Rhizoctonia solani transcriptional activator interacts with rice WRKY53 and grassy tiller 1 to activate SWEET transporters for nutrition

Graphical abstract

Highlights

• •Heterotroph pathogen R. solani hyphal growth is enhanced by high sugar concentrations.
• •AOS2 secreted by R. solani, GT1 and WRKY53 form the transcriptional complex to activate SWEETs and excrete hexose significantly.
• •Hexose transporter gene SWEET2a and SWEET3a negatively regulate rice resistance to ShB.

Abstract

Introduction

Rhizoctonia solani, the causative agent of the sheath blight disease (ShB), invades rice to obtain nutrients, especially sugars; however, the molecular mechanism via which R. solani hijacks sugars from rice remains unclear.

Objectives

In this study, rice-R. solani interaction model was used to explore whether pathogen effector proteins affect plant sugar absorption during infection.

Methods

Yeast one-hybrid assay was used to identify Activator of SWEET2a (AOS2) from R. solani. Localization and invertase secretion assays showed that nuclear localization and secreted function of AOS2. Hexose transport assays verified the hexose transporter activity of SWEET2a and SWEET3a. Yeast two-hybrid assays, Bimolecular fluorescence complementation (BiFC) and transactivation assay were conducted to verify the AOS2-WRKY53-Grassy tiller 1 (GT1) transcriptional complex and its activation of SWEET2a and SWEET3a. Genetic analysis is used to detect the response of GT1, WRKY53, SWEET2a, and SWEET3a to ShB infestation. Also, the soluble sugar contents were measured in the mutants and overexpression plants before and after the inoculation of R. solani.

Results

The present study found that R. solani protein AOS2 activates rice SWEET2a and localized in the nucleus of tobacco cells and secreted in yeast. AOS2 interacts with rice transcription factor WRKY53 and GT1 to form a complex that activates the hexose transporter gene SWEET2a and SWEET3a and negatively regulate rice resistance to ShB.

Conclusion

These data collectively suggest that AOS2 secreted by R. solani interacts with rice WRKY53 and GT1 to form a transcriptional complex that activates SWEETs to efflux sugars to apoplast; R. solani acquires more sugars and subsequently accelerates host invasion.

Introduction

Sheath blight (ShB) is one of the most important diseases of rice, which causes yield losses of up to 50 % [1]. It is highly challenging to control the disease due to the wide host range of the pathogen, strong resistance of the sclerotia, and lack of rice resistance [2]. Rhizoctonia solani Kühn is the causative agent of ShB; it is divided into 14 anastomotic groups (AG1 to AG13 and AGBI) based on physiology, genetic composition, and mycelial interaction [3], [4]. AG1-IA, a subgroup of the AG1 fusion group, is the most destructive in rice. The pathogen infects all rice organs, leading to more than 50 % economic losses in extreme cases [5]. During infection, pathogens secrete effector proteins to reduce the hypersensitive reactions and increase susceptibility by interfering with the host immune defense response [6]. Generally, the effector proteins secreted by necrotrophic pathogens inhibit pattern-triggered immunity (PTI) at the early and transient establishment stages of infection, while others induce cell death at the late necrotrophic stages of infection and ultimately contribute to pathogen virulence [7]. Comparative genomics and secretome analyses have predicted 985 secretory proteins and 103 small cysteine-rich effector candidates in R. solani AG1-IA genome. Among these, RsAGLIP and RsIA_NP8 trigger cell death and inhibit PTI to promote disease development [5], [7], while RsAG8_06778, RsAG8_07159, and RsLysM participate in the interaction between R. solani and the host [8], [9]; however, the function of many other R. solani effectors has not been reported.

SWEETs (Sugar will eventually be exported transporters) are important sugar transporters of plants. They are mainly involved in sugar efflux and phloem loading and regulate the interaction between plants and pathogens [10]. Studies have shown that pathogen infection induces SWEET expression, which enhances host susceptibility [11], while the effectors secreted by pathogen induce the expression of host defense genes. For example, the secretion of hypersensitive response-inducing protein 1(UvHrip1) by the pathogen into rice downregulated OsPR expression [12]. In addition, TAL effectors, delivered to the plant cytoplasm through the type III secretory system, induced the expression of specific SWEET genes, ensuring sucrose delivery to the apoplasts of colonized cells [13]. The effectors of Xanthomonas oryzae pv. oryzae targeted OsSWEET11/Xa13, OsSWEET13/Xa25, and OsSWEET14 in rice [13], [14], [15], [16], [17]. Meanwhile, Avrb6, a TAL effector of X. citri subsp. malvacearum, activated the expression of the cotton sucrose transporter GhSWEET10 during its invasion [18]. These studies indicated that nutrient competition is a key mechanism involved in the interaction between pathogen and host; however, the role of fungal pathogen effectors in diverting the host nutrients is still unclear.

