Tobamoviruses CP Proteins Hijack Light‐Induced Protein (NbLIP1) to Promote Viral Replication by Facilitating VRO Formation

ABSTRACT

Tobamoviruses establish viral replication organelles (VROs) on the host endoplasmic reticulum (ER) for their replication, a process demanding substantial different types of lipids. However, how viruses efficiently transfer these lipids from other compartments like chloroplasts remains incompletely understood. Fibrillin (FBN) proteins are primarily localised to chloroplasts and are intimately linked with lipid metabolism and stress responses in chloroplasts. Here, we report that NbLIP1, a light‐induced FBN1 family protein in Nicotiana benthamiana, directly interacts with the coat protein (CP) of rehmannia mosaic virus (ReMV) and other related tobamoviruses (TMV, YoMV). Upon viral infection, this interaction leads to the relocalisation of NbLIP1 from chloroplasts to ER‐proximal viral replication sites. Functional assays demonstrated that overexpression of NbLIP1 significantly enhanced viral replication, viral protein accumulation, and VRO formation, while silencing NbLIP1 had the opposite effects. These findings unveil a novel viral infection mechanism whereby the viral CP hijacks the host lipid transfer protein NbLIP1 and recruits it to viral replication factories to promote viral replication, potentially by modulating lipid supply or the microenvironment remodelling at the replication sites. This study not only elucidates the role of NbLIP1 as a novel pro‐viral host factor but also provides new insights into how viruses exploit host resources across cellular compartments, suggesting NbLIP1 as a potential antiviral target.

1. Introduction

Replication of all positive‐sense single‐stranded RNA (+ssRNA) viruses is fundamentally dependent on the extensive remodelling of host intracellular membranes to form specialised ‘viral replication factories’ or organelles (VROs) (Laliberté and Sanfaçon 2010; Zhang et al. 2025). To achieve efficient genomic RNA replication, these viruses do not replicate freely in the cytoplasm. Instead, they ingeniously construct highly structured VROs on the surface of the host's ER membrane. These VROs, characterised by curved membrane and specific lipid composition, not only provide a dedicated and protected environment for the viral RNA‐dependent RNA polymerase (RdRp) to synthesise viral RNA, shielding it from host defences, but also serve to concentrate viral and co‐opted host factors that are essential for replication (Nagy and Feng 2021). However, this large‐scale membrane remodelling imposes a significant demand on the host's lipid metabolism, requiring a substantial and targeted supply of specific molecules like phospholipids and sterols (Nagy et al. 2016; Xu and Nagy 2015).

To meet the lipid demand, +ssRNA virus have evolved sophisticated strategies that go far beyond simple acquisition, systematically hijacking a suite of factors to orchestrate VRO formation (Wang 2015; Zhang et al. 2025). This process often begins by anchoring the viral replication organelles to pre‐existing subcellular membrane‐bound proteins, which serve as a foothold on specific organelles like the ER (Zhang et al. 2025). Subsequently, viruses actively recruit host membrane‐shaping and curvature‐inducing proteins, such as endosomal sorting organelles required for transport (ESCRT) proteins are essential for the formation of VROs (Barajas et al. 2009, 2014; Diaz et al. 2015; Kovalev et al. 2016), to sculpt the membrane into the characteristic high‐curvature structures of VROs. Critically, to secure a targeted and efficient supply of specific lipids, viruses hijack the host's own machinery for inter‐organellar communication. They induce the formation of virus‐induced membrane contact sites (vMCSs), which act as specialised conduits for non‐vesicular lipid transport (Stefan et al. 2017; Vance 2015). These vMCSs provide an ideal platform where co‐opted host lipid transfer proteins (LTPs) can efficiently shuttle essential lipids, like sterols and phospholipids, directly to the burgeoning replication sites (He et al. 2023; Nagy et al. 2016). Finally, to optimise this newly created niche, viruses also commandeer a range of other host pro‐viral factors (Hsp70/90, TOM1, Rabs, Pgk1) involved in metabolic processes and RNA synthesis, ensuring the VRO is a highly favourable environment for viral replication (Zhang et al. 2025). Collectivelly, this demonstrates that the co‐option of LTPs at these induced contact sites is an important viral replication strategy, allowing viruses to tap into the plant's own diverse and sophisticated network of lipid shuttling machinery. This strategic reliance on lipid‐mobilising proteins naturally leads to the question of whether viruses have also evolved to target other, less conventional protein families with lipid‐binding capabilities (Michaud and Jouhet 2019), such as the chloroplast‐resident Fibrillins, to secure their resource supply.

Rehmannia mosaic virus (ReMV), a member of the Tobamovirus genus, is a major plant pathogen that severely impacts the yield and medicinal quality of Rehmannia glutinosa (Feng et al. 2023). R. glutinosa is an important medicinal plant that relies on vegetative propagation and is thus highly vulnerable to viral accumulation (Park et al. 2009; Zhang et al. 2008). ReMV possesses a typical tobamovirus genome, encoding a replicase complex, a movement protein (MP), and a coat protein (CP). Evidences have shown that the CP of tobacco mosaic virus (TMV) is not absolutely essential for replication, as mutants lacking the CP can still replicate within individual protoplasts (Spitsin et al. 1999). This indicates that while the virus can replicate its RNA at the single‐cell level, its efficiency may be reduced because the CP plays an important regulatory role in the formation of VROs (Asurmendi et al. 2004). For instance, it has been found that the absence of CP leads to impaired VRO formation, thereby reducing overall replication efficiency (Asurmendi et al. 2004). This positions the CP as a prime candidate for mediating the acquisition of host resources, particularly the lipids essential for constructing these extensive replication membranes.

Concurrently, plant cells possess sophisticated systems to manage lipid metabolism and respond to environmental stress, with plastids (especially chloroplasts) playing a central role (Beltrán et al. 2018; Gao et al. 2020; Wilson et al. 2014). The Fibrillin (FBN) protein family, encoded by nuclear genes and primarily localised to plastids, is intimately linked to plastid lipid biology and stress responses (Gámez‐Arjona et al. 2014; Singh and McNellis 2011). FBN proteins are major structural components of lipoprotein particles within plastids called plastoglobules (PGs) (Deruère et al. 1994; Singh and McNellis 2011; Winkenbach et al. 1976). PGs, often associated with thylakoid membranes, serve as dynamic hubs for the storage and metabolism of various crucial lipophilic compounds, including the photosynthetic electron carrier plastoquinone (PQ), antioxidants like tocopherols, and the energy storage lipid triacylglycerol (TAGs) (Singh and McNellis 2011; Youssef et al. 2010). The connection between FBNs and lipids is further suggested by a conserved lipid‐binding motif and their structural similarity to Lipocalins (Laizet et al. 2004; Singh and McNellis 2011; Ting et al. 1998). Functional studies solidify this link: loss of FBN1 isoforms in Arabidopsis leads to reduced leaf TAG levels, while FBN4 is essential for proper PG development and the accumulation of lipids within them (Singh et al. 2012, 2010; Youssef et al. 2010). Furthermore, FBN gene expression is regulated by light, development, and various stresses, positioning them as key nodes connecting environmental signals with plastid lipid metabolism (El‐Sappah et al. 2024). Recent evidence suggests that wheat yellow mosaic virus NIb protein interacts with a wheat FBN protein (TaLIP) to counteract host defence responses (Zhang et al. 2019). However, reports on FBN‐virus interactions remain scarce. This leaves a crucial question: whether viruses can exploit the metabolic functions of FBNs for a pro‐viral purpose, rather than simply targeting them to suppress host defences.

Here, we identified a pro‐viral factor, NbLIP1 (XP_016504394.1), a chloroplast‐resident FBN1‐family protein. NbLIP1 interacts with CP proteins of tobamoviruses. Overexpression of NbLIP1 promotes tobamoviruses infection while silencing impairs viral infection. Crucially, we then elucidate the biochemical basis for this hijacked protein's pro‐viral role, showing that NbLIP1 is a phospholipid‐binding protein and that this activity is indispensable for promoting viral infection. Our work thus reveals a sophisticated, cross‐compartment hijacking strategy, establishing a new paradigm for how a plant virus can breach organellar boundaries to fuel its proliferation and identifying the CP‐NbLIP1 axis as a promising new target for antiviral intervention.

2. Materials and Methods

2.1. Plant Materials and Growth Conditions

N. benthamiana plants were grown in a controlled environment chamber under long‐day conditions (16 h light/8 h dark) at 25°C. Plants approximately 4–6 weeks old were used for agroinfiltration experiments.

2.2. Virus Inoculation

Agrobacterium tumefaciens strain GV3101 harbouring the infectious clone of ReMV‐HN (Feng et al. 2023) in the pCB301 vector was cultured to an OD₆₀₀ of 0.6. The bacterial cells were collected by centrifugation and resuspended in infiltration buffer (10 mM MgCl₂, 10 mM MES, 200 μM acetosyringone, pH 5.6) and kept at 25°C for 3 h. The bacterial suspension was infiltrated into the lower leaves of four‐week‐old N. benthamiana plants using a 1 mL needleless syringe. For mock inoculation, an Agrobacterium suspension carrying the empty pCB301 vector was used. Systemic chlorotic or mosaic symptoms were observed and photographed at different days post‐inoculation (dpi).

