GmPBS1, a Hub Gene Interacting with Rhizobial Type-III Effectors NopT and NopP, Regulates Soybean Nodulation

Abstract

Soybean is a legume crop rich in protein and oil. Symbiotic nitrogen fixation plays an important role in the growth of soybean. The type-III effectors such as NopT and NopP are the important signaling factors for the establishment of symbiosis in soybean. In this study, the analysis of nodulation in soybean after inoculation with HH103ΩNopT, HH103ΩNopP, and HH103ΩNopT&NopP indicated crosstalking between NopT and NopP. Further, we aimed to identify the genes of soybean involved in the pathway underlying the crosstalk between NopT and NopP using RNA-seq analysis. Five of the identified candidate genes were confirmed to be induced by NopT and NopP. The expression of GmPBS1 significantly increased to a much larger extent than that of the other four genes after soybean was inoculated with HH103ΩNopT, HH103ΩNopP, or HH103ΩNopT&NopP. The interaction between NopT and GmPBS1 was confirmed via bimolecular fluorescence complementation. Finally, nodulation analysis after GmPBS1 overexpression in the hairy roots indicate that GmPBS1 can regulate the negative effect of NopP on the nodulation, and this regulation is related to NopT. Collectively, our results suggested that during the nodulation in soybean, NopT and NopP have a crosstalking network and GmPBS1 is the hub gene.

1. Introduction

Soybean (Glycine max L. Merr) is a major source of plant-based protein for people worldwide. It provides a healthy, affordable, and environmentally friendly alternative to animal-based protein [1]. According to the current agrochemical practice, soybean is grown without inoculation with rhizobia, and due to this, high doses of mineral nitrogen are normally applied to the soil. Excessive use of nitrogen fertilizer leads to environmental pollution and soil degradation. Therefore, it is urgent to find an effective way to replace traditional nitrogen fertilizers [2].

Biological nitrogen fixation (BNF) by legumes and rhizobia is an important natural source of nitrogen in agricultural ecosystems [3]. In legumes, the process of BNF occurs in root nodules, which are highly specialized organs containing bacterial symbionts. The total nitrogen fixation accounts for 50–60% of the total nitrogen absorbed by plants [4].

The process of symbiotic nitrogen fixation by rhizobia and legumes is the result of the specific interaction between them, and the development of symbiosis is jointly regulated by several signals [5]. Generally, legume roots release unique flavonoids into soil, attracting rhizobia to the plant rhizosphere [6]. Flavonoids can activate the expression of nodule (Nod) genes in rhizobia. The proteins encoded by Nod genes are involved in the synthesis of lipid oligosaccharides, namely, Nod factors (NFs), which contain species-specific modification [7]. NFs are perceived by a plasma-membrane-localized receptor complex consisting of the pseudokinase Nod factor perception (MtNFP) and receptor-like kinase (RLK) LysM domain receptor-like kinase 3 (MtLYK3) in Medicago truncatula and Nod factor receptor 1 (LjNFR1) and LjNFR5 in Lotus japonicus. Thus, the symbiotic nodulation signal of the plant is activated, inducing the formation of infection threads and leading to curling of hairy root [8,9].

PBS1 (a protein kinase) and some PBS1-like proteins, such as Botrytis-induced kinase1 (BIK1), PBL1, and PBL2 belong to the RLCKs VII subfamily. All of them are related to FLS2 and transmit immune signals through cell surface immune receptors [10]. Flagellin induces phosphorylation of these RLCKs. BIK1 directly phosphorylates the NADPH-oxidase respiratory burst oxidase homolog D (RBOHD) at specific sites, thereby controlling the production of reactive oxygen species (ROS) and stomatal immunity. In plants with mutant BIK1, PBL1, PBL2, or PBS1, the defense responses were compromised to varying degrees [11,12,13]. However, in contrast to BIK1 mutants, PBS1 mutants only exhibited a defect in defense response against pattern-triggered immunity.

