The serine/threonine/tyrosine kinase STY46 defends against hordeivirus infection by phosphorylating γb protein

Abstract

Protein phosphorylation is a common post-translational modification that frequently occurs during plant–virus interaction. Host protein kinases often regulate virus infectivity and pathogenicity by phosphorylating viral proteins. The Barley stripe mosaic virus (BSMV) γb protein plays versatile roles in virus infection and the coevolutionary arms race between plant defense and viral counter-defense. Here, we identified that the autophosphorylated cytosolic serine/threonine/tyrosine (STY) protein kinase 46 of Nicotiana benthamiana (NbSTY46) phosphorylates and directly interacts with the basic motif domain (aa 19–47) of γb in vitro and in vivo. Overexpression of wild-type NbSTY46, either transiently or transgenically, suppresses BSMV replication and ameliorates viral symptoms, whereas silencing of NbSTY46 leads to increased viral replication and exacerbated symptom. Moreover, the antiviral role of NbSTY46 requires its kinase activity, as the NbSTY46_T436A mutant, lacking kinase activity, not only loses the ability to phosphorylate and interact with γb but also fails to impair BSMV infection when expressed in plants. NbSTY46 could also inhibit the replication of Lychnis ringspot virus, another chloroplast-replicating hordeivirus. In summary, we report a function of the cytosolic kinase STY46 in defending against plant viral infection by phosphorylating a viral protein in addition to its basal function in plant growth, development, and abiotic stress responses.

The cytosolic serine/threonine/tyrosine kinase STY46 negatively regulates Barley stripe mosaic virus replication by phosphorylating and interacting with cb protein.

Introduction

In nature, plants are attacked by various pathogens, including fungi, oomycetes, viruses, and bacteria. Plants have evolved different mechanisms to defend against virus infection, including RNA silencing (Pumplin and Voinnet, 2013; Zhao et al., 2016), RNA decay (Li and Wang, 2019), ubiquitination- and autophagy-mediated degradation (Alcaide-Loridan and Jupin, 2012; Wu et al., 2019; Ismayil et al., 2020), plant hormone signaling pathways (Mazen and Lin, 2015; Collum and Culver, 2016), and resistance (R) gene-mediated defense responses (Chisholm et al., 2006; Jones and Dangl, 2006). Virus recognition is one of the most important steps for plants to combat virus invasion (Wang, 2015). Diverse kinds of host factors have been reported to recognize viral proteins in different steps of virus infection, including virion disassembly, viral genome translation, virus replication complex formation, virus cell-to-cell movement, and systemic viral movement (Wang, 2015).

Protein kinases are a class of proteins often found to promote virus infection in plants by phosphorylating and activating viral proteins. For example, the plant protein kinase CK2 could target diverse viral proteins, including potyvirus coat protein (CP; Ivanov et al., 2003), Bamboo mosaic virus (BaMV) CP (Hung et al., 2014), Tomato mosaic virus (TMV) movement protein (MP; Matsushita et al., 2003), and Barley stripe mosaic virus (BSMV) triple gene block 1 (TGB1; Hu et al., 2015), and thereby promote viral infections. The cAMP-dependent protein kinase (PKA) phosphorylates Beet black scorch virus (BBSV) CP to facilitate assembly and stability of virus particles (Zhao et al., 2015). PKA can also phosphorylate BSMV γb protein and leads to inhibition of viral-induced plant RNA silencing and cell death responses (Zhang et al., 2018). The replication and cell-to-cell movement of BaMV also requires the kinase activity of chloroplast phosphoglycerate kinase (chl-PGK) and the Nicotiana benthamiana Ser/Thr kinase-like protein (NbSTKL; Cheng et al., 2013a, 2013b). There are also examples of the involvement of plant protein kinases in antiviral responses. The SNF1-related protein kinase 1 (SnRK1) could interact with and phosphorylate βC1 protein encoded by the betasatellite of Tomato yellow leaf curl China virus (TYLCCNV), and plays an essential role in plant defense reactions against geminivirus infection (Shen et al., 2011, 2014; Zhong et al., 2017).

The serine/threonine/tyrosine (STY) protein kinases belong to a plant-specific protein kinase family with dual-specificity, which occupies both Ser/Thr and Tyr phosphorylation catalytic domains (Rudrabhatla et al., 2006). Protein kinases from this family consist of conserved N-terminal ACT domains (Asp kinase, Chorismate mutase, and Tyr A prephenate dehydrogenase) which could regulate kinase activity by combining small molecules (Grant, 2006; Eisa et al., 2019), as well as 11 repeats of typical kinase domains in the C-terminal (Lamberti et al., 2011b). The STY protein kinases can autophosphorylate themselves. The STY kinase activity is manganese-dependent and could be inhibited by Tyr kinase inhibitors (Reddy and Rajasekharan, 2006; Lamberti et al., 2011b). There are 57 different STY protein kinases in Arabidopsis thaliana (Rudrabhatla et al., 2006). Among them, AtSTY8, AtSTY17, and AtSTY46 were reported to phosphorylate chloroplast transit peptides and are involved in chloroplast differentiation in Arabidopsis (Lamberti et al., 2011a, 2011b). The sty46 single or sty8/sty46 double mutants in Arabidopsis showed dwarfed growth and abnormal thylakoid formation (Lamberti et al., 2011b). A recent study demonstrated that the STY46 kinase in Arabidopsis promotes the conversion of starch stored in chloroplasts to sugar during stress responses (Dong et al., 2020). However, the role of STY protein kinases in plant antiviral responses is not well understood.

BSMV is the type member of the genus Hordeivirus, whose genomes consist of three positive-stranded RNAs designated RNAα, RNAβ, and RNAγ (Jackson et al., 2009; Jiang et al., 2021). The RNAα encodes the αa replicase, which is essential for virus replication (Petty et al., 1990). The RNAβ encodes four proteins including capsid protein and TGB proteins (TGB1, TGB2, and TGB3), which collectively contribute to virus movement (Petty et al., 1990; Lim et al., 2008). The RNAγ encodes the γa protein, which is the polymerase subunit of the RNA-dependent RNA polymerase (RdRp) complex (Petty et al., 1990), and the sgRNAγ encodes a 17 kDa cysteine-rich protein γb. The γb protein is a multifunctional protein that participates in different stages of virus infection. Its functions include single-stranded viral RNA binding during replication (Donald and Jackson, 1996), suppression of RNA silencing (Yelina et al., 2002; Bragg and Jackson, 2004), inhibition of peroxisomal reactive oxygen species burst (Yang et al., 2018a), subversion of autophagy-mediated antiviral defense (Yang et al., 2018b), mediating systemic virus movement (Yelina et al., 2002), viral pathogenesis (Donald and Jackson, 1994), and facilitating viral transmission through seed (Edwards, 1995). BSMV replicates its RNA genome on the cytosolic side of the chloroplast outer membrane (Zhang et al., 2017; Jin et al., 2018). The γb protein is targeted to the chloroplast by binding to the αa replicase and promotes virus replication (Zhang et al., 2017). In addition, γb protein also enhances ATP-mediated viral movement complex assembly and promotes viral cell-to-cell movement (Jiang et al., 2020). We previously found that γb is phosphorylated by PKA-like kinase for its activity in suppression of RNA silencing and cell death responses (Zhang et al., 2018). The multifunctional role of γb makes it a major player in BSMV infection, but whether γb can be targeted by the host antiviral machinery is currently unknown.

