe virus (RSV; Tenuivirus oryzaclavatae), and Pns11 of rice dwarf virus (RDV; Phytoreovirus alphaoryzae), is required for their PTGS suppression activities [81–89]. Given the importance of nuclear localization for the function of many VSRs, we predicted the presence of nuclear localization signals (NLS) within the protein sequences of VSRs from geminiviruses. We found that all VSRs in geminiviruses were predicted to possess an NLS, although no canonical NLS was identified in the PYLCV V2 protein (Table 1). In spite of this, our intracellular localization studies demonstrated the presence of V2 in the nucleus, suggesting that V2 may utilize a non-canonical NLS or be transported into the nucleus via interaction with a host protein. Nevertheless, considering the size of the V2 protein (37.5 kDa; 38.1 kDa when fused to GFP), it is possible that it can freely diffuse in and out of the nucleus [90–93].

Additionally, our results show that PYLCV V2 is also localized to the ER. Notably, a closer examination of the nucleus revealed a distinct dot accumulating in sub-nuclear structures that corresponded to the Cajal body, which may have implications for the biology of the infected cell (Figs. 2 and 3). Both GFP-V2 and V2-GFP demonstrated localization within the nucleus and the ER network; ER and Cajal body localization were confirmed by co-localization with the markers RFP-HDEL and fibrillarin, respectively (Fig. 4). Interestingly, maximum intensity projection views derived from Z-stack images of V2-GFP revealed the presence of small punctate structures within the cytoplasm (Fig. 3), of which the identity remains to be determined.

Cajal bodies are critical sub-nuclear structures involved in plant–virus interactions, playing essential roles in systemic infection as well as in the cell-to-cell and systemic movement of viruses. In the case of DNA viruses, CBs are also involved in the suppression of transcriptional gene silencing (TGS) and the methylation of the viral genome [94]. Notably, the V2 protein from TYLCV interacts predominantly with AGO4 in CBs, where it accumulates and inhibits methylation of the viral genome [26]. In the present study, PYLCV V2 localized to the Cajal bodies and co-immunoprecipitated with NbAGO4-1, suggesting that this divergent V2 protein may also target AGO4-associated RdDM processes. Together with the TGS suppression phenotype observed in 16c-TGS plants, these results support a potential mechanistic link between Cajal bodies localization, AGO4 interaction, and suppression of transcriptional silencing by PYLCV V2. Further work will be required to determine whether PYLCV V2 inhibits AGO4-dependent methylation of viral DNA or host transgenes directly.

In N. benthamiana, the expression of the V2 protein encoded by geminiviruses using a PVX-based vector accelerates systemic necrosis (SN) and induces a necrotic cell death with HR-like features. This study reveals that expressing PYLCV V2 from a PVX vector leads to the formation of necrotic lesions, inducing a necrotic response locally in N. benthamiana leaves. Additionally, we observed that the upper systemic leaves exhibited systemic mosaic patterns, severe leaf curling, and necrotic symptoms (Fig. 7). Notably, the V2-induced HR failed to restrict the long-distance movement of PVX-V2, ultimately leading to systemic plant death. Our findings align with previous studies demonstrating that various geminiviruses, including papaya leaf curl virus (PaLCuV; Begomovirus caricae), tomato yellow leaf curl Sardinia virus (TYLCSV; Begomovirus solanumflavusardiniaense), CLCuMuV, and tomato leaf curl Java virus (ToLCJV; Begomovirus solanumjavaense), induce HR and SN through their V2 protein [21, 34, 35, 68, 78]. The impact of V2 on plant development suggests that V2 may play a significant role in symptom determination. It should be noted that PVX-mediated expression provides a useful heterologous system to assess the capacity of viral proteins to alter host phenotypes, but it does not necessarily reproduce the expression level, timing, subcellular context, or infection biology of PYLCV. Moreover, whereas wild-type PYLCV infection induced upward leaf curling, PVX-V2 expression was associated with more severe PVX-like symptoms and downward leaf curling. Therefore, although our data support a role for PYLCV V2 in modulating symptom severity or host phenotypic alterations, they do not definitively establish V2 as the sole or direct symptom determinant of PYLCV. This will require future experiments using V2-defective or V2-mutated PYLCV infectious clones, complementation assays, and time-course analyses of viral accumulation and symptom development. However, the precise mechanisms by which V2 influences these processes and its potential role as a virulence factor remain to be fully elucidated.

Conclusion

In this study, we provide evidence that the PYLCV V2 protein is a multifunctional protein with several populations in distinct subcellular localizations. Importantly, this work represents the first identification and characterization of a viral suppressor of RNA silencing (VSR) for PYLCV, providing a foundation for future investigations into the molecular mechanisms of PYLCV pathogenesis. Subsequent research should prioritize elucidating the molecular underpinnings of these activities of V2 and assessing their conservation with previously described V2 proteins from other geminiviruses. Understanding the interactions between V2 and the host cell will be essential for developing targeted antiviral strategies to combat PYLCV infection and related geminivirus diseases.

Abbreviations

AF

Autofluorescence

AGO4

Argonaute 4

ANOVA

Analysis of variance

BF

Bright field

BLAST

Basic local alignment search tool

CB

Cajal body

CD

Conserved domain

CFP

Cyan fluorescent protein

dpi

Days post-infiltration

dsRNA

Double-stranded RNA

ER

Endoplasmic reticulum

EV

Empty vector

GFP

Green fluorescent protein

HDA6

Histone deacetylase 6

hpi

Hours post-infiltration

HR

Hypersensitive response

IR

Intergenic region

kDa

Kilodaltons

LB

Luria–Bertani

LD

Long-day

NCBI

National center for biotechnology information

NLS

Nuclear localization signals

ORF

Open reading frame

PTGS

Post-transcriptional gene silencing

RdDM

RNA-dependent DNA methylation

RDR6

RNA- dependent RNA polymerase 6

RFP

Red fluorescent protein

SDT

Sequence demarcation tool

SGS3

Suppressor of gene silencing 3

SN

Systemic necrosis

sRNA

Small RNA

TGS

Transcriptional gene silencing

VSR

Viral suppressor of RNA silencing

WT

Wild type

Authors’ contributions

Hasan Zeitooni and Laura Medina-Puche carried out and analyzed the experiments and wrote the manuscript, with input from all authors; Masoud Shams-Bakhsh and Rosa Lozano-Durán conceived and designed the study.

Funding

Work in the Lozano-Durán lab is partially funded by the Excellence Strategy of the German Federal and State Governments.

Data availability

The data that supports the findings of this study are available in the Supporting Information of this article.

Declarations

Ethics approval and consent to participate

Not applicable. This manuscript does not involve any studies with human participants or animals conducted by the authors.

Consent for publication

All authors agree to publish.

Competing interests

The authors declare no competing interests.