Our previous transcriptome study found that R. solani, the causative agent of sheath blight (ShB), strongly induced the expression of GT1 [19]. GT1 expression was lower in the ShB resistant cultivar ‘Teqing’ than the susceptible ‘Lemont’ [20]. In addition, studies have related GT1, a class I homeodomain leucine zipper (HD Zip) gene [21]; GT1, also named HOX12, regulates tiller branching and panicle exertion in rice [22], [23]; however, its function in rice defense against ShB is unclear. The WRKY transcription factors, an important gene family involved in forming regulatory networks, are also involved in plant response to pathogen infection and abiotic stress [24]. WRKY interacts with various proteins to create a network that regulates plant development and disease resistance [25]. Our previous study reported the negative role of WRKY53 in regulating rice resistance to ShB [19]. Subsequent studies showed that ShB activated WRKY53 to induce SWEET2a, a negative regulator of ShB resistance [26]. However, the possible molecular mechanism of R. solani-host interaction, especially the sugar competition between host and pathogen, remains unclear.

The present study found that R. solani secreted AOS2 increases the virulence of fungi through nutritional competition. The effector AOS2 interacts with WRKY53 and GT1 to form a transcription factor complex which activates SWEET genes to increase the susceptibility of rice. This finding will reveal the nutrient competition mechanism between pathogen and host and provide novel ideas for controlling ShB in rice.

Materials and methods

Plant growth and pathogen inoculation

Wild-type (WT) (Oryza sativa L. Dongjin, Longjing 11, and Zhonghua 11), wrky53 mutant, WRKY53 overexpressing (OX) [27], GT1 OX (#1, #2), GT1 RNAi (#4, #6) [21], and genome-edited gt1, sweet2a (#1, #2) and sweet3a (#1, #2, #3 and #4) rice plants were cultured in a glass greenhouse under 24–30℃ and natural lighting condition for one month until inoculated with R. solani. Nicotiana benthamiana plants were cultured in an incubator at 22–24℃ and 16 h/8 h light and dark cycle for four weeks at the Shenyang Agricultural University, China.

Rhizoctonia solani AG1-IA strain was cultured in a PDA medium at 30℃. The hyphal growth was measured after three days of incubation on a medium supplemented with 1, 5, 10, and 20% glucose or sucrose. Wood veneers (1 cm × 0.5 cm) or the PDA mediums (0.7 m diameter) with AG1-IA strains were used to inoculate rice sheaths or leaves. Rice leaves treated with RNase-free water (Mock), dsAOS2, and dsAOS2 with 5 mM glucose were inoculated similarly. The length or area of the lesion was measured using ImageJ software two weeks (sheaths) or three days (leaves) after inoculation.

Vector construction

Further, to screen the R. solani effectors that induced SWEETs during infection, 2 kb of SWEET2a promoter sequences and 200 R. solani AG1-IA proteins with signal peptide were cloned into pHisi (pHisi-pSWEET2a) and pGBKT7 yeast vectors (BD-putative effector), respectively [28]. Similarly, 3.0 kb SWEET3a promoter sequences with AOS2, GT1, and WRKY53 were cloned into pHisi (pHisi-pSWEET3a) for yeast one-hybrid assay. GT1 sequence was cloned into pGAD424 and pGBKT7 vectors (AD-GT1 and BD-GT1) for yeast two-hybrid assay. SWEET2a, SWEET3a, AOS2, GT1, and WRKY53 sequences were cloned into pDONR221 and then into pEH19 (SWEET2a-GFP, SWEET3a-GFP, AOS2-GFP, AOS3-GFP and GT1-GFP in tobacco), pDRf-eGFP-GW (SWEET2a-eGFP and SWEET3a-eGFP in yeast), pXNGW (AOS2-nYFP and GT1-nYFP), and pXCGW (WRKY53-cCFP and GT1-cCFP) by homologous recombination for localization, hexose transport and bimolecular fluorescence complementation assays, respectively. Primers used for vector construction are listed in Table S1.

Yeast hybrid and self-activation assays

The 200 BD-putative effectors were transformed into the yeast strain YM4271 containing pHisi-pSWEET2a, and AD-GT1, AD-WRKY53, and BD-AOS2 were transformed into the yeast containing pHisi-pSWEET2a and pHisi-pSWEET3a to screen and verify the activation of SWEET2a and SWEET3a by effector protein AOS2. The transformed yeasts were grown on the SD/-L or SD/-H for 48 h. BD-AOS2/BD-GT1/BD-AOS3 and AD-GT1/AD-WRKY53 were transformed into Y187 and AH109, respectively. Transformed yeasts were mated and cultured on SD/-LT or SD/-TLH with or without 3-AT to test the interaction between AOS2 or AOS3, GT1, and WRKY53. The yeast cells transformed with BD-GT1 and BD were grown on SD/-T or SD/-H media to self-activate GT1 at 30℃. Images were captured after 48 h of incubation.