In this study, we demonstrated that the cytosolic protein kinase STY46 of N. benthamiana directly interacts with BSMV γb in vitro and in vivo, and the autophosphorylated NbSTY46 kinase phosphorylates γb. Transient overexpression or constitutive overexpression of NbSTY46 inhibits wild-type (WT) BSMV replication, but not replication of the mutant BSMV lacking γb. Down-regulation of NbSTY46 expression aggravates BSMV symptoms and increases viral accumulation in N. benthamiana. Further experiments utilizing an inactive mutant of NbSTY46 showed that the kinase activity of NbSTY46 is required for its role in antiviral responses. NbSTY46 also inhibits the replication of another hordeivirus. Thus, we propose a new function for NbSTY46 in restricting plant virus infection.

Results

BSMV γb interacts with NbSTY46 in vivo and in vitro

To identify host proteins involved in the BSMV infection cycle, a yeast-two-hybrid (Y2H) screen was performed by using γb protein as a bait to prey possible interaction partners from mixed cDNA libraries of three Nicotiana species (Nicotiana tabacum, N. benthamiana, and N. glutinosa; Shanghai Ziming Biotech. Co., China). The STY protein kinase STY46 in N. tabacum was found to bind to γb. Nicotiana STY46 genes (NtSTY46 or NbSTY46) encode a 565-amino acid protein (63.73 Da). Phylogenetic analysis showed high homology among NbSTY46, NtSTY46, and Arabidopsis AtSTY46 (Rudrabhatla et al., 2006). In addition, the peanut (Arachis hypogaea) STY-like kinase family (Family 1.2), including STY46, STY8, and STY17, formed a single taxon in the phylogenetic tree, suggestive of their close relationship during evolution (Supplemental Figure S1). Similar to AtSTY46, Nicotiana STY46 also encodes the N-terminal ACT domain and the C-terminal repeats of 11 kinase subdomains containing the three conserved motifs (motifs 1–3), which mediate substrate specificity (Lamberti et al., 2011b; Supplemental Figure S2).

In order to verify the results from the previous Y2H screen, the full-length coding region of N. benthamiana STY46 (NbSTY46; https://solgenomics.net, Niben101Scf28650g00001.1) was cloned and inserted into the pGADT7 vector. At the same time, γb was expressed from the pGBKT7 vector. Yeast cell growth and blue-color precipitated within cells transformed with pGAD-NbSTY46 and pGBK-γb, indicating the interaction between NbSTY46 and γb (Figure 1A). To further investigate the interaction between NbSTY46 and γb in vivo, we performed bimolecular fluorescence complementation (BiFC) and co-immunoprecipitation (Co-IP) assays. We fused either the N- or the C-terminal half of yellow fluorescent protein (YFP) to the C-terminus of γb, and fused the C- or N-terminal half of YFP to the C-terminus of NbSTY46. Reconstituted YFP signals were observed when NbSTY46-YFPn and γb-YFPc, or NbSTY46-YFPc and γb-YFPn, were coexpressed in N. benthamiana cells (Supplemental Figure S3, A). Next, we used recombinant virus clones BSMV_γb-YFPc (pCB301-α + pCB301-β + pCB301-γ_γb-YFPc) and BSMV_γb-YFPn (pCB301-α + pCB301-β + pCB301-γ_γb-YFPn), which drive γb-YFPn/c expression in the context of virus infection, and tested for the presence of YFP signals (Supplemental Figure S3, B). When BSMV_γb-YFPc and NbSTY46-YFPn or BSMV_γb-YFPn and NbSTY46-YFPc were coexpressed, YFP signals were observed (Figure 1B). These results suggested that NbSTY46 interacts with γb in planta under virus infection. In addition, immunoprecipitation of γb-3xFlag from total protein lysate of plant leaves coexpressing γb-3xFlag and GFP-NbSTY46 leads to co-precipitation of GFP-NbSTY46 but not GFP (Figure 1C), further demonstrating the interaction between NbSTY46 and γb in vivo.

Figure 1.

Interactions between γb and NbSTY46 in vitro and in vivo. A, Interaction of γb with NbSTY46 in a Y2H system. The combinations of Y2H vectors are shown in the figure. Self-interaction of γb was used as a positive control, and EVs pGAD and pGBK were used as negative controls. B, BiFC assay verifying the interaction between γb and NbSTY46 in the context of BSMV infection in N. benthamiana leaves. NbSTY46-YFPn or NbSTY46-YFPc were coexpressed with BSMV_γb-YFPc, BSMV_γb-YFPn, RbcL-YFPn, or RbcL-YFPc. YFP fluorescence was visualized by confocal microscopy at 3 dpi. Bars = 20 μm. C, Co-IP analysis of the interaction between γb and NbSTY46. The γb-3xFlag and YFP-NbSTY46, or γb-3xFlag and YFP were co-agroinfiltrated in N. benthamiana leaves. Total proteins were extracted at 3 dpi and immunoprecipitated with anti-Flag beads. Input and co-immunoprecipitated proteins were analyzed by western blot with anti-GFP or anti-Flag antibodies. D, GST pull-down identified the interaction between γb and NbSTY46, and the essential region of γb that is responsible for the interaction with NbSTY46. GST-tagged γb protein or its truncated mutants were incubated with His-NbSTY46. After incubation with glutathione-sepharose beads, the eluted proteins were analyzed by immunoblot with anti-His or anti-GST antibodies. His-GFP was used as a negative control.

To determine which region of γb is responsible for its interaction with NbSTY46, different truncations of GST-tagged γb were expressed from Escherichia coli and purified in the presence of His-NbSTY46. We found that His-NbSTY46 could be co-purified with GST-γb, but not with GST-GFP (Figure 1D). Moreover, five truncated mutants of γb (γb_1-85, γb_86-127, γb_1-24, γb_19-47, and γb_60-85; Jackson et al., 2009) were tested for their ability to interact with His-NbSTY46; the results showed that the basic motif (BM, aa 19–47) of γb was sufficient for interaction with NbSTY46 (Figure 1D).