In Arabidopsis thaliana cells the effector AvrPphB cleaves PBS1 after infection with Pseudomonas syringae, resulting in a conformational change in PBS1 that activates RPS5-mediated resistance [14]. AvrPphB is a cysteine protease that targets serine–threonine kinases involved in pathogen-associated molecular pattern-triggered immunity, and it is highly homologous to the type-III effector NopT. PBS1 forms a preactivation complex with RPS5 and triggers RPS5 activation upon the cleavage of AvrPphB [15,16]. Although TaPBS1 (the PBS1 homolog) in wheat can be cleaved by AvrPphB and can bind to RPS5, it cannot trigger RPS5-mediated hypersensitivity (HR) in transient experiments. Both PBS1 and TaPBS1 are phosphorylated by flagellin [17].

It is known that the type-III secretion system (T3SS) transports a cocktail of effector proteins directly into the host cell, each with multiple functions. Previous studies have reported that double mutants of type-III effectors can be used to detect the synergistic relationship between different effectors. We hypothesized that a gene in the host can simultaneously respond to various type-III effectors, and the host signaling network may be regulated and affected by a specific hub gene during the establishment of symbiosis. To identify the genes in soybean that respond to both NopT and NopP, we performed RNA-seq analysis to determine differentially expressed genes (DEGs) after inoculation of soybean with single and double mutants of NopT and NopP. Further, we screened the candidate genes and elucidated their interaction at both transcript and protein levels.

2. Materials and Methods

2.1. Construction of Rhizobial NopT and NopP Mutants

The HH103ΩNopT and HH103ΩNopP (NopT and NopP single mutants [18,19], respectively) and HH103ΩNopT&NopP (double mutant) were constructed using a triparental mating method. First, NopT with an inserted Spectinomycin antibiotic (Spec) resistant gene was cloned and ligated into the vector PJQ2000 to generate PJQ2000-NopT-Spec [20]. The Spec coding sequence was inserted at the site 18-bp downstream of the start codon of NopT. Further, Escherichia coli carrying PJQ2000-NopT-Spec (Figure 1a), helper strain, and Sinorhizobium fredii HH103 (HH103) were mixed in the ratio 1:1:3 as described in a previous triparental mating protocol [21]. The constructed NopT mutant was confirmed using polymerase chain reaction (PCR) performed with appropriate primers and the KOD One^™ PCR Master Mix Kit (TOYOBO, Beijing, China) as per the manufacturer’s instructions. Similarly, the single mutant HH103ΩNopP and double mutant HH103ΩNopT&NopP were constructed. The detailed method is as follows. The digestion site on the nopP gene was designed, and the specific digestion site BamHI was selected. The GAATTG sequence was mutated to the GGATTC sequence at 8-bp downstream of the nopP gene start codon (ATG) to acquire a digestion site (Figure 1b). Two primers were designed within 30-bp upstream and downstream of the mutation site, and the two primers had inverse complementary sequences. For the Kanamycin (Kan) resistance gene on pEASY-T1 vectors, primers were designed, and the SpeI digestion-site sequence was added at 5′ end of the primers. After amplification of the two pairs of primers using PCR, nopP and Kan ligation product was obtained. The specific amplification primers of NopT and NopP are given in Table S1.

2.2. Soybean Nodulation Experiment

The soybean cultivar used in the experiment was Suinong 14 (SN14), and the rhizobial strain was Sinorhizobium fredii HH103 (hereafter referred to as HH103), along with its derived mutants HH103ΩNopT, HH103ΩNopP, and HH103ΩNopT&NopP. Soybean plants were grown under long-day (16 h/8 h light/dark) and 60% relative humidity conditions at 25 °C. Prior to rhizobial inoculation, SN14 seeds were sterilized with chlorine and cultured until the first trifoliate leaf was uncovered. The rhizobial suspension was prepared and mixed with 10 mM MgSO₄ to obtain an OD₆₀₀ of 0.6, and each plant was inoculated with 1.5 mL of this suspension. The nodulation phenotype was evaluated 30 days after the inoculation with rhizobia.