Bimolecular fluorescence complementation and localization assays

To further verify the interaction between AOS2, GT1, and WRKY53, the WRKY53-cCFP, GT1-cCFP, AOS2-nYFP, and GT1-nYFP plasmids were transformed into GV3101, an Agrobacterium tumefaciens strain. Interaction protein pairs were transformed into tobacco cells by the infiltration method. Here, pXNGW and pXCGW were used as the negative control, while H2B-RFP was used as a nuclear marker. The A. tumefaciens strains with SWEET2a-GFP, SWEET3a-GFP, AOS2-GFP, and GT1-GFP were infiltrated into tobacco cells for localization analysis. Finally, the images were captured with an Olympus FV3000 instrument (Olympus Corporation, Japan) after 48 h of infiltration.

Hexose transport and invertase secretion assays

SWEET2a-eGFP and SWEET3a-eGFP plasmids were constructed and transformed into EBY4000, a yeast strain with hexose transport deficiency, to determine the ability of the SWEETs to transport glucose, fructose, and galactose. The yeast cells expressing SWEET2a, SWEET3a, AtSWEET1 (positive control), and pDRf-eGFP-GW (negative control) were cultured on the YP medium supplemented with 2 % maltose, 2 % glucose, 2 % galactose, 2 % fructose, or 1 % maltose with 1 % 2-deoxyglucose (2-DG, a toxic homolog of glucose), maintaining a pH of 5 or 7 using MES-Tris buffers. The fluorescence signal of the SWEET2a-GFP plasmid in yeast was collected after culturing in a YPM medium at 30℃ for 12 h.

Further, the invertase secretion assay was performed to test the secretory activity of AOS2. The predicted signal peptide of AOS2 with recombinant SUC2 was transformed into YTK12 (a SUC-deficient yeast strain) to help the transformed yeast secrete invertase and hydrolyze sucrose. Yeasts with the recombinant SUC2 were cultured on a CMD-W medium containing yeast N base without amino acids, tryptophan dropout supplement, sucrose and glucose, and a YPRAA medium (yeast extract, peptone, raffinose, and antimycin A). The yeast colonies multiplied on the above medium at 30 ℃ for 72 h as previously described [29]. The SUC2 guided by UV7823 was used as the negative control, while UV44 was used as the positive control.

Transactivation assay

The protoplasts were transformed with the pSWEET2a and pSWEET3a reporters and the 35S:AOS2, 35S:AOS3, 35S:GT1 and 35S:WRKY53 effectors following the polyethylene glycol (PEG) method, using 35S:LUC as an internal control. The protoplast proteins were extracted, and GUS activity was measured with the GUS assay kit (Promega, Madison, USA) at 12 and 24 h after transformation [30], [31], [32].

RNA extraction and qRT-PCR analysis

Total RNA was extracted from the rice seedlings or inoculated leaves, and cDNA was synthesized using the RNAiso Plus reagent and the PrimeScript RT reagent kit with gDNA Eraser (Takara, Dalian, China), following the manufacturer’s instructions. Quantitative RT-PCR was performed with the ChamQ universal SYBR qPCR Master Mix (Vazyme, Nanjing, China) on an Agilent Mx3000P instrument (Agilent, Palo Alto, CA, USA) to analyze the gene expression. The 2^-ΔΔCt method was used to calculate the relative expression levels of the genes, using Ubiquitin as the internal reference. Primers used for qRT-PCR are listed in Table S1.

Soluble sugar content assays

To verify that the effector affects sugar absorption in rice during ShB infection, the sheaths of one-month-old ZH11, gt1, sweet2a, sweet3a, DJ, GT1 OXs, LJ11, wrky53, and WRKY53 OX seedlings were collected for extraction of soluble sugars. These seedlings included both plants that had been inoculated with R. solani AG1-IA and control plants that had not been inoculated. The plant samples were purified using 80% ethanol that had been kept at 80℃ for 1 h. The extracts were concentrated and dissolved in 1 ml of distilled deionized water (ddH₂O). High-performance liquid chromatography (HPLC) was used for the identification of the glucose, sucrose, and fructose contents using an NH₂ analytical HPLC column and refractometer with parallax detection (Waters, Milford, MA, USA). The mobile phase was acetonitrile: ddH₂O (v:v, 75:25) and the flow rate was 1.0 ml·min⁻¹. The determination was repeated three times for each sample, according to the manufacturer’s instructions [33].

Statistical analysis

All data were expressed as mean ± standard deviation (SD) and statistically analyzed using the Prism 5 software (GraphPad, San Diego, CA, USA.). Group comparisons were performed using a one-way analysis of variance (ANOVA) and Student’s t-test (* P < 0.05).