Taken together, we demonstrated that the host protein kinase NbSTY46 directly interacts with BSMV γb in vivo and in vitro and the BM region of γb is required for their interaction.

Autophosphorylated NbSTY46 kinase phosphorylates γb protein

A previous study reported that the Arabidopsis STY protein kinases could autophosphorylate themselves, and the kinase activity was manganese-dependent (Reddy and Rajasekharan, 2006). We purified recombinant His-NbSTY46 protein from E. coli, which was fused with a 6xHis tag on its N-terminus (Supplemental Figure S4), and incubated it with different ions in the presence of adenosine triphosphate labeled on the gamma phosphate group with ³²P. Our data show that His-NbSTY46 could phosphorylate itself in vitro, and the kinase activity strongly depends on Mn²⁺ (Figure 2A). The His-NbSTY46 could also phosphorylate γb recombinant protein in vitro (Figure 2B).

Figure 2.

Autophosphorylated NbSTY46 kinase phosphorylated γb protein. A, An in vitro phosphorylation assay was performed with 2 μg purified NbSTY46 in the presence of Mg²⁺, Mn²⁺, and/or Ca²⁺; the autophosphorylated bands were detected by autoradiography. Coomassie brilliant blue (CBB) staining shows equal amounts of NbSTY46. B, In vitro phosphorylation of His-tagged recombinant γb protein. NbSTY46 was used as a kinase resource. A reaction containing the NbSTY46 kinase and γb protein produced an autoradiography band, and reactions lacking γb protein served as negative controls. CBB-stained γb was used as a loading control. C, The kinase inhibitors JNJ-10198409 and tyrphostin were added to the in vitro kinase assay with increasing concentrations. Constant amounts of 2 μg NbSTY46 kinase and 5 μg γb protein were used. The reaction containing the highest concentration of DMSO served as a control.

The kinase activity of AtSTY8 toward Ser or Thr is inhibited by specific Tyr kinase inhibitors (Lamberti et al., 2011b). We selected the kinase inhibitors JNJ-10198409 and tyrphostin (Cayman) to test their effects on NbSTY46 kinase activity. The solvent dimethyl sulfoxide (DMSO) was used as a negative control. Unlike AtSTY8 that can be inhibited by both chemicals, the autophosphorylation and substrate phosphorylation activity of NbSTY46 was only inhibited by JNJ-10198409, but not tyrphostin (Figure 2C). This difference may be due to minor differences within the protein sequences of AtSTY8 and NbSTY46 (Supplemental Figure S2).

NbSTY46 impairs BSMV replication

To understand whether STY46 functions in BSMV infection, we first detected the expression level of NbSTY46 during virus infection. We observed no significant changes in either the mRNA or the protein level of NbSTY46 upon BSMV infection (Supplemental Figure S5, A and B). The cytosolic localization of NbSTY46 was also not affected by the virus (Supplemental Figure S5, C).

Next, we transiently overexpressed NbSTY46 in leaves of N. benthamiana plants infected with BSMV and detected viral CP or RNA accumulation (Kovalev and Nagy, 2013). We found that compared with the empty vector (EV) control, overexpression of NbSTY46 reduced virus CP or RNA accumulation by ∼50% in plants (Figure 3A). In order to exclude the possible inhibition of NbSTY46 on viral movement, we agroinfiltrated plants with RNAα + RNAγ that lack the RNAβ coding for the CP and the TGB MPs (Zhang et al., 2017; Yang et al., 2018b). The movement-deficient virus also accumulated to a lesser extent when NbSTY46 was transiently overexpressed (Figure 3B). These results indicated that NbSTY46 inhibits BSMV replication.

Figure 3.

NbSTY46 impairs BSMV replication. A, Effects of NbSTY46 overexpression on BSMV protein and viral RNA accumulation. Agrobacterium tumefaciens harboring the EV pMDC32 or pMDC32-NbSTY46-3xFlag was infiltrated into N. benthamiana leaves. Mixtures of A. tumefaciens containing the BSMV infectious clone were infiltrated into the same leaves on the next day. Two days later, total protein and RNA were extracted for western blotting and northern blotting. In the western blot, viral CP and NbSTY46 were detected with anti-CP antibodies and anti-Flag antibodies, respectively. In the northern blot, viral gRNAs were detected with the 3′-UTR-specific probe. The large subunit of Rubisco (RbcL) levels and methylene blue-stained rRNAs are shown as protein and RNA loading controls. B, Effects of NbSTY46 overexpression on BSMV replication. After infiltration of pMDC32 and pMDC32-NbSTY46-3xFlag, the same leaves were agro-inoculated with A. tumefaciens mixtures containing movement-deficient BSMV (RNAα + RNAγ). NbSTY46 and viral γb proteins were detected by western blotting with anti-γb antibodies and anti-Flag antibodies, respectively, and viral RNAα and RNAγ were detected by northern blotting with the 3′-UTR-specific probe. C, Effects of HvSTY46 overexpression on BSMV accumulation. Agrobacterium tumefaciens harboring the EV pMDC32 or pMDC32-HvSTY46-3xFlag was infiltrated into N. benthamiana leaves. Mixtures of A. tumefaciens containing the BSMV infectious clone were infiltrated into the same leaves on the next day. Two days later, total proteins were extracted for western blotting. Anti-CP antibodies and anti-Flag antibodies were used for the detection of the viral CP and HvSTY46, respectively. Rubisco (RbcL) levels were used as loading controls. D, Systemic symptoms of BSMV-infected NT, two NbSTY46-OE lines transgenic N. benthamiana (4c and 8c), and two NbSTY46-KD transgenic N. benthamiana lines (6a and 10a) at 10 dpi. E, Western blot detection of the virus accumulation in Figure 3D with anti-CP antibodies at 10 dpi. Rubisco levels indicate equal loading. F, Effects of NbSTY46 on mutant virus BSMV_mγb accumulation, which contained a substituted γb initiation codon ATG with TTG. Western blot detection of the mutant viral CP accumulation in the inoculated leaves of NT, NbSTY46-OE, and NbSTY46-KD transgenic N. benthamiana plants at 2 dpi. Rubisco levels were used as a loading control. In A, C, E, and F, the viral CP bands were quantified by ImageJ software.