2.3. RNA Extraction and Gene Expression Analysis

Total RNA was extracted from N. benthamiana leaves using TRIzol reagent (Invitrogen) according to the manufacturer's instructions. The concentration and purity of the RNA were determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher). First‐strand cDNA was synthesised from 1 μg of total RNA using the HiScript III RT SuperMix for qPCR (+gDNA wiper) (Vazyme, #R323‐01). To analyse the relative transcript abundance of NbLIP1, semi‐quantitative RT‐PCR was performed. The cDNA template was amplified using gene‐specific primers for NbLIP1 and the internal reference gene NbEF1α. The PCR reaction was carried out for 28 cycles. The band intensities were used to qualitatively assess the changes in gene expression.

For quantitative real‐time PCR (qRT‐PCR), reactions were performed using ChamQ Universal SYBR qPCR Master Mix (Vazyme, #Q711‐02) on a CFX96 Real‐Time PCR Detection System (Bio‐Rad) according to the manufacturer's instructions and previous studies (Zhang et al. 2013). The NbEF1α gene was used as an internal reference gene for normalisation. Relative gene expression levels were calculated using the 2^−ΔΔCt method. To ensure data reliability, all qRT‐PCR analyses were performed in triplicates. All primers used for these analyses are listed in Tables S1 and S2.

2.4. Virus‐Induced Gene Silencing

To silence the endogenous NbLIP1 gene, a tobacco rattle virus (TRV)‐based VIGS system was employed (Zhang et al. 2023). A conserved 200‐bp fragment of the NbLIP1 gene was cloned into the pTRV2 vector to generate pTRV2‐NbLIP1 (Table S2). Agrobacterium cultures containing pTRV1 were mixed in a 1:1(v/v) ratio with cultures containing pTRV2‐NbLIP1, pTRV2‐NbPDS (as a positive control) (Senthil‐Kumar and Mysore 2014), or pTRV2‐GUS (Jefferson 1989). The mixed cultures were co‐infiltrated into the leaves of five‐week‐old N. benthamiana. At approximately 10–14 dpi, when the photobleaching phenotype was visible on the NbPDS‐silenced control plants, the other silenced plants were used for subsequent virus inoculation experiments. The gene silencing efficiency was confirmed by qRT‐PCR.

2.5. Generation and Identification of Transgenic Plants

To generate stable NbLIP1‐overexpressing N. benthamiana plants, the coding sequence of NbLIP1 was cloned into the pSuper1300‐Myc vector under the control of the cauliflower mosaic virus (CaMV) 35S promoter. The resulting construct was introduced into wild‐type N. benthamiana via the Agrobacterium‐mediated leaf‐disc transformation method (Zhang, Wang, et al. 2021). Putative T0 transgenic plants were selected and regenerated on MS medium supplemented with 50 mg/L kanamycin. Positive transformants were first confirmed by PCR and then transferred to soil to produce T1 seeds. Homozygous T2 lines exhibiting high expression levels of the NbLIP1 protein and mRNA, as determined by western blot and qRT‐PCR, were selected for further experiments.

2.6. Enzyme‐Linked Immunosorbent Assay (ELISA)

The methodology was adapted with slight modifications from previously established protocols (Gu et al. 2024) for quantifying viral accumulation using a double‐antibody sandwich ELISA. A 96‐well microtiter plate was coated with purified anti‐CP^ReMV polyclonal antibody (1:1000 dilution) in carbonate buffer (pH 9.6) at 4°C overnight. After washing with PBST buffer (1 × PBS, 0.05% TWEEN 20, pH 7.4), the plate was blocked with 2% non‐fat milk in PBST for 2 h. Plant leaf samples were ground in extraction buffer, and the supernatant was added to the wells and incubated at 37°C for 2 h. After washing, HRP‐conjugated anti‐CP^ReMV antibody was added and incubated at 37°C for 1 h. Finally, the 3,3′,5,5′‐tetramethylbenzidine (TMB) substrate solution was added for colour development, and the reaction was stopped with 2 M H₂SO₄. Absorbance was measured at 450 nm using a microplate reader.

2.7. Polyclonal Antibodies Preparation

To generate polyclonal antibodies against NbLIP1, CP^ReMV and CP^YoMV, the corresponding purified His‐tagged recombinant proteins were used to immunise rabbits according to standard protocols as described previously (Zhang, Zhuang, et al. 2021). Briefly, purified proteins emulsified with Freund's adjuvant were injected into the ear margin vessels of New Zealand white rabbits. Antiserum was collected after several boosts, and the antibodies were purified using protein affinity chromatography (Genscript, Cat. L00210). The specificity and titre of the antibodies were verified by Western blotting.

2.8. Subcellular Localization Analyses

A.tumefaciens cultures (OD₆₀₀ = 0.5) expressing different truncated proteins were co‐infiltrated with tomato bushy stunt virus (TBSV) P19 (to suppress gene silencing) into N. benthamiana leaves. At 2 dpi, the co‐localisation of GFP fluorescence (excitation at 488 nm, emission at 500–530 nm) and chloroplast autofluorescence (excitation at 633 nm, emission at 650–700 nm) was observed under a confocal microscope (Leica SP8).

2.9. Bimolecular Fluorescence Complementataion (BiFC) Assay

To investigate the interaction between NbLIP1 protein and CP proteins of ReMV, YoMV, and TMV, all BiFC plasmids (Table S2) were transformed into A.tumefaciens strain EHA105. The infiltration contained 0.5 of YFP^N, 0.5 of YFP^C and 0.3 of P19^TBSV. Infiltration leaves were harvested and observed at 48 h post‐infiltration (hpi) using a confocal microscope (excitation at 512 nm, emission at 527 nm).

2.10. Co‐Immunoprecipitation (Co‐IP) Assay

Co‐IP experiments were performed as previously described (Zhuang et al. 2025). A. tumefaciens harbouring pGD‐Flag‐NbLIP1 with different expression vectors (Myc‐CP^ReMV, Myc‐CP^YoMV and Myc‐CP^V) were co‐infiltrated into N. benthamiana leaves. Plant materials were homogenised in liquid nitrogen and resolved in protein extraction buffer (10% [v/v] glycerol, 25 mM Tris–HCl pH 7.5, 1 mM EDTA, 0.1% NP40, 2% [w/v] polyvinylpolypyrrolidone, 10 mM DTT, 1 mM PMSF). The homogenate was placed on ice for 30 min and centrifuged for 15 min at 5000 g and 4°C. The supernatant was then centrifuged again and gently mixed with anti‐DYKDDDDK IP Resin (GenScript, Cat. No. L00425) at 4°C for 4 h. The mixtures were washed thrice with IP buffer (10% [v/v] glycerol, 25 mM Tris–HCl pH 7.5, 1 mM EDTA, 0.1% NP40, 1 mM PMSF). Finally, the IP products were analysed via immunoblotting using anti‐Flag (Sangon Biotech, 1:5000 dilution, Cat. D191041) and anti‐Myc antibodies (Sangon Biotech, 1:5000 dilution, Cat. D191042).

2.11. Protein Expression and Purification From the E. coli

Protein expression and purification were performed as previously described with minor modification (Zhang et al. 2023). The pGEX‐4 T‐1‐NbLIP1 plasmid, encoding the GST‐NbLIP1 fusion protein, and the pET‐28a‐based plasmids, encoding His‐CP^ReMV, His‐CP^YoMV, His‐CP^V and His‐EGFP, were individually transformed into E. coli strain BL21 (DE3). Recombinant protein expression was induced by the addition of 1 mM isopropyl‐β‐D‐thiogalactopyranoside (IPTG), followed by incubation for an additional 16 h at 16°C. Cells were harvested by centrifugation, resuspended in buffer T (50 mM Tris–HCl, 150 mM NaCl, 1 mM EDTA, 1% Triton X‐100, 1 mM PMSF, pH 7.5), and disrupted by ultrasonication (Model 500 Homogeniser, Fisher Scientific, Pittsburgh, USA). The crude lysates were clarified by centrifugation at 12000 × g for 20 min at 4°C. The supernatant was passed through a Glutathione Resin (GenScript, Cat. L00206) or Ni‐NTA affinity resin (GenScript, Cat. L00250) three times. His‐fused proteins were eluted with T buffer containing varying concentrations of imidazole and concentrated using Amicon‐Ultra‐15 filters (10 kDa; Millipore). GST‐fused protein was eluted with T buffer containing 250 mM L‐glutathione and concentrated using Amicon‐Ultra‐15 filters (30 kDa; Millipore).