Results

R. solani protein AOS2 activates rice SWEET2a to increase virulence

Sugar is an important nutrient source for plant pathogens. Therefore, to analyze the effects of sugar on R. solani growth, 1, 5, 10, and 20% of glucose or sucrose was added into the PDA or PSA media, and R. solani hyphal growth was measured. The results indicated that higher sugar content in the medium promoted R. solani hyphal growth (Fig. 1A, 1B). During host-pathogen interaction, SWEET sugar transporters play key roles and help the pathogen hijack sugar from the host [28]. The SWEET expression patterns were examined upon R. solani infection to analyze whether SWEETs play key roles in the interaction between rice and R. solani. The qRT-PCR analysis showed that R. solani infection had induced SWEET2a, SWEET2b, SWEET11, SWEET13, and SWEET14, with the highest induction detected for SWEET2a among the 21 SWEETs tested (Fig. 1C).

Fig. 1.

Rhizoctonia solani infection induces the expression of SWEETs. AR. solani AG1-IA cultured on PDA and PSA medium supplemented with 1, 5, 10, and 20% glucose or sucrose, respectively. Black triangle indicates sugar concentration increases from the left to the right. Glu and Suc indicate glucose and sucrose, respectively. B Hyphal growth of R. solani cultured at 30℃ for 3 days. Different letters above the bars indicate significant differences at P < 0.05. C RT-qPCR analysis of SWEET gene expression levels 0, 24, 48, and 72 h after R. solani inoculation.

Furthermore, a yeast-one hybrid was performed using 2 kb of SWEET2a promoter sequence and 200 putative secretory proteins of R. solani. These 200 proteins with a predicted signal peptide [4] were cloned into the pGBKT7 yeast vector (data not shown). Yeast one-hybrid assay results indicated that AG1IA_03763, a predicted hypothetical protein and named activator of SWEET2a (AOS2) in this study, activated the SWEET2a promoter (Fig. 2A). Subsequently, the transient assay in the protoplast cells showed that AOS2 and the positive control WRKY53 [26] activated the SWEET2a promoter (Fig. 2B). Analysis of the expression patterns in rice using RT-qPCR revealed that AOS2 expression was induced after R. solani infection (Fig. 2C). AOS2 contained a predicted signal peptide within the first 24 amino acid residues at the N-terminal. Furthermore, an invertase secretion assay was performed using previously reported methods [29]. The nucleotide sequences encoding the first 24 amino acid residues of AOS2 were cloned into the N-terminal of the signal peptide-deleted SUC2, an invertase gene, and expressed in YTK12, a SUC-deficient yeast strain, which cannot hydrolyze sucrose. Typically, if YTK12 expressing the recombinant SUC2 guided by the AOS2 signal peptide secretes the protein, it will allow the yeast to grow on the YPRAA medium with sucrose. As expected, the cells with UV44 (positive control) and signal peptide of AOS2 grew in the YPRAA medium, while those with the negative control UV7823 failed to grow, suggesting that AOS2 is a secretory protein (Fig. 2D). Further analysis showed that AOS2-GFP with the nuclear maker H2B-RFP [34] colocalized in the nucleus, while free GFP localized in the nucleus and the cytosol (Fig. 2E).

Fig. 2.

Secretory protein AOS2 directly activates SWEET2a. A Schematic diagram shows the recombinant vectors BD-putative effector, pSWEET2a-pHisi-1, and the yeast strain YM4271 with two vectors. The yeast cells were co-transformed with 2 kb of the SWEET2a promoter and BD-AOS2 or BD empty vector. B Transient assay performed to analyze AOS2 activation on SWEET2a promoter. The 35S:WRKY53 plasmid was used as the positive control. Different letters indicate significant differences at P < 0.05. C qRT-PCR analysis of AOS2 expression levels 0, 12, 24, 48, 72, and 96 h after R. solani inoculation. Different letters indicate significant differences at P < 0.05. D Signal peptide of AOS2 plays a secretory role in yeast. Schematic representation of the recombinant proteins constructed with different signal peptides linked to invertase. SUC-deficient yeast strain YTK12 transformed with various signal peptide-invertase recombinant proteins were cultured on CMD-W and YPRAA media. The signal peptides of UV7823 and UV44 were used as the negative and positive controls, respectively. E Localization of AOS2-GFP and GFP alone in tobacco leaves. H2B-RFP was used as a nuclear marker. Scale bar = 20 µm.

Then, to analyze the function of AOS2 during R. solani invasion, the AOS2 dsRNA was synthesized and treated via the spray induced gene silencing (SIGS) method on the rice leaf surface[35]. Inoculation of R. solani identified that dsRNA treatment significantly reduced the disease severity (Fig. 3A, 3B). Subsequent RT-qPCR showed dramatic suppression of AOS2 in the dsRNA-treated group compared to the mock control (Fig. 3C). The R. solani-mediated induction of SWEET2a was also significantly inhibited in the AOS2 dsRNA-treated leaves compared to the mock (Fig. 3D).

Fig. 3.