Since barley (Hordeum vulgare) is the natural host of BSMV, we also identified the potential barley ortholog of NbSTY46, named HvSTY46. Phylogenetic analysis showed that HvSTY46 formed a single taxon in the phylogenetic tree with NbSTY46, NtSTY46, AtSTY46, AtSTY8, and AtSTY17, which suggests their close relationship during evolution (Supplemental Figure S1). Sequence alignment showed that similar to AtSTY46 and NbSTY46, HvSTY46 also has the N-terminal ACT domain and the C-terminal repeats of 11 kinase subdomains containing the three conserved motifs (motifs 1–3; Supplemental Figure S2). First, we performed a BiFC assay by using HvSTY46-YFPn and BSMV_γb-YFPc, and found that HvSTY46 also interacts with γb during BSMV infection (Supplemental Figure S6). Furthermore, the transient overexpression experiment suggested that overexpression of HvSTY46 in N. benthamiana plants decreased BSMV CP accumulation to ∼50% (Figure 3C), which indicates that HvSTY46 also inhibits BSMV infection. These results suggested that like NbSTY46, HvSTY46 also interacts with γb protein and inhibits BSMV infection, which implies that STY46 might be a conserved kinase to defend against BSMV infection in both monocots and dicots hosts.

Furthermore, we generated transgenic N. benthamiana plants overexpressing NbSTY46-3xFlag (NbSTY46-OE) or downregulating NbSTY46 (NbSTY46-KD). Compared with non-transgenic (NT) plants, NbSTY46-OE and NbSTY46-KD plants showed no abnormal developmental phenotype (Supplemental Figure S7, A and C). The western blotting analysis confirmed that the NbSTY46-3xFlag protein was overaccumulated in NbSTY46-OE plants, and the endogenous NbSTY46 was silenced in NbSTY46-KD plants (Supplemental Figure S7, B and D). We inoculated BSMV on NT and NbSTY46 transgenic plants. At 10 d post-infection (dpi), the systemically infected leaves were chlorotic and mosaic in NT control plants, whereas in NbSTY46-OE transgenic plants, systemically infected leaves displayed less and minor chlorotic spots in a stochastic pattern. In contrast, almost all of the infected leaves showed remarkable virus infection symptoms in NbSTY46-KD transgenic plants (Figure 3D). The systemically infected leaves were used to measure virus accumulation. The results showed that virus accumulation was significantly less in NbSTY46-OE transgenic plants compared with that in the WT plants, while BSMV accumulation was ∼30%–50% higher in NbSTY46-KD plants (Figure 3E).

Since γb can interact with NbSTY46, we sought to test whether γb is the target of NbSTY46 for its inhibition in BSMV replication. We used a BSMV infectious cDNA clone BSMV_mγb that has a mutation on the γb initiation codon ATG (Zhang et al., 2017; Yang et al., 2018b). We found that NbSTY46 transgenic overexpression or silencing had no or only marginal effects on BSMV accumulation (Figure 3F). Thus, we conclude that NbSTY46 negatively regulates BSMV infection through interacting with γb.

The kinase activity of NbSTY46 is essential for inhibiting virus accumulations

The kinase domain of NbSTY46 is located on its C-terminus, while the N-terminal ACT domain allosterically regulates the kinase activity (Lamberti et al., 2011b; Figure 4B). We tested whether the kinase activity of NbSTY46 is required for its inhibition of BSMV replication.

Figure 4.

Kinase activity of NbSTY46 is essential for inhibiting BSMV infection. A, In vitro phosphorylation assay of purified γb with NbSTY46, and its truncated mutants NbSTY46_N (aa 1–269) and NbSTY46_C (aa 270–565); the autophosphorylated bands of NbSTY46 and phosphorylated bands of γb are indicated in the picture. CBB-stained NbSTY46 and γb were used as loading controls. B, Schematic representation of NbSTY46 protein. The ACT domain in NbSTY46_N (aa 1–269) and the kinase domain in NbSTY46_C (aa 270–565). A conserved Thr in the C-terminal (position 436 in NbSTY46) is highlighted, which was substituted by an Ala in NbSTY46 and leads to complete loss of kinase activity. C, Transient overexpression experiment of NbSTY46, NbSTY46_N, and NbSTY46_C in N. benthamiana. Viral CP proteins were evaluated with anti-CP antibodies, and the accumulation of NbSTY46 and its mutants was detected with anti-Flag antibodies. CBB-stained Rubisco (RbcL) was used as a loading control. D, In vitro phosphorylation assay of purified γb with NbSTY46 and NbSTY46_T436A. The autophosphorylated bands of NbSTY46 and phosphorylated bands of γb were detected by autoradiography. CBB-stained NbSTY46 and γb were used as loading controls. E, GST pull-down assay verifying the interaction of γb and NbSTY46 or NbSTY46_T436A. GST-tagged γb was incubated with His-NbSTY46 or His-NbSTY46_T436A. After incubation with glutathione-sepharose beads, the eluted proteins were analyzed by western blotting with anti-His and anti-GST antibodies. GST-GFP and His-GFP served as negative controls. F, Transient overexpression experiment of NbSTY46 and NbSTY46_T436A in N. benthamiana. Viral CP proteins and NbSTY46 were detected with anti-CP antibodies or anti-Flag antibodies. CBB-stained RbcL was used as a loading control. In C and F, the viral CP bands were quantified by ImageJ software.

First, we purified two versions of truncated NbSTY46 recombinant protein containing the ACT domain (NbSTY46_N, aa 1–269) or the kinase domain (NbSTY46_C, aa 270–565), and tested their in vitro kinase activity (Figure 4A). We found that NbSTY46_C had reduced kinase activity, and NbSTY46_N had no kinase activity. When these two proteins were overexpressed during BSMV replication, only NbSTY46_C suppressed viral accumulation (Figure 4C), suggesting that the inhibitory effect of NbSTY46 is associated with its kinase activity.

To confirm this result, we explored a kinase-inactive mutant of NbSTY46. Previous studies have shown that the conserved threonine (position Thr-439 in AtSTY8, Thr-445 in AtSTY17, and Thr-443 in AtSTY46) in the activation domain is essential for the kinase activities (Lamberti et al., 2011b). By aligning the Nicotiana STY46 protein sequences with the Arabidopsis homologs, we found that the conserved Thr of NbSTY46 is at position aa 436, and the Motif 2 within the Thr-436 site is identical between two Nicotiana STY46s and AtSTY46 (Supplemental Figure S2). We then substituted the threonine (T) with alanine (A) to construct the NbSTY46_T436A mutant. The in vitro kinase assay showed that NbSTY46_T436A could not phosphorylate itself and γb (Figure 4D). Further a GST pull-down assay showed that the NbSTY46_T436A mutant could not interact with γb protein either (Figure 4E). Then we tested the effect of NbSTY46_T436A on inhibition of BSMV replication. We found that overexpression of NbSTY46_T436A could not inhibit, but slightly increased BSMV accumulation (Figure 4F). It is possible that NbSTY46_T436A is dominant-negative toward the WT NbSTY46 function in BSMV replication. These results suggested that the kinase activity of NbSTY46 is required for the inhibition of BSMV replication.