2.3. Next-Generation RNA Sequencing and Analysis

The experimental material for transcriptome sequencing was SN14, a famous soybean cultivar in China. The roots of soybean were inoculated with HH103, HH103ΩNopT, HH103ΩNopP, or HH103ΩNopT&NopP in three biological replicates. Further, 36 h after rhizobial inoculation, 1 cm long roots were collected from the hypocotyl of the roots, and 3 biological replicates were prepared.

DEGs were identified using DEseq2-edgeR. The screening criteria were false discovery rate <0.01 and fold change >1.5. Further, KEGG and GO annotations were used to understand their functions in soybean. As our aim was to identify the potential genes involved in the same signaling network, the Weighted Gene Co-expression Network Analysis (WGCNA) was used, and the network was plotted using TBtools software (V1.108).

2.4. Construction of a Vector for Overexpressing GmPBS1 in the Hairy Roots

Total RNA was isolated from SN14 using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). The full-length cDNA of GmPBS1 was obtained using PCR with KOD One^™ (TOYOBO, Beijing, China) DNA polymerase (TransGen Biotech, Beijing, China) and a pair of primers (forward primer: 5′-ATTGATTAGAGATCTTCTAGAATGATGGGTAGTTGCCCTTGTT-3′; reverse primer: 5′-AATGTCGACGGTACCGGATCCTTTGCTTCCTTCAGGACTACTTTG-3′). The amplified products were purified and cloned into the Fu28 vector with homologous recombination. pSOY1-35S: GmPBS1: GFP was obtained by replacing the target fragment on the FU28 vector into the eukaryotic expression vector pSOY1 through LR recombination reaction.

2.5. RNA Extraction and RT-qPCR Analysis

Soybean roots were collected 36 h after rhizobial inoculation. RNA was extracted from soybean root cells using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA) as per the manufacturer’s protocol. cDNA was synthesized using HiScript II Reverse Transcriptase. RT-qPCR was performed on the Roche lightcycl480 system (Roche, Mannheim, Germany) and TB Green (Takara, San Jose, CA, USA). The experiment involved 3 biological replicates and 3 technical replicates. The specific primers for the analysis of genes are listed in Table S2.

2.6. Subcellular Localization of GmPBS1 in Tobacco

According to the coding sequences of GmPBS1 and NopT, forward and reverse primers were constructed to introduce EcoRV and KpnI sites, respectively. With cDNA of SN14 seeds as the template, coding DNA sequence (CDS) was amplified using high-fidelity Taq polymerase and ligated to the vector Fu28 with green fluorescent protein (GFP) label. The product Fu28-GmPBS1: GFP (Fu28-NopT: GFP) was used to transform E. coli DH5α, and the positive monoclonal colonies identified by PCR were sequenced. Further, the target fragment GmPBS1: GFP (NopT:GFP) was transferred into a plant expression vector pSoy1 driven by CaMV35S promoter through Gateway^™ LR reaction, and the constructed pSoy1-GmPBS1:GFP (pSOY1-NopT:GFP) plasmids were transformed into Agrobacterium tumefaciens EHA105. A. tumefaciens EHA105 with pSoy1-NopT:GFP and pSOY1-GmPBS1:GFP was used to infect tobacco (Nicotiana benthamiana) leaves to study the sub-cellular localization of GmPBS1. The plants were grown at 26 °C under 16 h light/8 h dark and 60% relative humidity conditions for 48 h [22]. GFP fluorescence was observed using a Zeiss LSM 700 confocal laser scanning microscope (Zeiss, Oberkochen, Germany). Transient infiltrated proGmPBS1: GFP N. benthamiana leaves were used as location controls. The plant expression vector pSoy1 is presented by Professor Fu Yongfu (Institute of Crop Sciences, Chinese Academy of Agricultural Sciences).

2.7. Bimolecular Fluorescence Complementation (BiFC) Assay to Study NopT–GmPBS1 Interaction

According to the coding sequence of GmPBS1 and NopT, forward and reverse primers were constructed to introduce the AhdI cleavage site. With the cDNA from the seeds of SN14 as the template, the CDS of the target gene was amplified using high-fidelity Taq polymerase and ligated into the vector pGWC. The ligation product pGWC-GmPBS1 (pGWC-NopT) was used to transform E. coli DH5α, and the positive clones identified using PCR were sequenced. The target fragment of GmPBS1 (NopT) was constructed from pEarlyGate 201 and pEarlyGate 202, which contained cYFP and nYFP, respectively. The constructed plasmid was used to transform A. tumefaciens EHA105. The resulting clones were verified using sequencing. The plasmids were introduced into A. tumefaciens EHA105, which was used to infect tobacco as described previously. The vectors required for the experiment is shown in Table S3. Infected tissues were analyzed at approximately 48 h after infection. Fluorescence of yellow fluorescent protein (YFP) was observed using a Zeiss LSM 700 confocal laser scanning microscope (Zeiss, Oberkochen, Germany).

2.8. Hairy Root Transformation and Detection of Positively Transformed Hairy Roots

To obtain transgenic composite plants, transformation of soybean hairy roots was performed according to a previously published method using the cultivar Dongnong 50 (DN50). The plant was adapted to the environment by growing in high humidity for 1 week. For nodulation phenotype analysis, each plant was inoculated with 1.5 mL suspension of HH103, HH103ΩNopT, HH103ΩNopP, or HH103ΩNopT&NopP with OD₆₀₀ of 0.6. The nodulation phenotype was evaluated 30 days after the inoculation with rhizobia. The positively transformed hairy roots with overexpressed GmPBS1 were identified by detecting fluorescence with GFP antibody and a hand-held fluorescent lamp (Luyor 3415RG; Ruyo, Shanghai, China).

2.9. Phenotypic Statistics and Data Analysis

SPSS22.0 (statistical Product and Service Solutions) software was used for statistical analysis of nodule number and nodules dry weight. Due to the small number of samples (n < 30), ANOVA was used, and it was also used for the analysis of RT-qPCR results. p < 0.01 was defined as extremely significant and p < 0.05 as significant.

3. Results

3.1. Effect of NopT and NopP on the Nodulation in Soybean

In our previous study, we constructed the single-mutant rhizobial strains HH103ΩNopT and HH103ΩNopP (shown in Figure 1a,b, respectively). Additionally, we constructed the double mutant HH103ΩNopT&NopP using HH103ΩNopT and HH103ΩNopP. PCR was performed to detect the special DNA region in the chromosome of the candidate mutant strains to confirm the successful construction of the mutant rhizobia. As expected, PCR results indicate that the double mutant had acquired Kan-resistance and Spec-resistance gene coding sequences. Compared with the wild-type rhizobia, genomes of both the single and double mutant strains had an inserted DNA fragment (Figure 1b). No DNA fragment was amplified in the wild-type rhizobia when the primers designed for the antibiotic-resistance coding sequence were used. These results confirmed the successful construction of the double mutant HH103ΩNopT&NopP. The NopP and NopT complementary strains were constructed based on relative mutants reported in a previous study [23].

To evaluate the relevance of NopT and NopP in the soybean-rhizobia symbiosis, the nodule numbers were recorded in G. max plants at 30 days post-inoculation (dpi) with the S. fredii strains. The results revealed that the NopT and NopP genes exhibited positive and negative effects on nodulation in SN14, respectively (Figure 2e). Furthermore, the double mutant strain carrying both NopT and NopP mutations exhibited a reduction in nodulation compared with the wild-type strain, indicating a crosstalk between NopT and NopP in the regulation of nodulation. The results indicate that the double mutant had a unique regulatory characteristic compared with the NopT and NopP single mutants. The dry weight of nodules exhibited a similar trend, further supporting the results of the nodule number (Figure 2f). Nodulation dry weight statistics of each group represented 10 plants.