Silencing of Rhizoctonia solani AOS2 increases disease resistance by inhibiting SWEET2a expression. A Leaves treated with the mock (RNase-free water) or dsAOS2 (#1, #2, and #3) were inoculated with R. solani. B Percentage of lesion area on the leaves treated with the mock or dsAOS2 (#1, #2, and #3). Different letters above the bars indicate significant differences at P < 0.05. C The expression level of AOS2 in the mock or dsAOS2 groups. Different letters above the bars indicate significant differences at P < 0.05. D The expression level of SWEET2a in the mock or dsAOS2 groups 0 or 48 h after inoculation. Different letters above the bars indicate significant differences at P < 0.05.

AOS2 interacts with rice transcription factor WRKY53 and GT1 to form a complex that activates SWEET2a and SWEET3a

Then, to understand the components interacting with AOS2, a yeast two-hybrid assay was performed using AOS2 as a bait. Among the factors isolated, WRKY53 and GT1 were further examined for their functions. Yeast two-hybrid and BiFC assays confirmed that AOS2 interacts with WRKY53 or GT1 in the nucleus (Fig. 4A, 4B). Further subcellular localization examination showed that WRKY53 interacts with GT1 in the nucleus (Fig. 4C, 4D). Next, the yeast one-hybrid assay showed that AOS2, GT1, and WRKY53 activate the SWEET2a promoter (Fig. 5A). We found that the inoculation of R. solani activated a few SWEETs, except SWEET2a; therefore, we examined the role of AOS2, GT1, and WRKY53 in activating the SWEET promoters (data not shown). The results showed that AOS2, GT1, and WRKY53 activated the SWEET3a promoter (Fig. 5B). Additionally, the transient assay revealed that AOS2, GT1, or WRKY53 alone activated SWEET2a and SWEET3a promoters, and the coexpression of AOS2, GT1, and WRKY53 demonstrated an additive effect on the activation of SWEET2a and SWEET3a promoters (Fig. 5C, 5D). Further RT-qPCR showed that dsRNA-mediated suppression of SWEET2a significantly enhanced after 48 h in the AOS2 dsRNA treated leaves compared to the mock (Figure S1).

Fig. 4.

Interaction between GT1, WRKY53, and AOS2. A Yeast two-hybrid assay between AOS2, WRKY53, or GT1. Transferred yeast cells were cultured on a synthetic medium lacking Trp and Leu (-LT), Trp, Leu, and His (-LTH) or -LTHA with 10 mM 3-AT (3-amino-1,2,4-triazole, a histidine synthase inhibitor). B Bimolecular fluorescence complementation assay between AOS2-nYFP and WRKY53-cCFP or GT1-cCFP. H2B-RFP was used as a nuclear marker. Scale bar = 20 µm. C Interaction between WRKY53 and GT1 in yeast two-hybrid assay. Transformed yeast cells were cultured on SD/-LT, SD/-LTHA, or SD/-LTH with 2.5 mM 3-AT. D Interaction between WRKY53-cCFP and GT1-nYFP in bimolecular fluorescence complementation assay. Scale bar = 20 µm.

Fig. 5.

AOS2-WRKY53-GT1 transcriptional complex activates SWEET2a and SWEET3a. A Yeast strain YM4271 co-transformed with pSWEET2a-pHisi-1 and AD-AOS2, AD-WRKY53, AD-GT1, or empty vector (EV). B Yeast strain YM4271 co-transformed with pSWEET3a-pHisi-1 and AD-AOS2, AD-WRKY533, AD-GT1, or EV. The yeast cell growth was monitored on SD/-L and SD/-LH with 10 mM 3-AT. The transient assay was performed using protoplast cells by co-expressing pSWEET2a-GUS (C) or pSWEET3a-GUS (D) with EV, AOS2, WRKY53, GT1, or AOS2 + WRKY53 + GT1. Different letters indicate significant differences at P < 0.05.

GT1 negatively regulates rice resistance to ShB

We previously identified WRKY53 as an activator of SWEET2a that negatively regulates rice resistance to ShB[26]. Our previous transcriptome study identified that GT1 is highly sensitive to R. solani infection[19]. The present study’s RT-qPCR confirmed that R. solani infection significantly induced GT1 expression (Fig. 6A). In addition, the GT1-GFP was found colocalized with the nuclear marker H2B-RFP in the tobacco leaves (Fig. 6B), and GT1 showed transcription activation activity in yeast (Fig. 6C). To verify GT1 function in rice resistance to ShB, GT1 overexpressing (OX) and RNAi plants were generated in a Japonica cultivar Dongjin (DJ). Analysis of these lines showed higher GT1 expression in the OXs and lower expression in the RNAi lines (Fig. 6D). Inoculation of R. solani revealed that GT1 OXs were more susceptible to ShB, while GT1 RNAi plants were less susceptible than the wild-type control DJ (Fig. 6E, 6F). In addition, CRISPR/Cas9-mediated genome-edited mutant (Zhonghua11 [ZH11] background) with a single base pair (T) addition in the third exon generated a new stop codon after E²¹⁵ residue. Inoculation of R. solani showed that the gt1 mutant was less susceptible to ShB compared to the wild-type control ZH11 (Fig. 6G, 6H).

Fig. 6.