NbSTY46 has no discernable effect on the VSR activity of γb

As a viral suppressor of RNA silencing (VSR), BSMV γb protein promotes virus pathogenicity and long-distance movement (Yelina et al., 2002; Bragg and Jackson, 2004). To determine the effects of NbSTY46 on γb VSR activity, we measured whether NbSTY46 can interfere with the suppression of positive-sense GFP (sGFP)-induced RNA silencing by γb (Dong et al., 2016; Zhang et al., 2018). NbSTY46 and its mutants (NbSTY46_N, NbSTY46_C, and NbSTY46_T436A in Figure 4B) were co-infiltrated with pGD-sGFP and pGD-γb-3xFlag. At 5 dpi, we observed that the intensity of the GFP fluorescence, as well as GFP protein accumulation, showed no difference between the EV control treatment and the leaf areas overexpressing NbSTY46 or its mutants (Figure 5A and 5B).

Figure 5.

NbSTY46 has no effect on the VSR activity of γb protein. A, Infiltrated N. benthamiana leaves coexpressing sGFP, γb, and NbSTY46-Myc or its mutants NbSTY46_T436A-Myc, NbSTY46_N-Myc, or NbSTY46_C-Myc were imaged under long-wave UV light at 5 dpi. The EV pGD, which co-expressed sGFP, γb-3xFlag, and EV, was used as a positive control. B, Western blot of GFP protein from co-infiltrated leaves at 5 dpi. In the western blot, anti-GFP, anti-Flag, and anti-Myc antibodies were used to detect the accumulation of GFP, γb, and NbSTY46 variants, respectively. CBB-stained RbcL was used as a loading control. C, The sGFP and γb were coinfiltrated into NT, NbSTY46-OE, and NbSTY46-KD transgenic N. benthamiana plants. The agroinfiltrated leaves were imaged under long-wave UV light at 5 dpi. The sGFP infiltrated NT plant was used as a negative control. D, Western blot analysis of GFP and γb protein accumulation in co-infiltrated leaves at 5 dpi with anti-GFP or anti-Flag antibodies. CBB-stained RbcL was used as a loading control.

Furthermore, we co-infiltrated pGD-sGFP and pGD-γb-3xFlag in NT, NbSTY46-OE, and NbSTY46-KD plants. GFP fluorescence and GFP accumulation were detected at 5 dpi. We found no observable difference in green fluorescence or GFP level among NT, NbSTY46-OE, and NbSTY46-KD plants (Figure 5C and 5D). These results demonstrated that NbSTY46 has no noticeable effect on the VSR activity of γb.

Identification of γb phosphorylation sites by NbSTY46

Kinase activity is important for the function of STY-like kinases. We have demonstrated that the kinase activity of NbSTY46 is essential for the suppression of BSMV infection (Figure 4). As NbSTY46 can phosphorylate γb (Figure 2B), we next identified the potential phosphorylation sites of γb by in vitro kinase assay and liquid chromatography/tandem mass spectrometry (LC–MS/MS) identification. Six potential phosphorylation sites in γb were identified (Figure 6A), including S30, S137, S141, S148, S149, and S151. As S136 is adjacent to S137, and S147 is adjacent to S148/S149, to exclude the influence of alternative phosphorylation on neighboring amino acids, we then mutated these serine residues to alanines (A), and obtained five γb derivatives S30A, S136/137A, S141A, 3SA (S147/148/149A triple mutant), and S151A. An in vitro phosphorylation assay showed that mutations at these sites could decrease the phosphorylation level of γb to varying degrees (Figure 6B left panel). To further explore whether these amino acids affect γb phosphorylation synergistically, we substituted these eight sites (S30, S136, S137, S141, S147, S148, S149, and S151) with alanines (A), named 8SA, and an in vitro kinase assay showed that the phosphorylation level of 8SA was decreased dramatically (Figure 6B right panel), which indicated that these sites might function synergistically.

Figure 6.

Screening the phosphorylation sites of γb by NbSTY46. A, LC–MS/MS results: the amino acids covered in LC–MS/MS are shown in bold. The identified potential phosphorylation sites are shown in red. The underlined amino acids were used for subsequent research. B, In vitro kinase assay of γb and its mutants. The autoradiographic bands were quantified by ImageJ software. CBB-strained γb and its derivatives were used as a loading control. 3SA indicates the S147A/S147A/S149A triple mutant. The 8SA mutant indicated γb derivatives by substituting all eight sites (S30, S136, S137, S141, S147, S148, S149, and S151) with alanines. C, Left panel: infiltrated N. benthamiana leaves coexpressing sGFP and γb or the 8SA mutant were imaged under long-wave UV light at 5 dpi. Right panel: western blot of GFP protein from co-infiltrated leaves at 5 dpi. Anti-GFP and anti-Flag antibodies were used to detect the accumulation of GFP and γb, respectively. CBB-stained RbcL are shown as loading controls. The pGD EV was used as a negative control. D, Upper panel: systemic symptoms of N. benthamiana plants induced by BSMV and BSMV_8SA at 10 dpi. Bottom panel: accumulation of BSMV and the mutant virus BSMV_8SA in inoculated leaves and systemically infected leaves of N. benthamiana. Viral accumulations were detected with anti-TGB1 and anti-CP antibodies. CBB-stained RbcL was used as a loading control. E, Left panel: systemic symptoms of barley induced by BSMV and BSMV_8SA at 14 dpi. Right panel: western blot detection of the virus accumulation in systemically infected leaves with anti-TGB1 and anti-CP antibodies. Rubisco levels indicate equal loading. In D and E, the viral TGB1 and CP bands were quantified by ImageJ software.

A transient expression assay showed that the 8SA mutant did not influence the VSR activity of γb (Figure 6C). To explore whether these amino acids affect the function of γb during virus infection, we firstly detected the viral accumulation of BSMV and its derivatives (BSMV_S30A, BSMV_S136/137A, BSMV_S141A, BSMV_3SA, and BSMV_S151A) via western blot in N. benthamiana; unfortunately, these mutant viruses had no obvious effect on virus infection (Supplemental Figure S8). Then, we infiltrated Agrobacterium tumefaciens containing WT BSMV and the BSMV_8SA mutant virus into N. benthamiana plants. At 10 dpi, the BSMV_8SA mutant caused severe viral symptoms in systemically infected leaves (Figure 6D upper panel), and western blot suggested that compared with WT BSMV, BSMV_8SA had more viral TGB1 and CP accumulation (Figure 6D bottom panel). Moreover, we also inoculated BSMV and BSMV_8SA on barley, the natural host of BSMV. Consistent with the result obtained from N. benthamiana, at 14 dpi, the BSMV_8SA mutant also caused severe viral symptoms in systemically infected leaves and had more viral protein (TGB1 and CP) accumulation (Figure 6E). These results imply that these eight phosphorylation sites, which are phosphorylated by STY46, might function redundantly in regulating virus infection.

