The P2 nucleic acid binding protein of Sugarcane bacilliform virus is a viral pathogenic factor

Source of plant materials

Sugarcane plants of the Badila cultivar that exhibited mottling, stunting, and chlorotic streaking symptoms were collected from Menghai County, Yunnan Province, China, and rapidly frozen in liquid nitrogen and then stored at −80 °C.

Multiple sequence alignment and phylogenetic analyses

Amino acid sequence similarities were determined for the P2 of six representative members of the genus Badnavirus and the sole member of Tungrovirus (Rice tungro bacilliform virus, RTBV); all sequences and the corresponding GenBank accession numbers are listed in Table 1. Sequence multiple alignment was performed with the Clustal Omega method using the SnapGene software. The corresponding sequences of the above viruses were aligned, and percent identities were determined in Clustal W. The phylogenetic tree was inferred via the Neighbor-joining method in MEGA11 with the bootstrap 1,000 replicates.

Construction of plasmids

Total genomic DNA was extracted using cetyltrimethylammonium bromide (CTAB)-based methods as previously described (Springer, 2010). Genomic DNA was purified after RNase A digestion and used as a template for PCR. The complete genome of SCBV was obtained by isothermal amplification with Phi29 MAX DNA Polymerase (Cat#: N106-01; Vazyme, Nanjing, China) according to the manufacturer’s instructions. The products of isothermal amplification were then used as templates for amplification of the complete genome using primer pair SCBV/F and SCBV/SmaI/R (Table S1) and Phanta EVO HS Super-Fidelity DNA Polymerase (Cat#: P504-d1; Vazyme, Nanjing, China). The SCBV complete genome sequence was then subcloned into the pCE2-TA/Blunt-Zero vector (Cat#: C601-01; Vazyme, Nanjing, China) and transformed into Escherichia coli Top10. The positive colonies were cultured and verified by PCR and Sanger sequencing. The complete genome of SCBV was then submitted to the National Center for Biotechnology Information (GenBank accession number: OR672147). The ORF2 coding sequence was amplified by PCR and cloned into PVX-based vector pGR106 by digested with ClaI and SalI followed by T4 DNA ligase (Cat#: EL0014,ThermoFisher Scientific, Shanghai, China) ligation to get the recombinant plasmid PVX-P2. Also, P2 was inserted into the pCHF3 binary expression vector or fused to the N-terminal of enhanced green fluorescent protein (eGFP) in pCHF3-eGFP by double enzyme digestion with SacI/BamHI. The corresponding recombinant plasmids were referred to as pCHF3-P2 and pCHF3-P2-eGFP, respectively. All primers used in this study are listed in Table S1.

Plant growth and agroinoculation

Wild-type, 16c (Voinnet & Baulcombe, 1997), 16-TGS (Raja et al., 2008) and RFP-H2B transgenic seedlings of N. benthamiana were grown to the 4- to 5-leaf stage in an insect-free chamber at a constant temperature of 25 °C and a 16-h/8-h day/night cycle. The binary plasmids pCHF3-P2 and empty pCHF3 were transfected into Agrobacterium tumefaciens strain EHA105, and the recombinant plasmids PVX and PVX-P2 were transformed into A. tumefaciens GV3101 by electroporation. Suspensions of A. tumefaciens cultures were regulated to OD₆₀₀ of 1.0, and the Agrobacterium containing PVX-P2 was infiltrated into N. benthamiana wild-type or 16-TGS seedlings using a 1-mL syringe without a needle. Transient PTGS suppression experiments were conducted as previously described (Johansen & Carrington, 2001; Li et al., 2014, 2015).

Plant transformation

The transgenic N. benthamiana plants over-expressing P2 or empty vector were generated by using the Agrobacterium-mediated leaf disc transformation method. The binary empty vector pCHF3 or recombinant plasmid pCHF3-P2 were transformed into A. tumefaciens strain EHA105 and used for transfection of N. benthamiana leaf discs. Potential transformants were selected on MS media containing 200 μg ml⁻¹ cefotaxime and 200 μg ml⁻¹ kanamycin. Kanamycin-resistant cluster buds were cut off, placed on rooting media, cultured to a height of 5–6 cm, and then transplanted into soil. Transgenic seedlings were verified by PCR with CaMV 35S promoter or P2-specific primers, respectively. Relative levels of P2 mRNA in transgenic plants were confirmed by RT-qPCR.

H₂O₂ detection in plants

H₂O₂ production was detected visually in N. benthamiana leaves using the 3,3′-diaminobenzidine (DAB) staining method (Cat#: A690009; Sangon Biotech, Shanghai, China) (Sharma & Ikegami, 2010) and making some modifications as previously described (Liang et al., 2023).

Subcellular localization analysis

For subcellular localization experiments, fluorescence in RFP-H2B transgenic N. benthamiana leaf epidermal cells or protoplasts inoculated with pCHF3-eGFP and pCHF3-P2-eGFP was examined by confocal microscopy (Leica TCS SP8MP; Leica, Mannheim, Germany) 2- to 3-days post inoculation (dpi) as described (Shen et al., 2011; Yoo, Cho & Sheen, 2007).

Protoplast preparation

To further observe the more precise subcellular localization of SCBV P2 protein, the protoplasts of RFP-H2B transgenic N. benthamiana leaf epidermal cells inoculated with pCHF3-eGFP or pCHF3-P2-eGFP were prepared by digesting with 1.5% (wt/vol) cellulase (Cat#: A002610; Sangon Biotech, Shanghai, China) and 0.4% (wt/vol) macerozyme R10 (Cat#: A004297; Sangon Biotech, Shanghai, China) as described (Yoo, Cho & Sheen, 2007).

DNA methylation analysis by restriction digestion

The genomic DNA of pCHF3 and P2 transgenic plants was extracted using the CTAB method. Digestion analysis of genomic DNA was performed by using a methylation-insensitive restriction endonuclease BamHI (Cat#: FD0054; Thermo Fisher Scientific, Shanghai, China) or a methylation-dependent endonuclease McrBC (Cat#: M0272; New England Biolabs, Ipswich, MA, USA), respectively. The restriction digestion reaction system (50 μL) consists of 10 μg of genomic DNA and 50 U of the respective endonuclease according to the manufacturer’s specifications. Digested products were instantly separated by electrophoresis through a 1.5% agarose gel.

RT-qPCR

RT-qPCR analysis was performed to measure the transcription level of core genes in the RNA-directed DNA methylation (RdDM) pathway. Total RNAs from PVX-inoculated N. benthamiana and P2 or empty vector-transgenic N. benthamiana plants were extracted as experimental and control groups using RNAiso Plus reagent (Cat#: 9108, Takara, Beijing, China) and the A₂₆₀/A₂₈₀ value and concentration of RNA products were measured by NanoPhotometer® N60/N50 (Implen, Munich, Germany). A total of 2 μg of high-quality RNA were converted to cDNA using HiScript III 1st Strand cDNA Synthesis Kit (+gDNA wiper) (Cat#: R312-01; Vazyme, Nanjing, China) according to the manufacturer’s instructions. The cDNA product was diluted 10-fold and served as a template for RT-qPCR. Each reaction mixture contained 2 μL of the diluted cDNA, 10 μL of 2 × ChamQ Universal SYBR qPCR Master Mix (Cat#: Q711-02; Vazyme, Nanjing, China), 0.4 μL of each of the forward and reverse primer (10 μM) in a total volume of 20 μL (primers are listed in Table S1). Three independent biological and experimental replicates were performed, and the reactions were run under the following program conditions: 95 °C for 30 s, 40 cycles of 95 °C for 10 s, and 60 °C for 30 s. Melting curves were derived (95 °C for 15 s, 60 °C for 60 s, and 95 °C for 15 s) for each reaction to ensure a single product. Reactions were performed using the LightCycler 96 real-time PCR system (F. Hoffmann-La Roche Ltd., Basel, Switzerland). The qPCR data was analyzed using the 2^−ΔΔCt method (Livak & Schmittgen, 2001). All primers used for qRT-PCR detection in this study are listed in Table S1.

Immunoblotting

As described previously, the protein of systemic leaves was extracted from PVX-infected plants (Xiong et al., 2009). Immunoblotting was performed with primary Mouse anti-PVX CP monoclonal antibodies, followed by HRP-conjugated Goat Anti-Mouse IgG (Cat#: D110087, Sangon Biotech, Shanghai, China). Blotted membranes were washed thoroughly and visualized using the SuperPico ECL Chemiluminescence Kit according to the manufacturer’s protocol (Cat#: E422; Vazyme, Nanjing, China).