Transcription factor GT1 acts as a negative regulator of ShB resistance. A qRT-PCR analysis of GT1 gene expression levels 0, 24, 48, and 72 h after Rhizoctonia solani inoculation. B Localization of GT1-GFP and free GFP in tobacco leaves. H2B-RFP was used as a nuclear marker. Scale bar = 20 µm. C Transcriptional activation activity using BD-GT1 into PJ69-4a yeast strain. The activity was measured on the SD medium -Trp or –His. BD was used as the negative control. D Expression levels of GT1 in DJ, GT1 OX (#1, #2), and GT1 RNAi (#4, #6) plants. E The sheaths from the DJ, GT1 OX (#1, #2), and GT1 RNAi (#4, #6) inoculated with R. solani. The length of lesions on the leaf surface was measured in (F). G Genomic structure of gt1 mutant. Black boxes represent exons, and the sequences below the third exon represent ZH11 and CRISPR/Cas9 induced gt1 mutant sequences. The base marked in red indicates the difference compared to ZH11. E215* indicates the stop codon after E215 residue. H Sheaths from the ZH11 and gt1 inoculated with R. solani. The length of the lesions was measured two weeks after inoculation. Different letters indicate significant differences at P < 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

SWEET2a and SWEET3a function downstream of the AOS1-GT1-WRKY53 complex and negatively regulate rice resistance to ShB

Transcriptome analysis using GT1 OX and RNAi plants indicated that SWEET2a and SWEET3a, which are positively regulated by GT1[22]. SWEETs play key roles in host-microbe interaction[13], [36]. The RT-qPCR showed that the expression levels of SWEET2a and SWEET3a were higher in GT1 OXs and lower in the GT1 RNAis or gt1 mutant compared with their corresponding wild-type controls (Fig. 7A, 7B). Similarly, SWEET2a and SWEET3a expression levels were lower in wrky53 but higher in WRKY53 OX compared to their control (LJ11) (Fig. 7C, 7D).

Fig. 7.

GT1 or WRKY53 significantly induces the expression of SWEET2a and SWEET3a. A, B Expression levels of SWEET2a and SWEET3a in DJ, GT1 OX (#1, #2), GT1 RNAi (#4, #6), ZH11, and gt1 plants. C, D Expression levels of SWEET2a and SWEET3a in LJ11, wrky53, and WRKY53 OX plants by qRT-PCR. The Ubiquitin was the internal reference. Different letters indicate significant differences at P < 0.05.

Then, we generated the CRISPR/Cas9-mediated of genome-edited mutants. The sequencing indicated a single base pair (A) addition in the first exon in sweet2a-1 and 13 base pairs deletions from the third exon in the sweet2a-2 mutant. Inoculation of R. solani demonstrated that sweet2a mutants were less susceptible to ShB than ZH11 (Fig. 8A, 8B). Among the sweet3a mutants, sweet3a-1, sweet3a-2, sweet3a-3, and sweet3a-4 showed 70 bp deletion, 103 bp deletion, 4 bp deletion, and 1 bp (T) addition, respectively. These sweet3a mutants were less susceptible to ShB than ZH11 (Fig. 8C, 8D), indicating that SWEET2a and SWEET3a, similarly to GT1 and WRKY53, negatively regulate rice resistance to ShB.

Fig. 8.

Membrane localized SWEET2a and SWEET3a negatively regulate ShB resistance with glucose transport activity. A Genomic structure of sweet2a mutant. The sequences below the first and third exons are the ZH11 and sweet2a genome-edited sequences. The base marked in red indicates the difference compared to ZH11. B Sheaths and lesion length in ZH11 and sweet2a (#1, #2) inoculated with R. solani. C Genomic structure of sweet3a mutant. The sequences below the second exon represent ZH11 and sweet3a genome-edited sequences. The base marked in red indicates the difference compared to ZH11. D Sheaths and lesion length in ZH11 and sweet3a (#1, #2, #3, and #4) inoculated with R. solani. E Localization of SWEET2a-GFP and SWEET3a-GFP in tobacco leaves. SWEET11-RFP was used as a membrane marker. Scale bar = 20 µm. F OsSWEET2a, OsSWEET3a, and AtSWEET1 in the EBY4000 yeast with hexose transport deficiency. The yeast cells were cultured on the medium supplemented with 2% maltose, 2% glucose, 2% galactose, 2% fructose, or 1% maltose with 1% 2-deoxyglucose (2-DG) at pH 5 and 7. Different letters indicate significant differences at P < 0.05. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

We also found that SWEET2a-GFP and SWEET3a-GFP colocalized with a plasma membrane marker SWEET11-RFP [37] in tobacco leaves (Fig. 8E). Further, the hexose transport activity of SWEET2a and SWEET3a was tested in EBY4000, a hexose transporter deficient yeast strain[38]. Compared with the positive control AtSWEET1, the SWEET3a transported glucose, fructose, and galactose in yeast; however, SWEET2a failed to transport glucose. Meanwhile, SWEET2a-GFP was not localized in the plasma membrane in yeast (Figure S2). However, it transported the toxic glucose analog 2-DG [39] independent of pH (Fig. 8F), suggesting that SWEET2a and SWEET3a are hexose transporters. Finally, to evaluate whether AOS2 suppression-mediated decline in R. solani virulence is associated with apoplast sugar levels, 5 mM glucose was sprayed onto the dsRNA treated plants. Here, glucose application rescued AOS2 suppression-induced weak virulence in rice (Figure S3).