3.2. Identification of DEGs Induced by NopT and NopP

RNA-seq analysis was conducted to identify the genes involved in the crosstalk network of NopP and NopT and to determine the DEGs 36 h after inoculation with various rhizobia (Figure 3a). Venn diagrams were plotted to show the DEGs between soybean plants inoculated with mutants and wild-type rhizobia (Figure 3b). Compared with the mock treatment (MgSO₄), inoculations of HH103ΩNopT, HH103ΩNopP, HH103ΩNopT&NopP, and HH103 obtained 2583, 3017, 2363, and 1826 DEGs, respectively. The number of shared DEGs between plants treated with HH103ΩNopT and HH103ΩNopP, HH103ΩNopT and HH103ΩNopT&NopP, and HH103ΩNopP and HH103ΩNopT&NopP were 563, 660, and 937, respectively. Collectively, the analysis revealed a significant number of DEGs among HH103ΩNopT, HH103ΩNopP, and HH103ΩNopT&NopP, suggesting a gene network redundancy mechanism between NopT and NopP.

3.3. Functional Enrichment of DEGs

To gain initial insights into the functions of the DEGs, GO and KEGG enrichment analyses were conducted. The DEGs were categorized into three classes for the GO enrichment analysis: biological process (BP), cellular component (CC), and molecular function (MF) (Figure 4a–c). The DEGs in plants inoculated with HH103ΩNopT, HH103ΩNopP, and HH103ΩNopT&NopP were significantly enriched in “cellular process”, “cell part”, and “binding” in BP (Figure 4a); “organelle” and “cell” in CC (Figure 4b); and “catalytic activity” and “transporter activity” in MF (Figure 4c). Furthermore, KEGG pathway enrichment analysis revealed that five pathways, including “Fatty acid elongation” “Plant hormone signaling” “Plant–pathogen interaction” and “MAPK signaling” were significantly enriched (q < 0.05). The DEGs in plants inoculated with HH103ΩNopT, HH103ΩNopP, and HH103ΩNopT&NopP were predominantly the genes involved in plant–pathogen interactions (Figure 5).

3.4. Weighted Gene Correlation Network Analysis (WGCNA)

A total of 672 genes with an average fragments per kilobase of exon model per million reads mapped (FPKM) > 1 in the plants inoculated with HH103ΩNopT, HH103ΩNopP, or HH103ΩNopT&NopP were selected for WGCNA. Based on the correlation coefficients of these genes, a gene coexpression network heatmap was plotted (Figure 6a). Modules can be interpreted as gene clusters with high correlation (Figure 6b), and the correlation coefficients between genes in the same cluster are relatively high. Five major gene clusters are marked in various colors in Figure 5a.

KEGG enrichment analysis was performed on five significantly correlated modules (q < 0.05). Black modules were mainly enriched in plant–pathogen interactions, phytohormones, and starch and sucrose metabolism (Figure 7a). Blue modules were mainly enriched in plant hormone signaling, phenylpropane biosynthesis, and plant–pathogen interactions (Figure 7b). The magenta modules were mainly enriched in phenylpropane biosynthesis and glutathione metabolism (Figure 7c). The red modules were mainly enriched in phenylpropane biosynthesis, starch and sucrose metabolism, and phytohormone signaling (Figure 7d). Grey modules were mainly enriched in phenylpropane biosynthesis, protein processing in the endoplasmic reticulum, and phytohormone signaling (Figure 7e). Therefore, starting from the black module, 11 genes with gene significance (GS) and module membership (MM) values >0.98 were selected, and the genes with weight values >0.5 were plotted to form a gene correlation network (Figure 7f). Among the 11 genes, 5 genes were specifically expressed in the roots or nodules, and the genes were screened and plotted to form a gene correlation network (Figure 7g).

3.5. Gene Expression Heatmap

For the heatmap, five genes specifically expressed in the roots or nodules were screened from the black module of WGCNA. In addition, six marker genes were included that have been extensively studied in plants and play a key role in the interaction between plants and pathogens. The FPKM values were obtained from RNA-seq for analysis.

Glyma.03g236300 and Glyma.16g212300 had higher FPKM after inoculation with NopP mutants. Therefore, we expected that these genes might have a more obvious response to NopP. Glyma.05g029800 had the highest FPKM after inoculation with HH103. Therefore, we expected that the infection with HH103 might inhibit the expression of this gene. Glyma.15g026400 exhibited the highest FPKM after inoculation with the NopT&NopP mutant, suggesting that this gene may be regulated by NopT and NopP at the same time. However, this gene has a weak response to a single effect factor. Glyma.19g098600, Glyma.17g217000 (GmCHIT5), and Glyma.02g138800 (GmMAPK2) exhibited higher FPKM after inoculation with the NopT mutant. Moreover, Glyma.19g098600 exhibited a higher FPKM after inoculation with the NopT&NopP double mutant and a higher FPKM after inoculation with the NopP mutant than after inoculation with HH103 (Figure 8). Therefore, it was speculated that the expression of Glyma.19g098600 (PBS1) may be involved in the crosstalking network of NopT and NopP.

3.6. Analysis of Difference in the Gene Expression after Inoculation with Various Rhizobia

RT-qPCR was performed to identify the candidate gene expression pattern in response to NopT and NopP single mutants and the NopT&NopP double mutant. Glyma.08G158100, Glyma.20g133200, Glyma.11g157100 (GmCML19), and Glyma.06g045400 (GmSCTF-1) were downregulated and Glyma.19G098600 (GmPBS1) was upregulated after inoculation with HH103ΩNopP, HH103ΩNopT, or HH103ΩNopT&NopP (Figure 9). This result further supported that the absence of NopT and NopP can affect the expression of PBS1. The result also indicate that the effect of crosstalk between NopT and NopP on nodulation may be related to PBS1.

3.7. Subcellular Localization of GmPBS1 in Tobacco

To assess whether there is a direct interaction between NopT and GmPBS1, we performed subcellular localization analysis using N. benthamiana. Both NopT and GmPBS1 were located in the cell membrane and nucleus (Figure 10). This supported the speculation that NopT and GmPBS1 might directly interact in the cell plasma membrane.

GmPBS1-cYFP and NopT-nYFP, as well as GmPBS1-nYFP and NopT-cYFP were mixed and injected into tobacco leaves. The two incomplete fluorescent reporter protein fragments in the cell membrane would be close to each other and would form an active fluorescent protein to produce fluorescence if NopT and GmPBS1 could interact with each other. A BiFC assay using split YFP complementation in N. benthamiana leaves indicates that NopT and GmPBS1 can interact (Figure 11), and the interaction occurs in the cytoplasm (Figure 11).

3.8. Effect of GmPBS1 Overexpression on the Nodulation

The soybean hairy roots transformed with pSOY1-35S: GmPBS1: GFP were transferred into vermiculite for 3 days and incubated with HH103, HH103ΩNopT, HH103ΩNopP, or HH103ΩNopT&NopP. GmPBS1 was successfully overexpressed as revealed by RT-qPCR (Figure S1). The nodule number of plants with pSOY1-35S: GmPBS1:GFP and inoculated with HH103, HH103ΩNopP, or HH103ΩNopT&NopP was significantly lower than that of control plants (containing empty vectors; Figure 12b). After inoculation with HH103ΩNopT, the number of nodules in plants with pSOY1- 35S: GmPBS1: GFP did not significantly change compared with those in control plants.

Therefore, the plants with GmPBS1 overexpression exhibited a reduced number of nodules after HH103 inoculation compared with wild-type plants. It can be speculated that PBS1 overexpression had an inhibitory effect on the nodulation in soybean. However, in the plants inoculated with the NopT single mutant, the nodule number did not significantly change after PBS1 overexpression compared with the control. Combining with the results of the BiFC assay, it can be concluded that the expression of PBS1 may require the presence of NopT. However, plants with PBS1 overexpression exhibited a reduced number of nodules after inoculation with the NopT&NopP double mutant. These results confirmed that GmPBS1 is a hub gene involved in the crosstalking network of NopP and NopT, triggering gene signaling transduction in soybean.

4. Discussion

NopT and NopP identified in some rhizobia can exhibit positive and negative effects, respectively, on the nodulation in soybean [18]. However, the double mutant HH103ΩNopT&NopP induced less nodules in soybean than wild-type HH103 (Figure 2e). These results indicate that the signaling triggered by NopT and NopP involved crosstalking. The RNA-seq analysis was performed to identify the candidate genes involved in the crosstalking between NopT and NopP. Interaction between NopT and GmPBS1 was confirmed using the BiFC analysis (Figure 10). In Arabidopsis, PBS1 was cleaved by AvrPphB; the cleavage of PBS1 can trigger the RPS5-activated programmed cell death [14]. Moreover, GmPBS1 can be cleaved by NopT [10]. GmPBS1 overexpression in soybean impaired the nodule formation. The presence of NopT might be essential for active HR in the plants. HR refers to the local cell death within hours of an infection to limit the proliferation and spread of a pathogen. The symbiosis between soybean and rhizobia is beneficial for the plant [24,25]. No HR phenotype was observed in soybean; however, the signaling should be present. NopP seemed to play a negative role on nodule formation; however, the NopT and NopP double mutant formed less nodules than HH103 and NopT and NopP single mutants. Overexpression of GmPBS1 could overcome the effect of the absence of NopP. This indicates that the NopP-triggered signaling pathway should be at the upstream of GmPBS1. NopP can determine the symbiotic incompatibility depending on the Rj2 genotype. In addition, the variation in NopP can be recognized by the Rj2-mediated incompatibility in soybean. GmNNL1 could interact with NopP from Bradyrhizobium diazoefficiens USDA110 [26]. GmNNL1 haplotype 1 can recognize NopP from B. diazoefficiens USDA110/USDA6, but not from USDA122, to trigger the HR in N. benthamiana leaves. In this study, NopP from HH103 was different from NopP in B. diazoefficiens USDA110/USDA6/USDA122. As the interaction between NopP and GmNNL1 is direct, GmNNL1 probably works in different ways to mediate the interaction with rhizobia.

In this study, we identified the expression pattern of three genes, namely, GmNNL1, GmPBS1, and GmRPS5, in response to NopT and NopP. RNA-seq data were used to detect their expression pattern. The results indicate that NopT and NopP can induce the expression of these genes. Moreover, the study reported that the double mutant of NopT and NopP can induce higher expression levels of GmRPS5 and GmNNL1 at 3 hpi (hours post infection) but not at 6 and 36 hpi (Figure S1). This suggested that NopT and NopP may play a pivotal role in inhibiting the defense response of soybean and promoting nodule formation during the early stages. However, further research is needed to elucidate the underlying regulation mechanism of NopT and NopP in symbiotic signaling transduction.

To detect the signaling pathway triggered by NopP and NopT, it is recommended to use a mutant with knockdown and overexpression of GmPBS1 and the downstream protein GmRIN4 [27,28,29]. These R proteins are known to respond to the T3SS of pathogens [28]. However, it is important to note that the rhizobial T3SS is used to establish a symbiotic relationship, which is different from pathogenic infection. Therefore, multiple T3Es should be used to modulate the host immunity at an appropriate level that allows rhizobial infection. Rhizobia use various T3Es to communicate with genes within the host for symbiotic purposes.

In conclusion, the interaction between NopT and GmPBS1 plays a role in regulating the immune response of soybean to rhizobial infection. Additionally, NopP can act as a partner to NopT to maintain an appropriate level of immunity, which allows effective symbiotic signaling communication with soybean genes. Moreover, GmPBS1 is a hub gene involved in the crosstalking network of NopP and NopT, regulating the soybean response to rhizobium.