NbSTY46 inhibits another chloroplast-replicating hordeivirus replication

Since γb is a conserved hordeiviral protein, we speculated that NbSTY46 could inhibit replication of another hordeiviruses through interaction with γb. Lychnis ringspot virus (LRSV) is another member of the genus Hordeivirus that infects N. benthamiana plants and replicates on the chloroplast (Jiang et al., 2018). First, we observed that NbSTY46 could interact with LRSV γb in vivo in a BiFC assay (Figure 7A). To test the effect of NbSTY46 on LRSV infection, LRSV was then inoculated onto NbSTY46-OE and NbSTY46-KD plants via agroinfiltration. The viral disease symptoms and viral accumulation were attenuated in NbSTY46-OE plants compared with those from LRSV-infected NT plants, while the symptom severity and the level of viral accumulation were both enhanced in NbSTY46-KD plants (Figure 7B and 7C). These results suggested that the NbSTY46 protein kinase is also an antiviral factor for replication of another hordeivirus LRSV.

Figure 7.

NbSTY46 inhibits other chloroplast-replicating hordeivirus replication. A, BiFC assay verifying the interaction between NbSTY46 and γb_BSMV and γb_LRSV in N. benthamiana leaves. YFP fluorescence was visualized by confocal microscopy at 3 dpi. Bars = 100 μm. B, Systemic symptoms of LRSV-infected NT, NbSTY46-OE, and NbSTY46-KD transgenic N. benthamiana plants at 10 dpi. C, Western blot detection of the virus accumulation in Figure 7B with anti-γb_LRSV antibodies at 10 dpi. Rubisco levels indicate equal loading. D and E, Effects of NbSTY46 on CMV or TuMV infection in the transient overexpression experiment. Viral accumulations were detected with anti-CMV CP antibodies (D) or anti-GFP antibodies (E). The anti-Flag antibodies were used to detect NbSTY46 expression. CBB-stained RbcL was used as a loading control. F, Effect of NbSTY46 on CMV or TuMV infection in NbSTY46 transgenic plants. CMV-GFP and TuMV-GFP were agroinfiltrated into NT, NbSTY46-OE, and NbSTY46-KD transgenic N. benthamiana plants. The GFP fluorescence of systemically infected leaves was photographed under long-wave UV light at 7 dpi.

We wondered if the antiviral role of NbSTY46 is hordeivirus-specific or can be extended to other virus families. We then tested the effect of NbSTY46 overexpression on the replication of a cucumovirus Cucumber mosaic virus (CMV) or a potyvirus Turnip mosaic virus (TuMV; Cillo et al., 2002; Beauchemin et al., 2007). We found that transient overexpression of NbSTY46 had no effect on the viral accumulation of CMV-GFP (see the “Materials and methods” section) or TuMV-GFP (Lellis et al., 2002; Figure 7D and 7E). To confirm this result, we inoculated CMV-GFP or TuMV-GFP onto NbSTY46-OE and NbSTY46-KD transgenic plants. The viral replication and movement were observed by the green fluorescence expressed from CMV-GFP and TuMV-GFP. These results indicated that compared with virus-infected WT N. benthamiana plants, the viral infection is not altered in NbSTY46-OE and NbSTY46-KD plants inoculated with either CMV-GFP or TuMV-GFP (Figure 7F). Taken together, these results showed that NbSTY46 negatively regulates the infection of chloroplast-replicating hordeiviruses.

Discussion

Reversible phosphorylation has been extensively studied in many kinds of post-translational modifications (PTMs), which plays a fundamental role in the regulation of protein activities. Several studies on diverged classes of viruses showed the relevance of the viral phosphorylation process in the establishment of infections (Jakubiec and Jupin, 2007). Our previous studies verified that hordeiviral proteins, including MP TGB1 and the multifunctional protein γb, are substrates of host protein kinases (Hu et al., 2015; Zhang et al., 2018). BSMV γb is phosphorylated at Ser-96 by PKA-like kinase to counteract both the host RNAi-based antiviral defense and the cell death response, allowing a dynamic balance between viral replication and its survival in the host plants (Zhang et al., 2018). However, plants and viruses have co-evolved, and our understanding of the mechanism underlying plant defense against viruses is far from complete. Here, we found that the NbSTY46 kinase interacts with and phosphorylates the viral γb protein in N. benthamiana (Figures 1 and 2B), and impairs replication of chloroplast-replicating hordeiviruses in a kinase activity-dependent manner (Figure 4), which might be a counter-counter-defense strategy employed by plants to defend from virus infection.

Many studies have reported the requirement of viral protein phosphorylation for optimal virus infection (Jakubiec et al., 2006). However, the inhibition effect of plant protein kinases in virus infection was only reported for the AMP-activated serine/threonine-protein kinase SnRK1. SnRK1 recognizes old-world geminivirus through recognizing βC1 carried by the β satellite (Shen et al., 2011; Shen et al., 2014; Zhong et al., 2017), and also restricts new-world geminivirus infection through interacting with and phosphorylating the viral protein AL2/C2 (Shen et al., 2014). Here, we provided another example of a protein kinase (NbSTY46) that recognizes and phosphorylates viral proteins. Plant STY8/17/46 are autophosphorylated and self-activated at conserved Thr residue at the domain VIII (Lamberti et al., 2011b). Different from SnRK1 that does not require its kinase domain to interact with TYLCCNB βC1, the kinase-inactive mutant NbSTY46_T436A neither interacts with γb (Figure 4E) nor inhibits BSMV replication (Figure 4F). These results indicate that the NbSTY46–γb interaction and the phosphorylation of γb by NbSTY46 are mutually dependent or functionally indistinguishable. Moreover, as a VSR protein, phosphorylation of βC1 weakens its suppression activity of RNA silencing (Shen et al., 2011, 2014; Zhong et al., 2017). Unlike βC1, NbSTY46-mediated phosphorylation of γb had no effect on its VSR activity (Figure 5). However, our results indicated that NbSTY46 inhibits BSMV infection at the virus replication level (Figure 3B).

Previous studies have shown that phosphorylation at different sites of the same protein might have different functions. For instance, Turnip yellow mosaic virus (TYMV) replicase 66K was phosphorylated at both the N-terminal and C-terminal. The phosphorylation of Thr-64 and Ser-80 at the N-terminal promotes the degradation of 66K protein, while the phosphorylation of Ser-326 at the C-terminal affects the assembly of the viral replication complex and viral RNA (Jakubiec et al., 2006). We previously showed that γb is phosphorylated at Ser-96 by PKA-like kinase to suppress both host RNA silencing and cell death response (Zhang et al., 2018). Here, we showed that the host cytosolic kinase STY46 could phosphorylate γb protein at other phosphorylation sites to inhibit BSMV infection. These results demonstrated that a viral protein could be phosphorylated at non-overlapping sites by different protein kinases, resulting in distinct infection phenotypes during the interplay between the plant and the virus.