GT1 and WRKY53 activate soluble sugar accumulation after infection with R. solani

To evaluate the potential association between sugar contents and disease resistance, the soluble sugar (glucose, fructose, and sucrose) contents were measured in the GT1 and WRKY53 mutants and overexpression plants. The results indicated that the sucrose, glucose, and fructose contents were similar between ZH11 and gt1 and between DJ and GT1 OXs (-1, -2) before inoculation of R. solani AG1-IA. The glucose, fructose, and sucrose contents were observed to be increased 12 and 24 h after inoculation, and the sugar levels were lower in gt1 than in ZH11 but higher in GT1 OXs compared with DJ (Fig. 9A-9C). Similarly, the glucose, fructose, and sucrose contents were similar between the LJ11, wrky53 and WRKY53 OX plants. After 12 and 24 h, the sugar contents were lower in wrky53 while higher in WRKY53 OX than in wild-type LJ11 (Fig. 9D-9F). GT1 and WRKY53 activated SWEET2a and SWEET3a, and sweet2a and sweet3a are also less susceptible to ShB; therefore, we then measured the soluble sugar contents in ZH11, sweet2a-1 and sweet3a-1 before and after inoculation of R. solani AG1-IA. The results showed that the soluble sugar contents were similar in ZH11, sweet2a and sweet3a before inoculation. After 12 and 24 h, the glucose, fructose and sucrose contents were lower in sweet2a and sweet3a than in ZH11, and the soluble sugar contents were similar in both sweet2a and sweet3a (Fig. 9G-9I).

Fig. 9.

Sugar contents in the sheaths of GT1, WRKY53, SWEET2a and SWEET3a genetic materials. (A) Glucose, (B) fructose and (C) sucrose contents in ZH11, gt1, DJ and GT1 OXs (-1, -2) 0, 12 and 24 h after inoculation of R. solani. (D) Glucose, (E) fructose and (F) sucrose contents in LJ11, wrky53 and WRKY53 OX 0, 12, and 24 h after inoculation of R. solani. (G) Glucose, (H) fructose and (I) sucrose contents in ZH11, sweet2a-1 and sweet3a-1 0, 12 and 24 h after inoculation of R. solani. The different letters indicate significant differences between groups at P < 0.05. ‘ns’: non-significant difference.

Discussion

Sheath blight severely affects rice production; therefore, an in-depth understanding of the interaction between R. solani and rice is necessary for management. A few effector proteins of R. solani have been reported to induce cell death in plants[7], [40], [41], [42]; however, the process via which R. solani hijacks sugar from the rice plants is unclear. SWEETs are the target of X. oryzae pv. oryzae, a causative agent of bacterial blight[11]. The effector proteins from X. oryzae pv. oryzae translocate into the nucleus and activate SWEETs, leading to efflux of cytosolic sugars to apoplast for feeding X. oryzae pv. Oryzae[13], [14], [15], [16], [17]. Other pathogens, such as fungi, nematodes and viruses, modulate plant SWEET genes[28]. We previously found that SWEET2a, SWEET11, and SWEET14 function in ShB resistance[26], [37], [43], suggesting the role of SWEETs during rice-R. solani interaction for sugar distribution. Here, using the R. solani-rice interaction model, we found that the rice transcription factor WRKY53 and GT1 interacted with the effector protein AOS2 from R. solani. This finding extends the known regulatory network of sugar hijacking by the pathogen from rice and provides a new perspective for the control of ShB by the inhibition of effectors such as AOS2.

The present study monitored sugar requirement for R. solani growth by applying different concentrations of sugar on PDA or PSA. The results showed that high sugar concentrations enhanced R. solani hyphal growth. Rhizoctonia solani is a heterotroph that had to hijack sugar from host plants to maintain the life cycle. It is a necrotrophic pathogen but requires rapid sugar acquisition at the early stage of infection, which might need SWEETs to efflux sugars to the apoplast. Further investigation revealed variations in the expression of a few SWEET genes after R. solani infection; here, SWEET2a showed the highest induction in rice after infection. In addition, the yeast one-hybrid and transient assays indicated that AOS2 is an activator of SWEET2a. AOS2 has a signal peptide at the N-terminal end that helps protein secretion in yeast. Meanwhile, AOS2 without the signal peptide localized into the nucleus, suggesting that AOS2 is a secretory protein that functions as a transcriptional activator. Suppression of AOS2 expression by SIGS significantly reduced disease severity, confirming the importance of AOS2 for R. solani invasion.