2.12. GST Pull‐Down Assay

GST Pull‐down assays were conducted as previously described (Gao et al. 2020). Bait GST‐NbLIP1 and His‐tagged protein were incubated with pre‐equilibrated Glutathione Sepharose 4B beads (GE Healthcare, Cat. 17–0756‐01) for 2 h at 4°C in binding buffer (50 mM Tris–HCl pH 7.5, 250 mM NaCl, 0.2% glycerol, 5 mM DTT, 0.25% Triton‐X100, and 100 mM PMSF). After washing the beads three times with wash buffer (50 mM Tris–HCl pH 7.5, 100–300 mM NaCl, and 0.25% Triton‐X100), the bound proteins were boiled in 2× SDS‐PAGE loading buffer for western blot use with a mouse monoclonal anti‐His antibody (Abmart, 1:5000 dilution, Cat. M20001) and a rabbit polyclonal anti‐NbLIP1 antibody (1:5000 dilution, prepared in‐house).

2.13. Protein‐Lipid Overlay Assay

Protein‐Lipid overlay assays were performed as previously described (He et al. 2023). To investigate the phospholipid‐binding specificity of NbLIP1 in vitro, a protein‐lipid overlay assay was performed using commercially available membrane lipid strips (Echelon Biosciences, Cat. P‐6001). These membranes are spotted with 15 different bioactive lipids, including C1P, PI3P, PI4P, PI5P, PI(3,5)P₂, PI(4,5)P₂, PI(3,4)P₂, PI(3,4,5)P₃, PE, PC, PS, PG, CL, PA, and PI and a blank control. The lipid strips were blocked for 1 h at 25°C with 3% (w/v) fatty acid‐free bovine serum albumin (BSA) in TBST buffer (20 mM Tris–HCl, 150 mM NaCl, 0.1% (v/v) Tween‐20, pH 7.4). The blocked membranes were then incubated overnight at 4°C with purified GST‐NbLIP1, GST‐NbLIP1^293AA295, or GST protein at a final concentration of 1 μg/mL. After extensive washing with TBST, the strips were incubated for 1 h at room temperature with a polyclonal anti‐NbLIP1 antibody. The strips were washed again and subsequently incubated for 1 h at room temperature with a horseradish peroxidase (HRP)‐conjugated goat anti‐rabbit IgG secondary antibody (1:5000 dilution). Following final washes, the signals were visualised by an enhanced chemiluminescence (ECL) detection kit using a chemiluminescence imaging system.

2.14. Phylogenetic Analysis

Phylogenetic analyses of NbLIP1 and other FBN family proteins from Arabidopsis, wheat, rice (Table S3), and other crops were conducted using MEGA12 Version 12.0.11 via the neighbour‐joining method. The amino acid sequences were aligned using DNAMAN version 8.0.

2.15. Quantification and Statistical Analysis

Western blot protein band intensities were analysed with ImageJ (https://imagej.net/ij/download.html), and data represent the means of three independent experiments. Student's t‐tests assessed significant differences in relative gene expression and viral accumulation. ‘***’ indicates t‐test, p < 0.001; ‘**’ indicates t‐test, p < 0.01; and ‘*’ indicates t‐test, p < 0.05. Significance was set at p < 0.05; ‘ns" indicates no significant difference. One‐way ANOVA in SPSS (version 22.0, IBM) was used to analyse data from multiple‐group comparisons. Significant differences among groups were evaluated using Duncan's multiple range test, and groups labelled with different letters indicate statistically significant differences (p < 0.05).

3. Results

3.1. Screening of the Host Range of Constructed Infectious cDNA Clone of ReMV‐HN Isolate From R. glutinosa

The rehmannia mosaic virus was found widely infecting R. glutinosa in Henan province, China, and we constructed the Agroinfiltration‐mediated infectious cDNA clone (ReMV‐HN) (Feng et al. 2023). To determine the host range of the ReMV‐HN isolate we selected various plant species for testing by inoculation, including N. benthamiana, Solanum lycopersicum , Capsicum annuum , Cucumis sativus , Helianthus annuus , and R. glutinosa . These plants represent different plant families and genera, which helps to comprehensively evaluate the host range of the ReMV‐HN isolate. Seven days post‐inoculation (dpi), the leaves of these plants appeared with chlorotic, mosaic, shrinkage, and curly symptoms to some extent (Figure S1A). Then, we detected viral CP gene expression by RT‐PCR, which revealed that this ReMV‐HN isolate could infect all tested plants (Figure S1B, upper panel). Western blot (Figure S1B) and ELISA analyses (Figure S1C) indicated that viral CP protein was actually expressed and accumulated highest in the natural host R. glutinosa . These molecular detections were consistent with the severity of symptoms appeared, that R. glutinosa showed the most severe symptoms, including prominent mosaic patterns, leaf curling, and growth retardation. N. benthamiana also exhibited severe symptoms. Tomato and pepper showed moderate symptoms, mainly characterised by slight mosaic patterns and leaf deformation, while cucumber and sunflower showed relatively mild symptoms, with only slight mosaic patterns (Figure S1A). These results collectively indicate that the ReMV‐HN isolate has a relatively broad host range, capable of infecting various plants, but with significant differences in infection severity among different hosts, with the most severe infections occurring in R. glutinosa .

3.2. CP Is Required for Efficient Replication and Systemic Infection of ReMV‐HN on N. benthamiana

To investigate the role of the ReMV‐HN coat protein (CP^ReMV) in viral systemic infection, we generated mutants targeting the CP gene within the pCB301‐ReMV‐HN infectious clone background. Specifically, a start codon mutant (pCB301‐ReMV‐HN‐CP^ATG‐AGA) and a large deletion mutant (pCB301‐ReMV‐HN‐ΔCP) were constructed (Figure 1B,D). These mutant viruses that lacked CP were inoculated into N. benthamiana leaves via Agrobacterium‐mediated infiltration. The infiltrated leaves (LL) and upper or systemic leaves (SL) (Figure 1A) were harvested at 5 days post‐infiltration (dpi) for analysis. To assess the effect of the CP on viral systemic infection, RNA and protein accumulations were analysed by RT‐PCR and immunoblot assays (Figures 1C and 1E). RT‐PCR detected both CP and MP transcripts in the infiltrated leaves of plants inoculated with pCB301‐ReMV‐HN‐CP^ATG‐AGA. However, no viral transcripts were detected in the corresponding systemic leaves (Figure 1C, upper panel). Consistent with the lack of systemic infection, western blot analysis failed to detect CP accumulation in the systemic leaves (Figure 1C, bottom panel). These results demonstrate that preventing CP translation initiation abolishes the systemic infectivity of ReMV‐HN isolate on N. benthamiana. To further confirm the requirement of CP for systemic infection, the pCB301‐ReMV‐HN‐ΔCP deletion mutant was generated (Figure 1D). As expected, due to the deletion, RT‐PCR and western blot analyses did not detect CP transcripts or protein in either infiltrated or systemic leaves (Figure 1E). While MP transcripts were detectable in the infiltrated leaves, no viral transcripts were observed in the systemic leaves (Figure 1E, upper panel). Furthermore, N. benthamiana inoculated with the pCB301‐ReMV^ΔCP mutant displayed a symptom distinctly different from the wild‐type ReMV‐HN infection (Figure 1D). In conclusion, ReMV CP is indispensable for systemic infection on N. benthamiana.

FIGURE 1.

CP is required for efficient replication and systemic infection of ReMV‐HN on N. benthamiana. (A) The inoculated leaf is designated as the ‘local leaf,’ and the upper, newly emerged leaves are the ‘systemic leaves.’ (B) Top: Schematic representation of the ReMV‐HN‐CP^AGA mutant, in which the start codon (ATG) of the CP gene was mutated to a non‐start codon (AGA). Bottom: Symptom phenotypes on N. benthamiana plants at 5 days post‐inoculation (dpi) with wild‐type (WT) ReMV‐HN or the ReMV‐HN‐CP^AGA mutant. White arrows indicate systemic leaves. Scale bar = 1 cm. (C) Molecular analysis of plants inoculated with ReMV‐CP^(AGA). Top: RT‐PCR detection of CP and Movement Protein (MP) transcripts in systemic leaves (SL) and local leaves (LL) from three independent plants (#1, #2, #3). Bottom: Western blot analysis of CP accumulation in systemic leaves from the same plants. N, a healthy plant (negative control); P, a plant infected with WT ReMV‐HN (positive control). (D) Top: Schematic representation of the ReMV‐ΔCP mutant, in which a large fragment of the CP gene was deleted. Bottom: Symptom phenotypes on N. benthamiana plants at 5 dpi with WT ReMV‐HN or the ReMV‐ΔCP mutant. Scale bar = 1 cm. (E) Molecular analysis of plants inoculated with ReMV‐ΔCP. Top: RT‐PCR detection of CP and MP transcripts. Bottom: Western blot analysis of CP accumulation. Samples and controls are as described in (C). (F) Time‐course analysis of ReMV‐HN replication in N. benthamiana local leaves. RT‐qPCR detection of viral RNA levels targeting the p126 region at different dpi. Data were normalised to EF1α and are presented as mean ± SD (n = 3). Both CP^ATG‐AGA and ΔCP mutants showed markedly reduced viral RNA accumulation compared with the wild type. (G) Quantitative comparison of viral RNA accumulation. Relative RNA levels of CP^ATG‐AGA and ΔCP mutants were significantly lower than those of WT ReMV‐HN. Statistical significance was determined by Student's t‐test (p < 0.01). (H) Schematic representation of the MP‐Flag–tagged infectious clones. A Flag epitope (DYKDDDDK) was fused to the C terminus of the MP coding region in the WT, CP^ATG‐AGA, and ΔCP backgrounds. (I) Symptom phenotypes of N. benthamiana plants inoculated with ReMV‐HN‐Flag and ReMV‐HN at 5 dpi. WT and MP‐Flag viruses induced typical mosaic and curling symptoms. Scale bar = 1 cm. (J) Western blot analysis of MP‐Flag and CP accumulation in local leaves (LL) of plants infected with ReMV‐HN‐Flag. Proteins were detected using anti‐Flag and anti‐CP antibodies. Rubisco staining served as the loading control. (K) Quantification of MP‐Flag accumulation in N. benthamiana local leaves. MP‐Flag levels were analysed by western blot at 5 dpi. Band intensities were quantified using ImageJ from three independent experiments. For each blot, the grayscale value of the first lane was set to 1 and used for normalisation. RbcL was included as a loading control. Data are presented as relative values (mean ± SD, n = 3).

RT‐qPCR analysis targeting the p126 replicase coding region with EF1α as an internal control revealed that viral RNA accumulation of the wild‐type ReMV‐HN increased continuously from 1 to 7 dpi, reaching a peak at 6–7 dpi (Figure 1F,G). In contrast, both the CP^ATG‐AGA and ΔCP mutants showed significantly reduced RNA accumulation levels during the whole infection period. These results indicate that deletion or translational inhibition of CP not only prevents systemic movement but also markedly decreases viral replication efficiency in local tissues. To further confirm the impact of CP on replication, a Flag tag was fused to the C terminus of the MP in the infectious clone (ReMV‐HN‐Flag), allowing the detection of MP as an indicator of viral replication (Figure 1H). The MP‐Flag recombinant virus exhibited no difference in infection timing or symptom severity compared to the wild‐type ReMV‐HN on N. benthamiana (Figure 1I), suggesting that the Flag insertion did not affect viral infectivity. Immunoblot analysis using anti‐Flag antibody showed that the MP‐Flag protein accumulated strongly in both infiltrated and systemic leaves of plants inoculated with ReMV‐HN‐Flag (Figure 1J). However, when CP was deleted or its translation was blocked, the accumulation of MP‐Flag was drastically reduced (Figure 1K). Together, these results demonstrate that the ReMV‐HN CP is essential not only for systemic infection in N. benthamiana but also for maintaining efficient viral replication and accumulation in inoculated leaves.

3.3. Tobamovirus CP Proteins Directly Interact With Light‐Induced Protein NbLIP1 Both In Vitro and In Vivo

ReMV is an important member of the genus Tobamovirus. Its CP shares 86.79% and 44.03% amino acid sequence identity with the CP of TMV and YoMV, respectively (Figure S2A). To further investigate the role of CP^ReMV protein in viral infection, we screened for proteins that interacted with the CP^ReMV using the yeast two‐hybrid system (Y2H). The results showed that CP^ReMV protein specifically interacts with the light‐induced protein NbLIP1 (Accession No. XP_016504394.1) in N. benthamiana. To verify this result, we constructed CP^ReMV‐YFPn, NbLIP1‐YFPc, CP^ReMV‐YFPc, and NbLIP1‐YFPn fusion expression vectors and further confirmed the association between CP^ReMV and NbLIP1 in vivo through bimolecular fluorescence complementation (BiFC) analysis (Figure 2A). In addition, western blot analysis verified the successful expression of all fusion proteins in the BiFC assay (Figure S2B). Interestingly, CP proteins of viruses with the same genus, such as TMV and YoMV, also interact with NbLIP1, suggesting that the interaction may be conserved among Tobamoviruses and has general significance. To verify the specificity of these interactions, we further performed GST pull‐down and Co‐IP assays for in vitro and in vivo interaction analysis. We first expressed and purified His‐tagged recombinant CP proteins of ReMV (His‐CP^ReMV), YoMV (His‐CP^YoMV) and TMV (His‐CP^V) (Figure S2C). The results demonstrated that CP^ReMV, CP^YoMV and CP^V interact with NbLIP1 in vitro and in vivo (Figures 2B and 3C). These results collectively indicate that NbLIP1 may be an important host factor in tobamoviruses' infection.

FIGURE 2.

Tobamovirus CP proteins directly interact with light‐induced protein NbLIP1 both in vitro and in vivo. (A) Bimolecular fluorescence complementation (BiFC) assay showing the association between tobamovirus CPs and NbLIP1 in N. benthamiana epidermal leaf. NbLIP1 and the CPs from ReMV, YoMV, and TMV were fused to the N‐terminal (nYFP) or C‐terminal (cYFP) halves of the yellow fluorescent protein (YFP). Recombinant protein was co‐expressed in N. benthamiana leaves via agroinfiltration. A representative image for the tobamoviruses CP and NbLIP1 interaction is shown. Reconstituted YFP fluorescence (green; excitation 514 nm, emission 520–560 nm), indicating protein interaction, was detected by LSCM. Bright‐field and merged images are also shown. (B) In vitro GST pull‐down assay confirming the direct interaction between NbLIP1 and tobamovirus CP. Recombinant GST‐tagged NbLIP1 or GST alone (as a control) immobilised on glutathione beads was incubated with purified His‐tagged CPs (His‐CP^ReMV, His‐CP^YoMV, and His‐CP^V). Bound proteins were analysed by western blot with an anti‐His antibody. (C) In vivo co‐immunoprecipitation (co‐IP) assay confirming the formation of a protein complex in planta. Flag‐tagged NbLIP1 was co‐expressed with Myc‐tagged CPs (from ReMV, YoMV, or TMV) in N. benthamiana. Total protein extracts were immunoprecipitated (IP) with anti‐Flag agarose beads. The input samples and immunoprecipitates were analysed by WB with anti‐Flag and anti‐Myc antibodies.

FIGURE 3.

Overexpression of NbLIP1 promotes tobamovirus infection. Experiments were performed using Agrobacterium‐mediated transient expression in N. benthamiana and transgenic N. benthamiana plants stably expressing NbLIP1 (OE‐NbLIP1). All observations and sampling were performed at 5 dpi. (A) Symptoms on N. benthamiana leaves co‐infiltrated with ReMV infectious clone and Agrobacterium expressing NbLIP1‐Myc or EGFP‐Myc (control). (B) Western blot analysis of ReMVCP accumulation in samples from (A). Fusion protein expression was detected using an α‐Myc antibody. Relative CP accumulation levels were assessed by densitometric analysis of Western blot bands using ImageJ software. Coomassie brilliant blue (CBB) staining of the large subunit of Rubisco (RbcL), which acts as the loading control. (C) Symptoms on wild‐type (WT) and OE‐NbLIP1 transgenic N. benthamiana plants inoculated with ReMV infectious clone. (D) Western blot analysis of ReMV CP accumulation in samples from (C). Relative CP accumulation levels were assessed by densitometric analysis of Western blot bands using ImageJ software. CBB staining shows equal loading. (E) Fusion protein expression was detected using an α‐Myc antibody. Relative CP accumulation levels were assessed by densitometric analysis of western blot bands using ImageJ software. CBB staining shows equal loading. (F) Western blot analysis of YoMV CP and Myc‐tagged protein accumulation in samples from (E). (G) Symptoms on WT and OE‐NbLIP1 transgenic plants inoculated with YoMV infectious clone. (H) Western blot analysis of YoMV CP accumulation in OE‐NbLIP1 and WT plants from (G). CBB staining shows equal loading. (I) Bright‐field (BF) and GFP fluorescence images of leaves from OE‐NbLIP1 transgenic plants and EGFP control plants inoculated with TMV‐GFP infectious clone. (J) Quantification of GFP fluorescent foci in samples from (I) to assess mean numbers of fluorescent foci. “***” indicates t‐test, p < 0.001. (K) Western blot analysis of GFP protein accumulation in samples from (I). Relative GFP accumulation levels were assessed by densitometric analysis of Western blot bands using ImageJ software. CBB staining shows equal loading. Scale bars are shown in relevant panels. Scale bar = 1 cm.

3.4. Overexpression of NbLIP1 Promotes Tobamovirus Infection

To investigate the function of NbLIP1 during Tobamovirus infection, we employed two strategies: overexpression of NbLIP1‐Myc in N. benthamiana using the Agrobacterium‐mediated transient over‐expression system and generation of transgenic plants stably over‐expressing NbLIP1 (OE‐NbLIP1). Plants transiently over‐expressing NbLIP1‐Myc exhibited more severe viral symptoms compared to control plants that were expressing EGFP‐Myc under ReMV‐HN infection (Figure 3A). Western blot analysis showed that viral CP accumulation was significantly higher in plants over‐expressing NbLIP1‐Myc than in control plants, with α‐Myc antibody detection confirming the expression of the fusion proteins (Figure 3B). To further verify these results, we performed the same experiment on OE‐NbLIP1 transgenic plants, which consistently showed increased viral accumulation and more severe symptoms upon NbLIP1 overexpression condition (Figure 3C,D). Similarly, when we changed ReMV with YoMV for inoculation and performed the same experiment, it was found that transient over‐expression of NbLIP1‐Myc led to aggravated symptoms (Figure 3E), and western blot analysis confirmed significantly increased viral CP^YoMV accumulation (Figure 3F). Tests in OE‐NbLIP1 transgenic plants further validated this result, displaying more pronounced symptoms than non‐transgenic plants (Figure 3G) and higher levels of CP^YoMV accumulation (Figure 3H). To visualise viral replication more directly, we used TMV‐GFP as a tool virus. In OE‐NbLIP1 transgenic plants, NbLIP1 overexpression resulted in enhanced GFP fluorescence (Figure 3I) and a significant increase in fluorescent foci compared to EGFP controls (EGFP = 38, NbLIP1 = 51) (Figure 3J). Western blot analysis showed that GFP protein accumulation was also significantly increased (Figure 3K), with accumulation levels approximately 5‐fold higher than the control group. These results consistently indicate that NbLIP1 promotes infection and replication of multiple tobamoviruses, serving as an important host pro‐viral factor in infection.

3.5. Silencing NbLIP1 Impairs Tobamovirus Infection

To further verify the function of NbLIP1 during tobamovirus' infection, we knocked down the expression of NbLIP1 using TRV‐mediated gene silencing. We cloned a NbLIP1‐specific fragment (sequence shown in Figure S6A, yellow region) into the TRV vector to specifically target NbLIP1 transcripts. To verify the effectiveness of the gene silencing system, we also constructed TRV‐NbPDS as a positive control (Table S2). The results showed that TRV‐NbPDS‐infected plants exhibited an obvious leaf photo‐bleaching phenotype (Figure 4A,D,G), indicating that the gene silencing system was working properly. RT‐PCR analysis showed that NbLIP1 expression levels were significantly reduced in TRV‐NbLIP1‐infected plants compared to TRV‐GUS‐infected (control) plants (Figure 4B,E,H). In the presence of ReMV infection, western blot analysis showed that viral CP^ReMV accumulation was significantly lower in NbLIP1‐silenced plants than in control plants (Figure 4C). Similarly, in YoMV‐infected plants, NbLIP1 silencing led to significantly reduced viral CP^YoMV accumulation (Figure 4F). Additionally, in TMV‐GFP‐inoculated plants, NbLIP1 silencing resulted in significantly reduced viral GFP accumulation (Figure 4H). These results indicate that silencing NbLIP1 can suppress infection and replication of multiple tobamoviruses, further confirming the important role of NbLIP1 in viral infection.

FIGURE 4.

Silencing NbLIP1 impairs tobamovirus infection. (A) Wild‐type N. benthamiana plants were subjected to tobacco rattle virus (TRV)‐mediated virus‐induced gene silencing (VIGS) by co‐agroinfiltration of TRV1 with TRV2‐NbPDS (positive silencing indicator), TRV2‐GUS (negative control), or TRV2‐NbLIP1 (experimental group). Leaf phenotypes are shown at 14 dpi, with photo‐bleaching in TRV1 + TRV2‐NbPDS inoculated plants. (B) Semi‐RT‐PCR analysis (28 cycles) of the NbLIP1 levels in plants from (A) (agroinfiltrated with TRV1 + TRV2‐GUS or TRV1 + TRV2‐NbLIP1; 10–12 dpi prior to ReMV inoculation). NbEF1α served as an internal control. (C) Plants with established VIGS (from Fig. B; NbLIP1 silenced or GUS control) were challenged with ReMV. Western blot analysis of CP^ReMV accumulation was performed using a specific α‐CP^ReMV polyclonal antibody. Relative CP^ReMV levels were quantified. RbcL served as a loading control. (D) Representative phenotypes of the target genes silenced N. benthamiana plants (as described in A) at 14 dpi. (E) NbLIP1 mRNA levels in VIGS‐treated plants (as in D; prior to YoMV inoculation), assessed by RT‐PCR as described in (B). (F) CP^YoMV accumulation in plants with the target genes silenced and challenged with YoMV, analysed by western blot as described in (C). (G) Leaf phenotypes of the target genes silenced N. benthamiana plants (as described in A) at 13 dpi, showing photo‐bleaching in TRV1+ TRV2‐NbPDS inoculated plants. (H) Upper panel: NbLIP1 levels in the target genes silenced plants (as in Fig. G; prior to TMV‐GFP inoculation), assessed by RT‐PCR as described in (B). Lower panel: GFP accumulation in the correspondingly treated plants of the target genes silenced and challenged with TMV‐GFP, analysed by western blot.

3.6. ReMV CP Interferes NbLIP1's Chloroplast Localization

To investigate the molecular mechanism by which NbLIP1 promotes viral infection, we first constructed a phylogenetic tree based on amino acid sequences of 24 FBN proteins from Arabidopsis, wheat, rice, and other crops (Table S3), which showed that NbLIP1 belongs to the FBN1 family (Figure S3A). Analysis of the NbLIP1 protein sequence using the TargetP‐2.0 (https://services.healthtech.dtu.dk/services/TargetP‐2.0) online prediction tool indicated that a chloroplast transit peptide (cTP) might exist within the first 56 amino acids at the N‐terminus (Figure 5A). To further characterise the domain organisation of NbLIP1, SMART domain prediction (https://smart.embl.de/) was performed, which revealed that NbLIP1 contains two conserved regions: an N‐terminal cTP domain and a C‐terminal PAP/FBN domain (Figure 5C). To visualise this domain arrangement and confirm the structural organisation, the three‐dimensional structure of NbLIP1 was modelled using AlphaFold3, showing that the predicted cTP forms a flexible N‐terminal extension, while the C‐terminal region folds into a stable PAP/FBN domain (Figure 5C). These results support the prediction that NbLIP1 is a chloroplast‐localised FBN1‐like protein with a distinct N‐terminal transit peptide.

FIGURE 5.

ReMV CP interferes with NbLIP1's chloroplast localization. (A) Prediction of an N‐terminal chloroplast transit peptide (cTP, amino acids 1–56) in NbLIP1 using TargetP‐2.0 (https://services.healthtech.dtu.dk/services/TargetP‐2.0). (B) Three‐dimensional structure of NbLIP1 generated by AlphaFold, showing an N‐terminal cTP region (red) and a C‐terminal PAP domain (cyan). (C) Schematic diagrams of the NbLIP1‐EGFP fusion constructs driven by the Cauliflower mosaic virus (CaMV) 35S promoter, including full‐length NbLIP1–EGFP, NbLIP1^cTP‐EGFP (1–56 aa), and NbLIP1^ΔcTP‐EGFP (57–322 aa, lacking the cTP). (D) Subcellular localization of full‐length NbLIP1‐EGFP, NbLIP1^cTP‐EGFP (cTP, aa 1–56), and NbLIP1^ΔcTP‐EGFP (ΔcTP, aa 56–322, lacking the SP) in N. benthamiana epidermal leaf, observed by LSCM EGFP fluorescence (green, excitation 488 nm, emission 500–550 nm); chloroplast autofluorescence (red, excitation 633 nm, emission 650–700 nm). Line‐scan intensity profiles (right panels) show the fluorescence overlap along the indicated regions. Scale bars = 20 μm. (E) Confocal microscopy of BiFC signals showing the interaction between CP^ReMV–Yn and NbLIP1–Yc in N. benthamiana leaves. BiFC fluorescence, which was originally generated by the yellow fluorescence protein (YFP) complementation, is shown in green for clarity. During ReMV infection, BiFC signals were also observed in association with the ER marker mCherry‐HDEL and the viral replication protein 126k^ReMV. Scale bars = 20 μm.

To verify the function of the predicted cTP on NbLIP1 chloroplast localization, we constructed different truncated proteins: NbLIP1^cTP‐EGFP containing only the N‐terminal chloroplast transit peptide (1–56 aa) and NbLIP1^ΔcTP‐EGFP lacking the N‐terminal signal peptide (56–322 aa) (Figure 5C). Laser Scanning Confocal Microscopy (LSCM) observations showed that the full‐length NbLIP1‐EGFP could co‐localise with chloroplasts (Figure 5D). When the N‐terminal signal peptide was removed, NbLIP1^ΔcTP‐EGFP could not co‐localise with chloroplasts, while the truncated protein containing only the chloroplast transit peptide (NbLIP1^cTP‐EGFP) still co‐localised with chloroplasts (Figure 5D). These results strongly indicated that from 1 to 56 aa of NbLIP1 is the chloroplast signal peptide. To further determine the critical region for NbLIP1 chloroplast localization, we further divided the possible cTP sequence of NbLIP1 (cTP^NbLIP1) into cTP1‐20^NbLIP1, cTP20‐40^NbLIP1, and cTP40‐56^NbLIP1, which revealed that all 1–56 aa are essential for NbLIP1's chloroplast localization (Figure S3B).