In conclusion, we identified that the protein kinase NbSTY46 in N. benthamiana interacts with and phosphorylates viral γb protein, and negatively regulates BSMV replication in a kinase activity-dependent manner. Overall, the data obtained in this study broaden our understanding of the functions of host STY46 protein kinase in plant–virus interactions besides its basal functions in plant growth, development, and abiotic stress response, and provide new insight into the multifunctional roles of phosphorylated γb protein during virus infection.

Materials and methods

Plant growth conditions

The WT and transgenic N. benthamiana plants were grown in a climate chamber at 23°C–25°C under a 14-h light/10-h dark photoperiod (Zhang et al., 2018).

Plasmid construction

The plasmids and corresponding primers used in this study are listed in Supplemental Table S1. All of the constructs described below were verified by DNA sequencing. The infectious BSMV agroinfiltration clones used in this study were described in our previous study (Yuan et al., 2011).

For Y2H and BiFC assays, the coding sequence of NbSTY46 was amplified from the N. benthamiana cDNA and inserted into EcoR1 and BamH1 sites of the pGBK vector or XbaI and BamHI sites of the pSPYNE-35S or pSPYCE-35S vectors (Walter et al., 2004). The pGAD-γb and pSPYNE-35S-γb or pSPYCE-35S-γb plasmids were described previously (Zhang et al., 2017).

For protein purification, the fragments of full-length NbSTY46, NbSTY46_N (aa 1–269), and NbSTY46_C (aa 270–565) were amplified from pGBK-NbSTY46 and cloned into the pET-30a(+) vector between NdeI and XhoI sites to generate pET-30a-NbSTY46, pET-30a-NbSTY46_N, and pET-30a-NbSTY46_C plasmids. Site-specific mutant plasmids pET-30a-NbSTY46_T436A were constructed by QuikChange Site-Directed Mutagenesis kit using corresponding primers (Supplemental Table S1). The pGEX-2T-γb vector was used as described previously (Zhang et al., 2017). The fragments of γb_1-85, γb_86-152, γb_1-24, γb_19-47, and γb_60-85 were amplified from pCaBS-γ and inserted into BamH1 and Xho1 sites of the pGEX-KG vector (Guan and Dixon, 1991).

To generate overexpression vectors, the coding sequences of NbSTY46, NbSTY46_N, NbSTY46_C, and NbSTY46_T436A were amplified and inserted into the Kpn1 and Spe1 sites of the pMDC32-3xFlag vector. For VSR detection vectors, the NbSTY46, NbSTY46_T436A, NbSTY46_N, and NbSTY46_C fragments were digested with BglII/Sal1 and cloned into the pGD-6xMyc vector. For subcellular localization and Co-IP assays, the full-length of NbSTY46 was cloned into the pGDG vector digested with XhoI/BamHI. The pGD-γb-3xFlag vector was used as described previously (Zhang et al., 2018).

For CMV-GFP, the infectious clone of CMV-RNA2-GFP was constructed according to a previously described method (Krenz et al., 2015). The CMV-RNA2 fragment, which lacks 30 amino acids at the C-terminal of 2b protein, was amplified from pCB301-CMV-RNA2 with the primers CMV-RNA2-2751-F(TGA)/CMV-RNA2-2661-R, and the sGFP fragment was amplified from pGD-sGFP with the primers sGFP-F/sGFP-R. Then two fragments were homologously recombined by using the one-step clone method following the manufacturer’s protocol (Vazyme, C112-02-AB) to generate the CMV-RNA2-GFP infectious clone.

Phylogenetic analysis

Phylogenetic analysis of NbSTY46, NtSTY46, and HvSTY46 with 57 different STY protein kinases from A. thaliana was performed by MEGA6 software with neighbor-joining method.

Y2H assay

The Y2H assay was performed as described previously (Zhang et al., 2017). Briefly, the pGAD-NbSTY46 and pGBK-γb vectors were transformed into yeast strain AH109 and Y187, respectively, then mated for 24 h at 30°C on a shaker at 50 rpm. The yeast cells were plated onto supplement dropout medium lacking adenine, histidine, leucine, and tryptophan (SD/-Ade/-His/-Leu/-Trp) to check for interactions between these two-hybrid proteins.

Bimolecular fluorescence complementation

The BiFC assay was performed as described previously (Zhang et al., 2018). All the pSPYNE-35S or pSPYCE-35S plasmids were transformed into A. tumefaciens strain EHA105 and co-infiltrated into N. benthamiana leaves. The infiltrated leaves were observed at 3 dpi by using an Olympus confocal FV1000 microscope. All images show GFP fluorescence recorded with λ = 488 nm.

Co-immunoprecipitation

The A. tumefaciens harboring pGD-γb-3xFlag and pGDG-NbSTY46, or pGD-γb-3xFlag and the pGDG EV were co-infiltrated into N. benthamiana leaves (Goodin et al., 2002). Total proteins were extracted from co-infiltrated leaves at 3 dpi by an extraction buffer (GTEN buffer [25 mM Tris/HCl pH 7.5, 1 mM EDTA, 150 mM NaCl, and 10% (v/v) glycerol] with 0.1% (v/v) NP40, 2% (w/v) polyvinylpolypyrrolidone, 10 mM DTT, and 1% (v/v) protease inhibitor cocktail). The homogenate was placed on ice for 30 min and centrifuged for 15 min at 8,000 × g at 4°C. Then the supernatant was centrifuged one more time. The final supernatant was incubated with Flag beads (Sigma) for 4 h at 4°C. The co-immunoprecipitated proteins were washed five times with IP buffer [GTEN buffer with 0.1% (v/v) NP-40] and subjected to western blot analysis with anti-Flag or anti-GFP antibodies.

Protein purification and pull-down assays

All plasmids used for protein purification were transformed into E. coli strain BL21 (DE3). His-tagged proteins were purified as described previously (Zhang et al., 2017). For purification of GST-fused proteins, the supernatants were passed through a glutathione–sepharose affinity column (GE Healthcare). The GST-fused proteins were eluted with T buffer [20 mM Tris–HCl, pH 9.0, 500 mM NaCl, 0.1% Triton X-100 (v/v), 10% glycerol (v/v), and 1 mM PMSF] containing 30 mM l-glutathione and concentrated with 30 kDa Amicon-Ultra-15 filters (Millipore).

For pull-down assays, about 10 µg of bait GST-fused proteins were incubated with about 5 µg prey His-tagged proteins with GST beads (GE Healthcare) at 4°C for 3–4 h in pull-down buffer [20 mM Tris–HCl, pH 7.5, 250 mM NaCl, 0.2% (v/v) glycerol, 0.25% (v/v) Ttriton-X 100, and 100 mM PSMF]. The samples were washed with pull-down buffer three to five times and boiled for 10 min followed by western blotting with anti-GST or anti-His antibodies.

In vitro phosphorylation assays

The in vitro kinase assays were carried out according to the protocol described previously with a minor revisions (Lamberti et al., 2011b). Briefly, 5 μg purified γb proteins were reacted with 2 μg NbSTY46 kinase in 1× reaction buffer (25 mM Tris–HCl pH 7.5, 10 mM MgCl₂, and 10 mM MnCl₂) with 5 μCi [γ-³²P] ATP (Perkin Elmer) and incubated for 30 min at 30°C. For the inhibition studies, the inhibitors JNJ-10198409 and tyrphostin (Cayman) were added to the reaction buffer at the indicated concentrations (Lamberti et al., 2011b). The reactions were terminated by the addition of 2× SDS sample buffer [100 mM Tris–HCl, pH 6.8, 4% SDS (v/v), 20% glycerol (v/v), 0.2% bromophenol blue (v/v), and 5% β-mercaptoethanol (v/v)] and boiled for 10 min. The samples were separated on 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS–PAGE). The gels were dried with a Model 583 Gel Dryer (Bio-Rad) and detected by autoradiography.

Virus inoculation and detection

For viral inoculation, A. tumefaciens EHA105 containing plasmids pCaBS-α, pCaBS-β, and pCaBS-γ were mixed at a 1:1:1 ratio and infiltrated into the leaves of 3–4-week-old plants (Yuan et al., 2011). Total proteins and RNAs were extracted from infiltrated leaves at 2 dpi to detect the accumulation of viral proteins and RNAs. In western blots, anti-BSMV CP antibodies were used to detect the viral protein accumulation. Northern blots were performed as described previously (Zhang et al., 2017). 3′-UTR-specific probe labeled by [α-³²P] UTP was used to detect the viral RNA accumulation.

Transient overexpression experiment in N. benthamiana

A transient overexpression experiment was performed as described previously (Kovalev and Nagy, 2013) with minor modifications. N. benthamiana leaves were infiltrated with A. tumefaciens cultures carrying plasmid pMDC32-3xFlag, or pMDC32-NbSTY46-3xFlag, pMDC32-NbSTY46_T436A-3xFlag, pMDC32-NbSTY46_N-3xFlag, and pMDC32-NbSTY46_C-3xFlag. The leaves were also co-agroinfiltrated with Tomato bushy stunt virus (TBSV) p19 for suppressing gene silencing (Lakatos et al., 2004). On the next day, the same leaves were infiltrated with A. tumefaciens cultures carrying pCaBS-α, pCaBS-β, and pCaBS-γ to initiate BSMV infection. Agrobacterium tumefaciens strains were at a concentration of 600 nm (OD₆₀₀): 0.2 for p19; 0.5 for pMDC32-NbSTY46 or its mutants; and 0.2 for pCaBS-α, pCaBS-β, and pCaBS-γ. Samples were collected at 2 dpi after BSMV infection.

Generation of NbSTY46 transgenic N. benthamiana plants

The pMDC32-NbSTY46-3xFlag and pMDC32-NbSTY46i plasmids were transformed into A. tumefaciens strain EHA105. Leaf disc transformation was used to generate the N. benthamiana plants (Horsch et al., 1989). After regeneration of leaf explants, total proteins were harvested and subjected to western blotting with anti-Flag or anti-NbSTY46 antibodies to screen the positive transgenic plants.

RNA silencing suppression experiments

VSR detection was performed as described previously (Dong et al., 2016; Zhang et al., 2017, 2018). Equal volumes of A. tumefaciens cultures harboring plasmids pGD-sGFP, pGD-γb-3xFlag, and pGD-NbSTY46-6xMyc or its mutants were mixed and co-infiltrated into 4–5-week-old N. benthamiana leaves. The agroinfiltrated leaves were illuminated under a long-wavelength UV lamp (UVP, Upland, USA) and photographed with a Canon digital camera (Canon, Tokyo, Japan) under a yellow filter at 5 dpi.

Statistical analysis

The intensity of the protein bands detected by western blot was analyzed by ImageJ software, the data shown are the means of three independent experiments. The RT-qPCT result (Supplemental Figure S5, A) was analyzed by Student’s t test (n.s. indicated no significant differences).

Accession numbers

Sequence data from this work can be found in Genebank/EMBL libraries, Solanaceae Genomics Network (https://solgenomics.net), or The Arabidopsis Information Resource (www.Arabidopsis.org) under the following accession numbers: NbSTY46 (Niben101Scf28650g00001.1), NtSTY46 (XM_016612812.1), HvSTY46 (AK357173.1), AtSTY46 (AT4G38470), AtSTY8 (AT2G17700), and AtSTY17 (AT4G35780).

Supplemental data

Supplemental Figure S1. Phylogenetic analysis of STY protein kinases from N. benthamiana (NbSTY46), N. tabacum (NtSTY46), and H. vulgare (HvSTY46) with 57 different STY protein kinases from A. thaliana.

Supplemental Figure S2. Sequence alignment between NbSTY46, NtSTY46, HvSTY46, and STY8, STY17, and STY46 in Arabidopsis.

Supplemental Figure S3. BiFC analysis of the interactions between γb and NbSTY46.

Supplemental Figure S4. The purification of NbSTY46 protein from E. coli.

Supplemental Figure S5. BSMV infection has no effect on the expression level and subcellular localization of NbSTY46.

Supplemental Figure S6. BiFC analysis of the interactions between γb and HvSTY46.

Supplemental Figure S7. Phenotype and identification of NbSTY46-OE and NbSTY46-KD transgenic N. benthamiana plants.

Supplemental Figure S8. Viral accumulation of BSMV and its mutants in inoculated N. benthamiana leaves.

Supplemental Table S1. Primers for plasmid constructions and RT-qPCR analysis.

 

D.L. and X.Z. conceived and designed the experiments. X.Z. performed most of the experiments with the assistance of X.W., Z.J., K.D., X.X., and N.Y. K.X., Y.Z., X.-B.W., C.H., and J.Y. analyzed the data. X.Z., D.L., K.X., Y.Z., and Z.J. wrote the manuscript. All the authors read and approved the manuscript.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is: Dawei Li (Dawei.Li@cau.edu.cn).