Phylogenetic relationships between SCBV P2 and other taxa of the genus Badnavirus

In this study, five sugarcane samples were collected and detected by RT-PCR; two samples were SCBV positive. One isolate’s complete genome of SCBV was sequenced and submitted to NCBI (GenBank accession number: OR672147). The ORF2 of SCBV is 372 nucleotides (nt) long and encodes a small protein of 123 amino acids (aa) called P2. The complete amino acid sequences of SCBV P2, Rice tungro bacilliform virus (RTBV, GenBank accession number: AAD30189), and several representative species of the genus Badnavirus (as seen in Table 1) were aligned and phylogenetic analyzed. Multiple sequence alignment shows that all P2 amino acid sequences have no obvious sequence similarity (Fig. 1A, Table 1). A coiled-coil like domain (₄₇-LLTLHGKITALEGRLQDLKEDIAKKADK-₇₄) was predicted in SCBV P2 by InterPro online prediction (https://www.ebi.ac.uk/interpro/). The coiled-coil like domain is relatively conserved in length and position in all the above viruses (Fig. 1A). Phylogenetic analysis shows that all the amino acid sequences of P2 proteins from eight individual viruses were clustered into three groups. SCBV P2 was assigned in a clade alone and had a relatively closer relationship with Banana streak OL virus and Cycad leaf necrosis virus P2, and these three viruses were clustered in the same group (Fig. 1B).We have previously demonstrated that SCBV P2 can bind to both homologous and heterologous nucleic acids in a sequence-nonspecific manner, and the coiled-coil like domain plays a vital role in P2-nucleic acids binding through self-interaction (Lou et al., 2023), and this is consistent with that of CSSV and RTBV (Jacquot et al., 1996; Jacquot, Keller & Yot, 1997). The above results suggest that although the badnaviruses P2 share little sequence similarity, they all contain a conserved coiled-coil like domain, which plays an indispensable role in P2-nucleic acids affinity binding. This property is conserved in Badnavirus and Tungrovirus.

Subcellular localization of the SCBV P2 protein

To determine the precise subcellular localization of SCBV P2, an enhanced green fluorescent protein (eGFP) was fused to the C terminus of P2 (P2-eGFP) and subcloned into the binary expression vector pCHF3 under the transcription of the Cauliflower mosaic virus 35S promoter. A. tumefaciens containing the recombinant plasmid pCHF3-P2-eGFP or the pCHF3-eGFP vector were infiltrated into 4- to 5-leaf stage RFP-H2B transgenic N. benthamiana plants (containing an RFP nucleus localization signal), respectively. Green fluorescence in inoculum leaves was observed at 2- to 3-dpi, using confocal microscopy. In plants expressing eGFP alone (35S-eGFP), the fluorescence was observed in the cytoplasm and the nucleus. Similarly, the fluorescence in P2-eGFP-infiltrated leaves was also found in the nucleus as well as cytoplasm (Fig. 2A). To further confirm the subcellular localization of P2, the inoculated leaves were digested with cellulase and macerozyme to obtain the protoplasts. The eGFP signal of 35S-eGFP infiltrated mesophyll protoplasts was distributed around the periphery of the cytomembrane and nucleus, and it is similar in the 35S-P2-eGFP infiltrated mesophyll protoplasts except for some dense bright fluorescent spots (Fig. 2B) The above results indicate P2 have a cytoplasm and nucleus subcellular co-localization.

RNA silencing suppressor activity of SCBV P2

RNA silencing is an efficient innate antiviral mechanism possessed by plants (Li & Wang, 2019). To repress the transcription of viral DNAs, host Dicer-like protein directs TGS through RdDM, or PTGS, which includes splicing and degradation or translational repression of recognized viral RNA (Boualem, Dogimont & Bendahmane, 2016; Matzke & Mosher, 2014). To date, the badnaviruses encoded RNA silencing suppressor (RSS) has been rarely reported. As we described previously that SCBV P2 can bind both DNA and RNA (Lou et al., 2023) and is localized in both the cytoplasm and nucleus, suggesting the possibility that it could function as a viral RSS and support viral infection. To test this hypothesis, we used a 16c-transgenic N. benthamiana line as the experimental plant, which can constitutively express a green fluorescent protein (GFP) signal localized in the endoplasmic reticulum (ER). A. tumefaciens harboring pCHF3-P2 or pCHF3 (negative control) and the p19 suppressor encoded by Tomato bushy stunt virus (TBSV) (positive control) was individually mixed with an equal volume of A. tumefaciens containing a recombinant plasmid expressing the RNA silencing inducer 35S-GFP and inoculated into 4- to 5-leaf aged 16c-transgenic N. benthamiana seedlings. By 4 dpi, the strength of green fluorescence in leaves infiltrated with the pCHF3 empty vector had decreased dramatically under UV light, and was almost undetectable under stereo fluorescence microscope. However, the intensity remained relatively high in patches expressing P2, and those inoculated with p19 had the highest fluorescence intensity (Fig. 3A), these data indicate that SCBV P2 is an RSS, which can suppress single stranded RNA (ssRNA) triggered local gene silencing.

To test whether SCBV P2 can suppress dsRNA-induced gene silencing, wild-type N. benthamiana plants of 4- to 5-leaf age were infiltrated with A. tumefaciens solutions containing the same volume of transient dsRNA elicitor 35S-GFP and 35S-dsFP together with pCHF3, pCHF3-P2, and pCHF3-p19, respectively. As shown in Fig. 3B, leaf spots infiltrated with neither pCHF3 nor pCHF3-P2 mixed with the silencing inducer did not show any GFP fluorescence at 4 dpi under UV light or stereo fluorescence microscopy, indicating that P2 fails to suppress dsRNA-induced RNA silencing. In contrast, a high intensity of green fluorescence was observed in patches expressing the p19 positive control (Fig. 3B). The above data suggest that SCBV P2 is a weak local ss-PTGS but not a ds-PTGS suppressor.

To determine whether P2 can also conquer TGS, a transgenic N. benthamiana line named 16-TGS (the CaMV 35S promoter of the GFP transgene is transcriptionally silenced) was used. Transgenic seedlings of 4- to 5-leaf stage were inoculated with GV3101 A. tumefaciens solutions (mock) or PVX-P2, PVX (negative control), or PVX-βC1 (the TGS repressor of Tomato yellow leaf curl China betasatellite (TYLCCNB), positive control), respectively. After 21 dpi, PVX-infected plants were almost asymptomatic, and no visible green fluorescence was detected in systemic tissues under a high-intensity UV lamp. However, PVX-βC1-infected plants exhibited severe dwarfing, stem deformation, and upward leaf curling along with the PVX-immanent mosaic symptoms, and GFP fluorescence was quite noticeable. As shown in Fig. 3C, systemic leaves infected with PVX-P2 showed severe mosaic and spotty mottling symptoms accompanied by visible green fluorescence under UV light (Fig. 3C), indicating that P2 is a potential TGS repressor.

SCBV P2 expression impacts RdDM signaling in the host

In plants, DNA methylation is a conserved epigenetic modification that regulates genome stability, gene expression, and antiviral defense (He et al., 2022; Wang et al., 2019; Zhang, Lang & Zhu, 2018). Moreover, the methylation level of the plant host is reprogrammed when challenged by an invading virus. To investigate the effect of P2 on the global methylation patterns of the host plant, we constructed a transgenic P2 line of N. benthamiana and tested methylation at the genome level using a methylcytosine-dependent endonuclease (McrBC). Genomic DNA from pCHF3 (Vec) and P2 transgenic N. benthamiana plants (5# and 9#) were extracted and subsequently processed with restriction digestion assays. A mock treatment without enzyme was performed as a control (Sham), and all genomic DNA samples remained unaltered. When a methylation-insensitive endonuclease (BamHI) was used, we found that all three genomic DNA samples were digested and formed a ‘smear’ pattern on the agarose gel during electrophoresis (Fig. 4A). However, McrBC treatment cleaved a portion of the genomic DNA from Vec transgenic lines. In the contrary, the DNA from the P2 transgenic lines showed high resistance and remained unchanged (Fig. 4A). The present data suggest that P2 can reduce DNA methylation on a genome-wide scale when transgenically expressed in N. benthamiana plants.

To explore the possible mechanism of SCBV P2 in repressing epigenetic TGS and genome-wide DNA methylation, the relative expression levels of the homologous genes of DNA methyltransferases, demethylases, histone deacetylase and essential genes related to the RdDM pathway were analyzed in P2 or the empty vector transgenic N. benthamiana plants. Specific primers were designed and synthesized for qRT-PCR detection of the homologs of DNA methyltransferases (MET1, DRM2, and CMT3), demethylases (ROS1, ROS2), argonautes (AGO1, AGO4), dicers (DCL3), and histone deacetylase 6 (HDA6). Total RNA from the empty vector (Vec) and P2 transgenic N. benthamiana plants was extracted 4 weeks after sowing, and total RNA from PVX- or PVX-P2-infected N. benthamiana plants was isolated at 15 dpi. The above RNA was reverse transcribed into cDNA and serves as templates for the subsequent RT-qPCR assays. As shown in Figs. 4B and 4C, the expression of NbAGO1 was significantly down-regulated in both P2 transgenic plants and PVX-P2-infiltrated plants, and the expression of NbAGO4 was dramatically reduced in P2 transgenic N. benthamiana plants but remained inconspicuous in PVX-P2-inoculated plants (Fig. 4B and 4C). Surprisingly, we found that the expression level of NbHDA6 was also significantly down-regulated in both P2 transgenic plants and PVX-P2-infiltrated plants. Further, the HDA6 protein has been proven to act as a histone deacetylase and cofactor of MET1, which stimulates DNA methylation. Taken together, we speculate that constitutive overexpression of SCBV P2 may suppress host TGS by inhibiting the expression of NbAGO1 and NbHDA6.

SCBV P2 induces a hypersensitive-like response (HLR) in N. benthamiana

SCBV infection usually causes patchy, chlorotic streaking symptoms (Viswanathan, Alexander & Garg, 1996) and induces broken chlorotic streaks in inoculated rice (Bouhida, Lockhart & Olszewski, 1993). We successfully constructed an infectious clone of SCBV by inserting a 1.06 copy of the tandem viral genome into the binary vector pCB301-2×35S-HDVRZ-NOS (Figs. S1A and S1B). After Agrobacterium-mediated inoculation, SCBV-infected rice seedlings showed stunting and broken chlorotic streaks on leaves at 21 dpi (Fig. S1C), consistent with the results of Bouhida and colleagues (Bouhida, Lockhart & Olszewski, 1993). We also investigated whether the necrotic streaks were the result of a hypersensitive-like response. The rice leaves that showed typical necrotic streaks were stained with 3,3′-diaminobenzidine (DAB), and the mock and pCB301-2×35S-HDVRZ-NOS empty vector infiltrated plants were used as controls. As shown in Fig. S1, brown necrotic spots appeared in leaves infected with SCBV, whereas plants inoculated with mock and empty vector remained transparent and speckless. The accumulation of viral genomic DNA was verified by PCR analysis, and the results indicate that infectious SCBV can successfully infect rice and cause H₂O₂ accumulation-induced HLR.

To evaluate whether SCBV P2 is a symptom elicitor in N. benthamiana, the P2 protein was ectopically overexpressed by a PVX-based vector. N. benthamiana plants of 4- to 5-leaf stage were inoculated with A. tumefaciens harboring PVX or PVX-P2, respectively. At 7 dpi, PVX-infected plants began to show typical mosaic and shriveling symptoms, whereas PVX-P2-infiltrated plants remained symptomless (Fig. 5A, upper panels). At 10 dpi, the symptoms of PVX-P2-infected plants resembled those of PVX-infected plants (Fig. 5A, middle panels). Symptoms of PVX-infected plants showed signs of recovery from 12 dpi, and vein chlorosis and mosaic phenotype disappeared by 20 dpi, whereas P2-expressing plants still exhibited mosaic and leaf wrinkling phenotypes (Fig. 5A, lower panels). These findings suggest that SCBV P2 can delay the onset of host symptoms to some extent and persist for a considerably longer period of time. To demonstrate the relationship between necrosis and H₂O₂ accumulation, the 3,3′-diaminobenzidine (DAB) staining assay was performed to detect the accumulation of H₂O₂ in empty or PVX-P2 inoculated plants at 10 dpi and 20 dpi. We found that P2-expressing leaves exhibited many brown necrotic spots at 20 dpi, whereas leaves inoculated with the vector remained spotless (Fig. 5B). Protein immunoblotting analyses confirmed that PVX CP accumulated in more significant amounts in PVX-P2-inoculated plants than that in PVX-inoculated plants (Fig. 5C), suggesting that P2 is a potential virulence factor that promotes PVX replication and accumulation. These results indicate that SCBV P2 protein is a viral pathogenicity factor that can induce a hypersensitive-like response and promote virus accumulation.