Subsequently, we investigated the interaction region between NbLIP1 and CP^ReMV. We found that CP^ReMV could interact with the NbLIP1^ΔcTP but not with the NbLIP1^cTP segment (Figure S3C). This indicates that the interaction region between viral CP and NbLIP1 is located in the non‐cTP region of NbLIP1. More importantly, when we further observe the subcellular localization of the interaction sites between CP^ReMV and NbLIP1 under viral infection conditions, we discovered that their interaction did not occur in chloroplasts but was localised around chloroplasts (Figure 5E) and co‐localised with the viral replication protein 126 k as well as the endoplasmic reticulum marker mCherry‐HDEL (Figure 5E). This result suggests that the virus, through its encoded CP protein, interacts with NbLIP1 and changes the subcellular localization of originally chloroplast‐localised NbLIP1, redirecting it to the vicinity of the endoplasmic reticulum (the replication site of tobamoviruses), thereby potentially facilitating viral replication.

To determine whether the subcellular localization pattern of NbLIP1–CP interaction is conserved among different tobamoviruses, BiFC assays were performed using CP proteins from YoMV and TMV. Under viral infection conditions, fluorescence signals derived from CP^YoMV–NbLIP1 and CP^V–NbLIP1 interactions were detected in close proximity to chloroplasts but did not overlap with chlorophyll autofluorescence (Figure S4A). These results indicate that, during infection, tobamovirus coat proteins hijack NbLIP1 and redirect it away from chloroplasts, resulting in loss of its chloroplast localization.

3.7. NbLIP1 Promotes Tobamovirus Replication by Enhancing Viral Replication Organelles Formation

To determine whether NbLIP1 associates with tobamovirus replication complexes, we first examined its subcellular distribution during ReMV infection. In infected leaves, RFP–NbLIP1 formed numerous discrete puncta in the cytoplasm that showed extensive colocalization with viral RNA reporters. Two mechanistically distinct reporter systems were employed: a B2/VP35‐based sensor that labels double‐stranded RNA (dsRNA) intermediates (Cheng et al. 2015), and a PUMHD‐based reporter that recognises a conserved positive‐strand RNA (+)RNA motif present in ReMV (Tilsner et al. 2009). The PUMHD domain of human Pumilio1 specifically binds two closely adjacent eight‐nucleotide sequences within the genomic RNA of tobacco mosaic virus (TMV). Given the high amino acid sequence identity between ReMV and TMV (86.79%; Figure S2A), ReMV is expected to contain the same or highly similar RNA motifs (Figure S4E), allowing efficient recognition by the PUMHD reporter system. Confocal imaging revealed that YFP‐labelled dsRNA and (+)RNA signals largely overlapped with RFP–NbLIP1 puncta (Figure 6A). High‐magnification views and corresponding fluorescence intensity line‐scan analyses in Figure S4B–D further confirmed the strong spatial colocalization between NbLIP1–RFP and the viral RNA reporters. Quantitative analysis confirmed a significant increase in both the number and fluorescence intensity of YFP puncta in the presence of NbLIP1 (Figure 6B), indicating that NbLIP1 accumulates in regions enriched for viral RNA replication.

FIGURE 6.

NbLIP1 promotes tobamovirus replication by enhancing viral replication organelles formation. (A) Confocal images of viral RNA species in transiently co‐expressed N. benthamiana leaves using two RNA visualisation systems: the B2/VP35 dsRNA reporter and the PUMHD v(+)RNA reporter. (B) Quantification of fluorescent puncta from both RNA reporter systems in transient NbLIP1‐overexpression assays, showing a significant increase in dsRNA and v(+)RNA foci compared with controls. (C) Triple‐colour co‐localization images showing YFP‐labelled viral RNA (dsRNA and v(+)RNA), RFP‐tagged NbLIP1, and BFP‐tagged CP in infected cells. Both RNA markers co‐localise with RFP‐NbLIP1 and BFP‐CP, demonstrating that NbLIP1 accumulates at viral replication sites enriched with viral RNAs and structural proteins. Scale bars = 2 μm. (D) LSCM images of ReMV‐126 K‐RFP fluorescent foci (indicator of viral replication organelles, VROs) in epidermal leaf of wild‐type (WT) N. benthamiana and two independent transgenic lines stably overexpressing NbLIP1 (OE‐NbLIP1‐1, OE‐NbLIP1‐2). Plants were transiently co‐expressing p126^ReMV‐RFP along with other ReMV components. RFP fluorescence (red, excitation 561 nm, emission 580–610 nm, bright‐field (Bright), and merged (Merge)) images are shown. Observations and samples were taken at 48 hpi post co‐expression. Scale bar = 20 μm. (E) Quantification of the number of p126^ReMV‐RFP fluorescent puncta from images as shown in (A) for WT, OE‐NbLIP1‐1, and OE‐NbLIP1‐2 plants. “*” indicates t‐test, p < 0.05. (F) Western blot analysis of p126^ReMV‐RFP, ReMV CP, and NbLIP1 accumulation in total protein extracts from leaves of WT, OE‐NbLIP1‐1, and OE‐NbLIP1‐2 plants treated as in (D). Proteins were detected using α‐RFP, α‐CP, and α‐NbLIP1 antibodies, respectively. RbcL served as a loading control.

To further investigate the role of NbLIP1 during viral infection, we simultaneously visualised YFP‐labelled viral RNA, RFP–NbLIP1, and BFP–CP in infected cells (Figure 6C). In both the B2/VP35‐dsRNA and PUM–(+RNA) reporter systems, viral RNA signals coincided with NbLIP1 and CP, forming discrete tri‐colour foci reproducibly observed across multiple cells and experiments. These results indicate that NbLIP1 is recruited by the virus to the vicinity of replication complexes, where it localises near active sites of RNA synthesis and capsid assembly. Such spatial convergence suggests that NbLIP1‐enriched regions represent replication–assembly interfaces that coordinate viral RNA production with structural protein accumulation.

To assess the functional relevance of NbLIP1, we next analysed ReMV replicase dynamics in NbLIP1‐overexpressing plants (OE‐NbLIP1‐1 and OE‐NbLIP1‐2) compared with NTcontrols. Confocal imaging of the replicase small subunit p126^ReMV–RFP revealed a marked increase in both the number and brightness of fluorescent foci in NbLIP1‐overexpressing leaves (Figure 6D). Automated quantification confirmed a statistically significant elevation in puncta density and mean fluorescence intensity (Figure 6E). Immunoblotting further demonstrated higher accumulation of p126^ReMV–RFP and CP in NbLIP1‐overexpressing plants relative to controls (Figure 6F). Collectively, these findings indicate that NbLIP1 recruitment to replication sites promotes the formation and stabilisation of viral replication factories, facilitating efficient tobamovirus replication.

3.8. NbLIP1 Promotes Tobamovirus Infection Through a Conserved Phospholipid‐Binding Motif

Given that NbLIP1 is redirected to the ER, a major site of lipid metabolism, we hypothesized that it might directly interact with phospholipids to exert its function. To identify potential lipid‐binding motifs, we conducted a MEMO‐based analysis of FBN family proteins across different species (Figure S5A,B). This revealed a conserved YLD domain, which has been previously implicated in lipid binding (Li et al. 2020). A corresponding motif was identified in NbLIP1, centering on residues 293 to 295 (Figure S5B). To examine the functional relevance of this motif, we generated a lipid‐binding–deficient mutant, NbLIP1^293AA295, in which the conserved tyrosine–leucine–aspartate (YLD) residues were substituted with alanine (Figure 7B).

FIGURE 7.

NbLIP1 promotes tobamovirus infection through a conserved phospholipid‐binding motif. (A) Protein‐lipid overlay assay. Membranes spotted with various phospholipids were incubated with purified GST‐NbLIP1, the lipid‐binding‐deficient mutant GST‐NbLIP1(293AA295), or GST alone as a negative control. (B) Schematic diagram of the wild‐type NbLIP1 protein and the lipid‐binding‐deficient mutant, NbLIP1(293AA295). The location of the targeted mutation at amino acid residues 293–295 is indicated. (C) Quantification of the binding signals for all phospholipids from the assay in (D). The grayscale values of the dots were quantified using ImageJ software. Data represent mean ± SEM from three independent experiments. Different letters above the bars indicate statistically significant differences (one‐way ANOVA, p < 0.05). (D) Symptoms on ReMV‐inoculated N. benthamiana leaves transiently expressing EGFP‐Myc (control), wild‐type NbLIP1‐Myc, or the mutant NbLIP1 (²⁹³AA²⁹⁵)‐Myc at 5 days post‐inoculation (dpi). Scale bar = 1 cm. (E) Western blot analysis of ReMV CP and Myc‐tagged protein accumulation in samples from (G). CBB staining of RbcL served as a loading control. (F) Figure was Created in BioRender. Chen (2025) https://BioRender.com/ithh6jh.

To test this, we performed a protein‐lipid overlay assay using purified proteins (Figure S2C) and commercially available lipid strips. The results showed that purified GST‐NbLIP1 robustly bound to several phospholipids, exhibiting particularly strong signals for phosphatidylinositol 3‐phosphate (PI3P), and phosphatidylinositol 4‐phosphate (PI4P), phosphatidylinositol 3,4‐phosphate [PI(3,4)P], as well as other phosphoinositides (Figure 7A). In contrast, the GST protein alone showed no binding activity (Figure 7A). Strikingly, the GST‐NbLIP1^293AA295 mutant, which targets a putative lipid‐binding motif, completely lost its ability to bind to any of the lipids on the membrane (Figure 7A). Quantification of the signal intensities (Figure 7C) confirmed this result: wild‐type NbLIP1 showed significantly higher binding signals to PI3P, PI4P, and PI(3,4)P₂ compared with other phospholipids, while the mutant exhibited background‐level values comparable to the GST control. The results not only confirm that NbLIP1 is a phospholipid‐binding protein but also demonstrate that the region around residues 293 and 295 is critical for the lipid binding activity.

To determine if this specific lipid‐binding activity is required for the pro‐viral function of NbLIP1, we transiently expressed the wild‐type NbLIP1‐Myc, the lipid‐binding‐deficient mutant NbLIP1^293AA295‐Myc, or an EGFP‐Myc control in the presence of ReMV systemic infection on N. benthamiana. Overexpression of wild‐type NbLIP1 significantly enhanced viral symptoms and CP^ReMV accumulation in local leaves, while the NbLIP1^293AA295 mutant (which lost lipid binding activity) completely failed to promote viral infection, as well as the control treatment (overexpression EGFP‐Myc) (Figure 7D). Western blot analysis confirmed that the levels of CP^ReMV accumulation in plants expressing the mutant were comparable to those plants that overexpressed the EGFP‐Myc control, despite the mutant protein being expressed at a similar level to the wild‐type protein (Figure 7E). This crucial result demonstrates that the lipid‐binding capability of NbLIP1 is indispensable for its function in promoting tobamovirus infection.

Finally, based on all these findings, we propose a working model for how NbLIP1 assists tobamoviruses infection (Figure 7F). In this model, under normal conditions (without tobamoviruses infection), the nuclear‐encoded NbLIP1 protein, its mRNA is translated in the cytoplasm, and the N‐terminal signal peptide (SP) guides the nascent polypeptide, with the assistance of chaperones like HSP70 and HSP90, through the Toc and Tic translocon complexes of the chloroplast outer and inner membranes, respectively, for proper localization within the chloroplast stroma, where it participates in chloroplast functions, potentially related to lipid metabolism or plastoglobule (PG) structure/function. Upon tobamoviruses infection (with tobamoviruses infection), after the virus uncoating, releasing its genomic RNA and proteins, including the coat protein (CP). The viral CP protein then ‘hijacks’ NbLIP1, potentially preventing its entry into chloroplasts or re‐routing it from chloroplasts, and directs it to the ER. On the ER, the viral replicase (p126, p183) along with other viral and host factors (such as ARL8, EF1A, JEF1B, MP, PABP5, TOM1) assemble to form VROs. As the formation and maintenance of VROs demand substantial amounts of host lipids, the hijacked NbLIP1, now localised near the ER, likely facilitates the recruitment of essential lipids (phospholipids and sterols) for the functional VROs formation. This might occur through its potential lipid‐binding/transport capabilities or by influencing host lipid synthesis/trafficking pathways. The efficient supply of lipids, mediated by NbLIP1, subsequently promotes the formation and function of VROs, enhances the activity of the viral replicase, and ultimately leads to increased accumulation of viral RNA and proteins (like CP), exacerbating viral infection.

4. Discussion

The establishment of viral replication organelles (VROs) is a hallmark of positive‐sense RNA viruses (Choudhary et al. 2020; de Beijer et al. 2024), including tobamoviruses, and this process invariably involves the extensive remodelling of host endomembranous and a substantial demand for host lipids (Laliberté and Sanfaçon 2010; Neufeldt et al. 2018; Strating and van Kuppeveld 2017). This study unveils a novel mechanism employed by tobamoviruses to secure these lipid resources. We demonstrate that the viral CP directly interacts with NbLIP1, a host FBN1 family protein primarily localised to chloroplasts. This interaction leads to the strategic relocalization of NbLIP1 from chloroplasts to ER‐proximal viral replication sites, ultimately promoting viral replication by VROs formation. The pro‐viral role of NbLIP1 was unequivocally established, as its overexpression significantly enhanced viral replication and VRO formation, while its silencing had the opposite inhibitory effect. This CP‐mediated, cross‐compartment relocalization of a pro‐viral host factor represents a novel mechanism for tobamoviruses to manipulate host cell machinery. Comparative analyses across multiple species indicated that FBN homologues from several plants, including N. benthamiana, S. lycopersicum and R. glutinosa , show similarly strong potential for interaction with CP (Figure S1D,E,F), suggesting that this CP–FBN association may be evolutionarily conserved and might underlie, at least in part, the variation in viral pathogenicity observed among different hosts.

The central question then becomes: how does NbLIP1, a chloroplast‐associated FBN1 protein, promote viral replication at the ER? Our study provides a direct biochemical explanation: NbLIP1 is a phospholipid‐binding protein. The protein‐lipid overlay assay unequivocally showed that NbLIP1 robustly binds to key structural and signalling phospholipids, most notably PI(3,4)P and various phosphoinositides (PIPs) (Figure 7D). Critically, this lipid‐binding capability is indispensable for its pro‐viral function. A single point mutation (NbLIP1^293AA295) that abrogated its lipid‐binding activity also completely abolished its ability to promote viral infection (Figure 7D,E). Considering the substantial lipid requirements for VRO biogenesis (Strating and van Kuppeveld 2017) and the well‐documented association of FBN proteins with plastid lipid metabolism, plastoglobules (PGs), and their potential lipid‐binding/transport capabilities (Gámez‐Arjona et al. 2014; Singh and McNellis 2011). Therefore, we propose that the hijacked NbLIP1, now positioned at the ER, acts as a lipid modulator to facilitate the recruitment, organisation or concentration of specific lipids required for the biogenesis and stability of the VROs, a model strongly supported by previous literature on FBNs and lipid metabolism. Moreover, neither overexpression nor silencing of NbLIP1 markedly altered the expression of representative SA‐ or JA‐pathway marker genes (Figure S6B), suggesting that its pro‐viral activity is not associated with major modulation of canonical plant defence signalling.

The phenomenon of viruses manipulating host lipid metabolism and membrane trafficking to create replication sites is well‐documented across diverse viral families, affecting both plant and animal hosts (Nagy et al. 2016; Neufeldt et al. 2018; Strating and van Kuppeveld 2017). Animal viruses, particularly members of the Flaviviridae family such as Dengue virus and Hepatitis C virus, exemplify sophisticated manipulation of host lipid resources (Neufeldt et al. 2018). They extensively remodel ER membranes and co‐opt lipid metabolic pathways, including fatty acid and cholesterol biosynthesis, and can even induce the expression of host genes required for lipid production to create a lipid‐rich environment conducive to their replication (Neufeldt et al. 2018). However, our study presents several unique aspects. Firstly, to our knowledge, this is the first report demonstrating the hijacking of an FBN family protein, specifically a chloroplast‐localised FBN1 member (Figure 6A), by a tobamovirus CP to promote viral replication. While FBN proteins are known for their roles in plant development, stress responses and plastid lipid organisation (Kim et al. 2015; Singh and McNellis 2011), their co‐option by viruses represents a novel facet of their function. Secondly, the mechanism detailed here involves a clear CP‐mediated, cross‐compartment relocalization of a host protein from chloroplasts to ER‐proximal VROs. While ER remodelling and recruitment of factors from other endomembrane systems are known for many viruses (Laliberté and Sanfaçon 2010), the direct recruitment of a chloroplast lipid‐associated protein to the ER for viral replication offers a new dimension to tobamovirus‐host interactions. However, the specific involvement of a chloroplast FBN protein redirected by CP, as demonstrated in our study with a plant tobamoviruses, highlights a distinct host‐virus interaction strategy. Thirdly, this study expands the known functional repertoire of the tobamoviruses CP beyond its established roles in virion assembly, RNA protection, and systemic movement (Ivanov and Mäkinen 2012), our data reveal a more active role for CP in the biogenesis or functional enhancement of VROs by directly recruiting a specific host metabolic protein to these sites.

The findings presented here have significant implications. Theoretically, they deepen our understanding of the intricate molecular dialogues between tobamoviruses and their hosts, unveil a novel function for FBN proteins in the context of viral infection, and provide new insights into how viruses coordinate host resources across different cellular compartments. Practically, NbLIP1, as a pro‐viral host factor, emerges as a potential target for developing novel antiviral strategies. Interfering with the CP‐NbLIP1 interaction or modulating NbLIP1 activity could offer a new avenue for controlling tobamovirus infections. Future research should aim to precisely delineate the molecular mechanisms by which NbLIP1 facilitates lipid supply to VROs, including identifying the specific lipids it might bind or transport and its impact on the overall lipid profile of replication sites. Investigating the consequences of NbLIP1 relocalization on its native functions within chloroplasts and exploring whether other FBN family members or other viruses employ similar hijacking strategies will also be crucial. Furthermore, elucidating the structural basis of the CP‐NbLIP1 interaction could provide a roadmap for targeted antiviral interventions.

Funding

This work was supported by the National Natural Science Foundation of China (32372486, 31801699), the Excellent Youth Fund of Jiangsu Natural Science Foundation (BK20220116), the Agricultural Science and Technology Independent Innovation Fund of Jiangsu Province (CX[24]3012), the ‘National Foreign Experts Project’ of the Ministry of Human Resources and Social Security of the People's Republic of China (H20240527), the Postgraduate Research and Practice Innovation Program of Jiangsu Province (SJCX24_2279), and the Chinese Government Scholarship (China Scholarship Council, CSC) (File No. 202108320223), and the Young and Middle‐aged Academic Leaders of the ‘Qinglan Project’ of Yangzhou University.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Data S1: Supporting Information.

Figure S1: Screening of the host range of constructed infectious cDNA clone of ReMV‐HN isolate from R. glutinosa. (A) Symptoms observed on various plant species mechanically inoculated with pCB301‐ReMV‐HN infectious clones. The upper row shows whole plants, and the lower row displays enlarged views of symptomatic leaves. Bars = 1 cm. (B) Immunoblot detection of ReMV coat protein (CP) accumulation in infected leaves. Equal amounts of total protein were separated by SDS–PAGE and probed with anti‐ReMV CP antibodies. M, molecular‐mass marker; P, positive control. (C) Quantification of viral accumulation by ELISA. Bars represent mean ± SD (n = 3). Different letters above bars indicate significant differences (p < 0.05, one‐way ANOVA followed by Tukey's test). Representative ELISA plates are shown below the graph. (D) Phylogenetic analysis of FBN family proteins homologous to NbLIP1 (XP_016504394.1) identified by BLAST searches across six plant species. The neighbour‐joining tree was generated using MEGA 11; bootstrap values (1000 replicates) are indicated at the nodes. (E) Predicted interaction models between ReMV CP and FBN homologues from different species generated by AlphaFold 3. Structural alignments highlight the conserved interface regions potentially mediating CP–FBN binding. (F) Binding affinities between ReMV CP and FBN homologues predicted using the PRODIGY web server (https://rascar.science.uu.nl/prodigy/). The estimated binding free energy (ΔG) and dissociation constant (Kd) values suggest that FBN proteins from Rehmannia glutinosa, N. benthamiana, and S. lycopersicum exhibit the strongest predicted interactions with ReMV CP. Although other factors may influence infection efficiency, these computational results provide complementary evidence supporting NbLIP1‐like proteins as potential CP‐interacting factors.

Figure S2: Sequence conservation of tobamovirus coat proteins and preparation of recombinant NbLIP1 proteins for lipid‐binding assays. (A) Multiple sequence alignment of coat proteins (CPs) from TMV, YoMV, and ReMV generated using ESPript 3.0. Conserved residues are shaded in red, and similar residues are shown in yellow. (B) Immunoblot detection of BiFC fusion proteins shown in Figure 2B. Total proteins extracted from Nicotiana benthamiana leaves were analysed by SDS–PAGE and probed with anti‐LIP (upper panel) and anti–TMV‐CP (lower panel) antibodies to verify the expression of NbLIP1‐ and CP‐fusion proteins, respectively. (C) Coomassie Brilliant Blue staining showing purified GST–NbLIP1 and GST–NbLIP1^293AA295 recombinant proteins expressed in E. coli . These preparations were used in the protein–lipid overlay assays shown in Figure 7A–C.

Figure S3: Phylogenetic analysis and functional characterisation of NbLIP1 chloroplast transit peptide and its interaction with ReMV coat protein. (A) Phylogenetic tree of FBN family proteins from Nicotiana benthamiana, Arabidopsis thaliana , Triticum aestivum , Oryza sativa , and other plant species constructed using MEGA 11 with the neighbour‐joining method. Bootstrap values (1000 replicates) are indicated at the nodes. NbLIP1 clusters within the FBN1 subfamily, supporting its classification as an FBN1‐like protein. (B) Subcellular localization of NbLIP1 chloroplast transit peptide truncations. Confocal images show N. benthamiana epidermal cells expressing GFP‐fused NbLIP1 fragments (NbLIP1^cTP‐EGFP, NbLIP1^cTP1–20‐EGFP, NbLIP1^cTP20–40‐EGFP, and NbLIP1^cTP40–56‐EGFP). RFP represents chlorophyll autofluorescence. Merged images and fluorescence intensity profiles (right) indicate that amino acids 1–56 are required for NbLIP1 chloroplast localization. Scale bars = 20 μm. (C) Interaction analysis between ReMV coat protein (CP^ReMV) and NbLIP1 truncations using BiFC assays. N. benthamiana leaves co‐expressing the indicated YFP‐split constructs were examined under confocal microscopy. CP^ReMV interacts with NbLIP1^ΔcTP but not with NbLIP1^cTP, indicating that the interaction region resides outside the chloroplast transit peptide. Scale bars = 20 μm.

Figure S4: Subcellular localization of NbLIP1 interactions with CPs from different tobamoviruses under viral infection conditions. (A) BiFC analysis of NbLIP1–CPYoMV (A) and NbLIP1–CPTMV (B) interactions under corresponding viral infection conditions. Nicotiana benthamiana leaves co‐expressing NbLIP1‐nYFP and CP‐cYFP constructs were examined by laser scanning confocal microscopy. Fluorescence signals from reconstituted YFP indicate protein–protein interactions, while chlorophyll autofluorescence marks chloroplasts. Merged images and fluorescence intensity line‐scan profiles (right panels) show that NbLIP1–CPYoMV and NbLIP1–CPTMV interactions occur near but not within chloroplasts. Scale bars = 20 μm. (B) Western blot analysis of protein expression in ReMV‐infected samples. Proteins were detected using antibodies against GFP and RFP. The upper panel (α‐GFP) shows the expression of GFP‐tagged proteins, while the lower panel (α‐RFP) shows the expression of RFP‐tagged proteins. The red arrows indicate the expected protein bands for NbLIP1 and control samples. EV denotes empty vector control, and NbLIP1 indicates the NbLIP1‐expressing construct.

(C, D) Line scan analysis of colocalization between NbLIP1–RFP and viral markers. The green and orange curves represent the relative fluorescence intensity profiles of NbLIP1–RFP and viral RNA markers, respectively, along a defined distance (pixels) across the selected regions. Panel C shows the intensity distribution for one colocalized region, and panel D for another region, both indicating strong colocalization of NbLIP1 with viral RNA. (E) Schematic representation of the ReMV construct used for PUMHD system recognition. The highlighted region (in red) represents the sequence motif recognised by the PUMHD system, a conserved eight‐nucleotide sequence (TGTAGATA) located within the 183 k gene, which allows the system to specifically label the positive‐strand RNA of ReMV. The PUM recognition motif (shown in red) is essential for efficient binding by the PUMHD reporter system.

Figure S5: Sequence analysis of NbLIP1 and its homologues reveals a conserved YLD motif. (A) Sequence logo of the conserved YLD motif across multiple NbLIP1 homologues from different plant species. The height of each letter represents the relative frequency of the amino acid at that position, indicating conserved regions of the motif. (B) Multiple sequence alignment of NbLIP1 and its homologues from, SlLIP1, CaLIP1, AtLIP1, OsLIP1, and BoLIP1 showing the conservation of the YLD motif (highlighted in yellow) and other key residues (highlighted in red). The alignment was generated using, using ESPript 3.0 (https://espript.ibcp.fr/ESPript/cgi‐bin/ESPript.cgi). Conserved residues are shaded in red, similar residues in yellow, and the YLD motif is highlighted.

Figure S6: Verification of TRV:NbLIP1 construct and expression analysis of defence‐related genes in NbLIP1‐silenced and overexpressing plants. (A) Nucleotide sequence of the NbLIP1 coding sequence (CDS). The yellow‐highlighted region corresponds to the target fragment incorporated into the TRV:NbLIP1 construct for virus‐induced gene silencing (VIGS). Primer‐binding sites are indicated by underlined sequences. (B and C) Relative expression levels of defence‐related marker genes in N.benthamiana plants under different conditions: non‐treated (NT), NbLIP1 overexpression (OE:LIP1), and NbLIP1‐silenced (TRV:LIP1) plants. Transcript levels of PR1, PR2, PR5, and NPR1 represent salicylic acid (SA)‐responsive genes, whereas COI1, OPR3, PR3 and PDF1.2 correspond to jasmonic acid (JA)‐related genes. Data represent means ± SD from three biological replicates.

Table S1: List of the specific pairs of primers used for RT‐PCR detection.

Table S2: List of the specific pairs of primers for construction of different vectors in this study.

Table S3: List of all LIP sequences for construct of phylogenetic tree in this study.

Data Availability Statement

The data that supports the findings of this study are available in the Supporting Information S1 of this article.