The further analysis detected an interaction between the components to form the AOS2-WRKY53-GT1 transcription factor complex. Mutation of WRKY53 or GT1 promoted rice resistance to ShB, while overexpression inhibited rice resistance compared to wild-type plants, suggesting WRKY53 and GT1 are the negative regulators of plant defense to ShB. Both GT1/HOX12 and WRKY53 are transcription factors[23], [44], and WRKY53 is known to activate SWEET2a and negatively regulates rice resistance to ShB[26]. Consistent with these earlier reports, the present study found that R. solani infection had induced GT1 and WRKY53, which activated SWEET2a and SWEET3a promoters, suggesting that the AOS2-WRKY53-GT1 transcription factor complex activates SWEET2a and SWEET3a. The SWEET2a and SWEET3a proteins were found localized in the plasma membrane. The sugar transporter assay using EBY4000, a hexose transport deficient yeast strain, demonstrated that SWEET2a and SWEET3a transported hexose. Meanwhile, the coexpression of AOS2-WRKY53-GT1 demonstrated an additive effect on SWEET2a and SWEET3a activation. The subsequent measurement of soluble sugar contents showed that inoculation of R. solani AG1-IA increased the soluble sugar contents in the sheaths and that GT1, WRKY53, SWEET2a, and SWEET3a mutants accumulated less soluble sugar, while the GT1 and WRKY53 overexpression plants accumulated more soluble sugars compared with the wild-type controls. These observations collectively suggest that R. solani activates WRKY53 and GT1 via an unknown pathway; R. solani simultaneously secretes AOS2 to enhance WRKY53 and GT1-mediated SWEET2a and SWEET3a activation for sugar efflux to apoplast by which R. solani obtains nutrients (Fig. 10).

Fig. 10.

Speculative model shows GT1-WRKY53 complex regulates SWEETs-related ShB susceptibility with or without AOS2 secretion in the nucleus.Rhizoctonia solani infection generates the GT1-WRKY53 complex that activates SWEET2a and SWEET3a in the nucleus. Hexose gets exported from cytoplasm to apoplast by SWEET2a and SWEET3a for R. solani absorption. If the host successfully recognizes the effector protein AOS2 secreted by R. solani, GT1-WRKY53-meidated SWEET2a and SWEET3a induction rises significantly. More hexose flows to the pathogen, increasing the virulence.

In addition, another putative effector AG1IA_04889 activated SWEET2a promoter in yeast-one hybrid screening, and named Activator of SWEET2a (AOS3) (Fig. S4A). AOS3-GFP was localized at the nucleus, and interacts with GT1 but not WRKY53 (Fig. S4B, S4C). Also, coexpression of AOS3, GT1, and WRKY53 has additive effects on SWEET2a activation (Fig. 4SD). These results suggested that multiple effectors from R. solani were able to activate SWEET2a. But AOS3 exhibited weak interaction with GT1 compared to AOS2, and no interaction with WRKY53, suggesting that AOS2 might be main effector to activate SWEETs in R. solani.

We also found that R. solani infection suppressed SWEET3a expression. Moreover, the SWEET3a mutant was less susceptible to ShB than the wild-type plant, suggesting that rice represses SWEET3a expression via an unknown pathway to protect rice from R. solani invasion. In addition, AOS2 suppression inhibited R. solani-mediated decline in SWEET3a expression, confirming that AOS2 activates SWEEET3a to hijack sugars. Thus, the present study proved that the R. solani secretory protein AOS2 interacts with WRKY53 and GT1 in the nucleus to activate the hexose transporters, SWEET2a and SWEET3a. These data improve our understanding of R. solani invasion and provide helpful information to protect rice from ShB.

This method allowed the rapid identification of candidate proteins that may directly regulate SWEETs and the exploration of the regulatory network of sugar redistribution during pathogen infection in both pathogen and plants, and, through genetic analysis, confirmed that GT1, WRKY53, SWEET2a, and SWEET3a negatively regulated rice resistance to ShB. However, there are several limitations, for example, difficulties in the measurement of soluble sugar levels in the apoplasts due to their lower sugar contents. In addition, the transformation approaches have not been well established for R. solani AG1-IA, reducing the effectiveness of gene function characterization. These issues need to be addressed by the development of new techniques in future studies.

Compliance with ethics requirements

This article does not contain any studies with human or animal subjects.

CRediT authorship contribution statement

Shuo Yang: Investigation, Data curation, Writing – original draft. Yuwen Fu: Investigation, Validation, Formal analysis. Yang Zhang: Data curation, Writing – review & editing. De Peng Yuan: Data curation, Writing – review & editing. Shuai Li: Validation, Formal analysis. Vikranth Kumar: Writing – review & editing. Qiong Mei: Methodology, Investigation, Validation. Yuan Hu Xuan: Funding acquisition, Conceptualization, Supervision, Project administration, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary material

The following are the Supplementary data to this article: