Introduction

The genus Phytophthora comprises over 100 species of parasitic oomycetes, the vast majority of which can cause severe diseases of various agriculturally important plants such as potato and soybean. Among them, Phytophthora sojae has a narrow host range and is primarily restricted to soybean, although some plant species such as lupin are also reported to be susceptible to the pathogen (Tyler 2007; Xiong et al. 2014). Phytophthora root and stem rot (PRR) caused by P. sojae is one of the most devastating and widespread diseases throughout many soybean-growing regions and is responsible for annual economic loss of 1–2 billion dollars (Sheng et al. 2015; Tyler 2007; Xiong et al. 2014). In recent years, PRR incidence has been increasing in several soybean-producing areas around the world, when soil conditions are suitable for pathogen development (Xiong et al. 2014). P. sojae can infect seeds, seedlings, and plants in all growth stages, causing seed decay, collapse of emerging seedlings, and root and stem rot in established plants. Presently, PRR is primarily managed based on host-mediated resistance provided by resistance (R) genes. In the P. sojae-soybean pathosystem, several types of resistance have been described, including race-specific resistance conferred by single dominant Rps (resistance to P. soja) genes and partial resistance provided by multiple genes (Qutob et al. 2000; Sugimoto et al. 2012). The resistance mediated by Rps genes encoding nucleotide-binding leucine-rich repeat (NB-LRR)-type proteins is race-specific and follows the gene-for-gene model (Bhattacharyya et al. 2005; Lin et al. 2014; Yin et al. 2013). To date, 18 Rps genes/alleles from soybean genetic sources have been identified and most of them have been mapped on the soybean genome (Lin et al. 2014; Sugimoto et al. 2012; Tyler 2007). Among them, seven Rps genes including Rps1a, Rps1b, Rps1c, Rps1 k, Rps3a, Rps6, and Rps8 are employed in commercial soybean production. However, PRR virulence pathotypes are very complex, with at least 55 P. sojae physiological races identified, and many virulence combinations occur because of mutations and sexual outcrossing (Abney et al. 2007; Burnham et al. 2003; Ryley et al. 1998; Tyler 2007). The diversity and variability of P. sojae races both within and between fields make it challenging to effectively control the pathogen based only on the Rsp gene-mediated resistance. Moreover, the use of resistant soybean cultivars with single Rsp genes can also promote the accumulation of new virulent pathogen strains, which is particularly evident in fields under continuous cultivation (Wang et al. 2006). Furthermore, although partial resistance which is inherited as quantitative trait loci (Wang et al. 2012), can provide more durable protection against P. sojae, it does not always prevent significant crop loss caused by the pathogen (Burnham et al. 2003). Therefore, new strategies for enhancing PRR resistance in soybean are still greatly needed.

To date, several attempts have been made to increase resistance of soybean to P. sojae infection by expressing foreign genes which confer resistance to fungal pathogens (Dong et al. 2015; Du et al. 2018; Fan et al. 2015). Overexpression of Gly m 4 l, a pathogenesis-related class 10 protein which plays an important role in the soybean defense system against P. sojae, was shown to enhance resistance to PRR (Fan et al. 2015). Expression of a novel ethylene response factor, GmERF5, in soybean significantly enhanced resistance to P. sojae and upregulated the expression of pathogenesis-related PR10, PR1-1, and PR10-1 genes involved in systemic acquired resistance (SAR) in plants (Dong et al. 2015). Recently, Du et al. (2018) showed that expression of the harpin-encoding gene hrpZm from Pseudomonas syringae enhanced resistance to P. sojae infections in transgenic soybean.

Harpin proteins encoded by the hrp (hypersensitive response and pathogenicity) genes are produced by several gram-negative plant pathogenic bacteria (Tampakaki et al. 2010). Harpins are secreted through type III secretion system and are mostly localized to the extracellular space in plant tissues, unlike bacterial effector proteins that act inside plant cells (Choi et al. 2013). Some harpins function as a part of translocator complexes involved in the translocation of effectors into plant cells (Choi et al. 2013). When harpins are directly applied to plants or expressed intracellularly, they trigger diverse beneficial processes such as induction of defense responses and enhancement of plant growth (Choi et al. 2013; Pavli et al. 2011; Peng et al. 2003; Shao et al. 2008; Sohn et al. 2007). Overexpression of harpin-encoding genes can enhance resistance to diseases and insects in plants such as tobacco, rice, rape, and cotton (Fu et al. 2014; Li et al. 2004; Miao et al. 2010; Rong et al. 2010; Wei et al. 1992). Recently, it has been reported that harpin overexpression in rice can even improve plant tolerance to abiotic stresses such as drought (Zhang et al. 2011). Multiple functions of harpin-encoding genes in plant development and defense against biotic and abiotic stresses make them a promising tool for enhancing plant disease resistance and improve crop yield through genetic engineering.

The harpinXooc-encoding hrf2 gene first isolated from the rice pathogenic bacterium Xanthomonas oryzae pv. oryzicola (Rong et al. 2010) has been shown to elicit hypersensitive response (HR)-related programmed cell death (PCD) in non-host plants and significantly enhance resistance to the fungal pathogen Sclerotinia sclerotiorum (Rong et al. 2010). In the present study, we generated transgenic soybean plants expressing hrf2 and assessed their resistance to the oomycete pathogen P. sojae.

Materials and methods

Agrobacterium tumefaciens-mediated transformation and generation of transgenic plants

A 414-bp fragment of the coding hrf2 region (as shown in Fig. S1) from X. oryzae pv. oryzicola strain RS105 (kindly provided by Prof. Xuewen Gao, Nanjing Agricultural University, China) was cloned into the SacI/XbaI sites of the compatible pCAMBIA3300 plasmid to generate the recombinant binary vector pCAMBIA3300-hrf2 (Fig. 1a). In the resulting construct, the total length of the T-DNA is 3.7 kb which covers the hrf2 coding region between the cauliflower mosaic virus (CaMV) 35S promoter and a polyA site, and the bar gene expression cassette which confers resistance to the herbicide glufosinate. The construct was then mobilized into A. tumefaciens strain EHA105. Soybean cultivar Shennong 9 which is susceptible to P. sojae, was used to generate transgenic soybean plants using the Agrobacterium-mediated transformation as described previously (Yang et al. 2017). The generated transgenic plants tolerant to glyfosinate (5 mg L−1) were transplanted into a greenhouse and analyzed by molecular screening.

Fig. 1
figure 1

Generation and molecular analysis of transgenic soybean plants. a Schematic diagram of the pCAMBIA3300-hrf2 vector for Agrobacterium- mediated genetic transformation. The coding region of hrf2 was inserted between the constitutive promoter CaMV 35S and a polyA site. The construct also contained a selectable marker bar for selection of glufosinate-resistant transgenic plants. Solid bars indicate probes (bar and hrf2) used for Southern blotting. Small arrows indicate primers used for PCR screening. b, c Screening of transgenic plants using LibertyLink strip (b) and PCR detection (c). M, DNA marker (2 K); Ct + , positive control; NC, non-transformed plants; Bk, ddH2O; 1–12, T0 transgenic plants. d, e Southern blotting analysis of transgenic plants. Genomic DNA was digested with SacI or HindIII, and hybridized with the corresponding DIG-labeled hrf2 (d) and bar (e) probes, respectively. M, DNA marker (15 K); Ct + , positive control; NC, non-transformed plants; 1–4, T2 transgenic events L13, L44, L16, and L32. f Western blotting analysis of transgenic plants using a polyclonal antibody raised against the recombinant HRF2 protein; M, protein molecular weight markers; Ct + , recombinant HRF2 protein expressed in E. coli. NC, non-transformed plants; 1–4, transgenic events L13, L44, L16 and L32

Full size image

Molecular analysis of transgenic soybean plants

Transgenic plants were first screened using LibertyLink® strips (EnviroLogix Inc., USA) and polymerase chain reaction (PCR) as described previously (Yang et al. 2018). Primers HRF2-F1 (5′-GTGGATTGATGTGATATCTCCACTG-3′) and HRF2-R1 (5′-CAAAGTCGCCGCCGCTGCTG-3′) specific for the CaMV 35S promoter and hrf2, respectively, were used to amplify a 492-bp fragment to confirm transgene presence. Glufosinate spray (1500 mg L−1) and PCR analysis were used to select positive progenies (generations T1 to T5). For Southern blotting analysis, total DNA was extracted from T2 transgenic and non-transformed control (NC) plants using a modified cetyltrimethylammonium bromide method (Telzur et al. 1999). Thirty micrograms of genomic DNA were digested with SacI or HindIII, subjected to electrophoresis in 0.8% agarose gels and subsequently transferred to positively charged nylon membranes (GE Amersham, USA). The transformation vector and non-transformed plants were used as positive and negative controls, respectively. Hybridization probes were prepared by amplifying the 414-bp hrf2 and 441-bp bar coding regions using primers HRF2-F2/HRF2-R2 (5′-ATGAACTCTTTGAACACAC. AATTC-3′/5′-TTACTGCATTGATGCGCTGTCGTTC-3′) and BAR-F1/BAR-R1 (5′-GCACCATCGTCAACCACTACATCGAG-3′/5′-TGAAGTCCAGCTGCCAGAAACCCAC-3′), respectively, and labeling them with digoxigenin (DIG)-High Prime (Roche, Germany). Hybridization was carried out at 42 °C for 12–16 h and staining was performed at room temperature with BCIP/NBT as substrate.

For western blotting analysis, leaf samples (0.1 g) were collected from T3 transgenic and NC plants and total proteins were extracted with buffer containing 100 mM NaCl, 10 mM EDTA, 200 mM Tris–HCl, 0.05% Tween-20, 0.1% SDS, 14 mM β-mercaptoethanol, 400 mM sucrose, and 2 mM phenylmethanesulfonyl fluoride. After sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) in 12% gels, proteins were transferred onto PVDF membranes (GE Healthcare, USA) and probed with a rabbit polyclonal antibody against the recombinant HRF2 protein (with molecular weight of 13.63 kDa) expressed in E. coli (1:500 dilution, GenScript Co., Ltd., Nanjing, China), and then with horseradish peroxidase (HRP)-labeled goat-anti-rabbit IgG (1:5000 dilution; Abcam, UK) at room temperature for 4 h. After extensive washing, protein bands were visualized using BiodlightTM Western Chemiluminescent HRP substrate (Bioworld Technology, Inc. USA).

Evaluation of transgenic soybean plant resistance to P. sojae

P. sojae race 1, a prevalent strain in China, was cultured in 10% V8 media at 25 °C in the dark as described previously (Erwin and Ribeiro 1996), and used to infect unifoliate soybean leaves fully expanded in the greenhouse. Twenty hypocotyls of 15-day-old soybean seedlings of each transgenic and NC plants were inoculated with macerated mycelia (inoculum) of P. sojae using a standard hypocotyl inoculation method (Abney et al. 2007). The inoculated plants were maintained for 18–24 h in the dark, and then switched to the 18-h light/6-h dark regime at 25 °C for symptom development. PRR symptoms were evaluated approximately 7–10 days after inoculation and the percentage of surviving seedlings was calculated. The same number of non-transformed plants were mock-inoculated (wounded) without the pathogen and used as controls. All experiments were performed independently three times with 20 plants per replicate.

Expression analysis of defense-related genes

Quantitative real time (qRT) PCR was performed to assess transcription levels of defense-related SA-dependent genes such as GmPR1a (AF136636), GmPR2 (β-1, 3-glucanase, M37753), GmPR3 (chitinase class I, AF202731), GmPR5 (pathogenesis-related group 5 protein, BU765509), GmPR12 (defensin, M37753), and GmPAL (phenylalanine ammonia lyase, X52953), JA-dependent genes such as GmAOS (allene oxide synthase, DQ288260) and GmPPO (polyphenol oxidase, EF158428), and HR-related genes such as GmNPR1-1 (FJ418594), GmNPR1-2 (FJ418596), GmSGT1 (NM_001249656), and GmRAR1 (FJ222386). Three fresh leaves from randomly selected T3 transgenic and NC plants were harvested and frozen in liquid nitrogen at 0, 4, 8, and 12 h after inoculation with P. sojae. Total RNA was isolated using the EasyPure PlantRNA Kit (TransGen Biotech, China) and treated with DNase I to remove genomic DNA according to the manufacturer’s protocol. RNA integrity was confirmed by agarose gel electrophoresis and its concentration quantified by spectrophotometry (Nanodrop ND-1000, USA). First-strand cDNA was synthesized with the ThermoScript RT-PCR system (Invitrogen, USA) and used as a template for qRT-PCR performed at the following conditions: 50 °C for 2 min, 95 °C for 10 min, and 45 cycles of 95 °C for 2 min, 62 °C for 30 s, and 72 °C for 30 s. Primers used in this study are listed in Supplementary Table S1. Relative transcription levels of the target genes were calculated relative to that of the constitutively expressed native soybean gene GmACT11 (GenBank No. BW652479) using the 2ΔΔCt method as previously described (Yang et al. 2018). The results were statistically analyzed based on at least three independent biological replications.

Agronomic traits of transgenic plants in field conditions

For evaluation of the agronomic performance, the transgenic plants were planted in the field station in Jilin province, China. Three replicates of triple-row plots per transgenic event were grown using a randomized block design, with each row consisting of twenty-thirty plants. At maturity, ten plants from each transgenic event and NC control (without P. sojae inoculation) were randomly sampled and evaluated for agronomic traits such as plant height, branch, node, pod, and seed numbers, and 100-seed weights.

Statistical analysis

Least significant differences (LSD) between mean values were analyzed by the t test at P = 0.05 or 0.01 using SPSS software (v. 17.0) and each transgenic event was compared with the corresponding NC plants.

Results

Generation of transgenic soybean plants and molecular screening

Sixty-four plants of soybean genotype Shennong 9 were produced from glufosinate-tolerant green shoots following the Agrobacterium-mediated transformation. Among these, thirty-seven plants were confirmed positive by LibertyLink® strip detection and PCR analysis. LibertyLink strip detection showed expression of the bar gene with two red lines appearing simultaneously for transgenic plants (Fig. 1b). PCR analysis of LibertyLink-positive plants showed amplification of the expected 492-bp fragment corresponding to the CaMV 35S-hrf2 coding sequence (Fig. 1c). After preliminary PRR resistance screening, four independent transgenic events L13, L16, L32 and L44 which exhibited higher resistance, were selected for further molecular analysis and resistance evaluation. Integration of the foreign gene was confirmed by Southern blotting with both DIG-labeled hrf2 and bar probes. The results showed that hybridization signals appeared in the selected transgenic plants and all the bands were greater than the expected 1.56-kb (SacI-hrf2-RB, Fig. 1d) and 3.44-kb (LB-bar-hrf2-HindIII, Fig. 1e) fragments, respectively. In contrast, no hybridization signals were observed in NC plants. Moreover, low copy numbers of T-DNA insertions (single copies for L13 and L32, two copies for L16, three copies for L44) were detected in the four independent transformation events. Western blotting analysis revealed the presence of a 13.63-kDa polypeptide corresponding to the HRF2 monomer in transgenic plants, whereas no signals were observed in NC plants (Fig. 1f), confirming expression of the HRF2 protein in transgenic events.

Transgenic plants exhibited stably enhanced resistance to PRR under greenhouse conditions

Phenotypic reactions of the transgenic events and their corresponding NC plants to P. sojae were evaluated in the greenhouse. At 7 days after inoculation with P. sojae, typical PRR symptoms, including necrotic and wilting leaves and plant death, were observed in most NC plants of each experimental replicate. In contrast, most of transgenic plants showed no visible symptoms or only slightly chlorotic leaves and developed normally, and mock-inoculated transgenic seedlings were asymptomatic (Fig. 2a). The percentage of surviving seedlings calculated to assess disease severity caused by P. sojae inoculation was higher in transgenic events (73.35–92.86%) compared to NC plants (38.50–40.37%) (Fig. 2b), although slight variations in the proportion of surviving seedlings for each transgenic event across the three replicates were observed, which might be attributed to minor differences in environmental conditions. Furthermore, transgenic events consistently exhibited enhanced resistance to P. sojae compared to NC over three generations (from T3 to T5) (Fig. 2b). To analyze the influence of hrf2 expression on agronomic characteristics of transgenic plants, we evaluated several traits such as leaf morphology, flower and hilum color, plant and podding height, node numbers, weight of 100 seeds, and maturity period under field conditions without P. sojae inoculation. The data showed that there were no significant differences between transgenic and NC plants, indicating no visible influence on agronomic performance caused by hrf2 expression (Table 1).

Fig. 2
figure 2

PRR resistance evaluation of transgenic soybean plants expressing hrf2.a Resistance response of T3 transgenic plants to P. sojae infection. b Increased survival rate of three consecutive generations (T3–T5) of transgenic plants inoculated with P. sojae. NC, non-transformed plants; mock, non-transformed plant wounded without P. sojae inoculation. L13, L44, L16, and L32, transgenic events. The data are presented as the mean ± SE of three independent experiments (20 plants per each replicate); **P < 0.01 compared to corresponding NC plants

Full size image
Table 1 Agronomic performance of transgenic soybean events in the field
Full size table

hrf2-expression upregulated multiple defense-related genes in transgenic plants

To investigate the molecular mechanisms underlying enhanced resistance to P. sojae in hrf2-expressing plants, we analyzed the transcription of 12 defense-related genes after P. sojae infection by qRT-PCR. First, we evaluated the time course of hrf2 expression in transgenic plants after inoculation with P. sojae. The results showed that hrf2 mRNA levels peaked at 4 h and remained high at 8 and 12 h post inoculation in all three examined transgenic soybean plants (Fig. 3a), although the abundance of the hrf2 transcript was slightly different among different transgenic events. Then, we compared mRNA levels of six SA-dependent genes (GmPR1, GmPR2, GmPR3, GmPR5, GmPR12 and GmPAL), two JA-dependent genes (GmAOS and GmPPO), and four HR-related genes (GmNPR1-1, GmNPR1-2, GmSGT1, and GmRAR) in transgenic and NC plants 4 h after P. sojae challenge. The results showed that the expression of all the 12 defense genes was significantly upregulated in response to P. sojae infection in transgenic plants but remained unchanged or increased to a lesser extent in NC plants (Fig. 3b–d). Especially high expression levels of the 12 defense-related genes were observed in transgenic event L13, which was consistent with the results of PRR resistance evaluation. Our data suggest that the expression of the hrf2 gene in soybean could elicit multiple resistance responses mediated by different signaling pathways, thus enhancing plant resistance to P. sojae infection and protecting soybean against PRR.

Fig. 3
figure 3

Relative expression of defense-related genes in transgenic soybean plants after P. sojae inoculation. a Expression changes of hrf2 mRNA in transgenic plants. Total RNA was extracted from fully grown leaves of T3 transgenic plants at 0, 4, 8, and 12 h after inoculation with P. sojae and mRNA levels of hrf2 were calculated relative to that of the GmACT11 gene. b–d Relative expression of six salicylic acid-dependent genes (b), two jasmonic acid-dependent genes (c), and four hypersensitive response-related genes (d) at 4 h after P. sojae inoculation. NC, non-transformed plants; L13, L16, and L32, transgenic events. The data are presented as the mean ± SE of three biological replicates; **P < 0.01 compared to corresponding NC plants

Full size image

Discussion

Harpin proteins are known to increase resistance of plants to fungi and insects, improve their tolerance to abiotic stresses, and promote growth (Fu et al. 2014; Li et al. 2004; Miao et al. 2010; Rong et al. 2010; Wei et al. 1992; Zhang et al. 2011). It has been reported that the introduction of the harpinXooc-encoding hrf2 gene in rapeseed could effectively enhance its resistance to S. sclerotiorum, an oxalate-secreting necrotrophic fungal pathogen (Rong et al. 2010). In the present study, we showed that the transgenic plants constitutively expressing the hrf2 gene also exhibited significantly enhanced resistance to the oomycete pathogen P. sojae, as demonstrated by the hypocotyl inoculation assay. Compared with non-transformed plants, the transgenic events showed milder PRR symptoms and had higher survival rates. These results are consistent with the multiple functional activity of harpin proteins which were reported to induce plant resistance against diverse plant pathogens including fungi, bacteria, and viruses either after external application or stable expression (Choi et al. 2013; Pavli et al. 2011; Peng et al. 2003; Shao et al. 2008; Sohn et al. 2007). The potential function of hrf2 in plant defense against P. soja and other pathogens such as S. sclerotiorum makes this gene a promising candidate for engineering disease resistance in soybean breeding.

Mechanisms underlying increased resistance of harpin-expressing plants to various infections have to be characterized for future combination with other defense and resistance components. Phytohormones SA and JA are known to play crucial roles in regulating defense mechanisms in plants (Pieterse et al. 2009; 2012; Robert-Seilaniantz et al. 2011; Sugano et al. 2014). It was reported that an extensive crosstalk existing between phytohormone-controlled signaling pathways contributes to the induction of strong defense responses in plants (Koornneef and Pieterse 2008). SA plays a crucial role in the activation of defense mechanisms in response to biotrophic and hemibiotrophic pathogens, whereas JA is typically associated with defense against necrotrophic pathogens, and the respective pathways are often mutually antagonistic. However, synergistic interactions have also been reported in some pathosystems. In this study, we observed that six SA-dependent genes (GmPR1, GmPR2, GmPR3, GmPR, GmPR12, and GmPAL) and two JA-dependent genes (GmAOS and GmPPO) were simultaneously upregulated in hrf2-expressing transgenic plants compared with NC plants which showed no or little transcriptional changes in response to P. sojae infection. Moreover, transgenic plants also exhibited significantly increased expression of genes involved in HR signaling. These findings suggest that the enhanced resistance to P. sojae infection in transgenic soybean constitutively expressing hrf2 may be attributed to significant transcriptional upregulation of a panel of genes involved in different plant defense-related signaling pathways. In addition, it was also reported that external application of harpins such as HrpNEa to plants or constitutive expression of harpin-encoding genes promote plant growth through the ethylene-mediated signaling pathway (Chuang et al. 2010; Dong et al. 2004; Oh and Beer 2007). However, in this study, we did not observe significant differences in agronomic traits such as plant height in transgenic soybean events compared to NC plants in our preliminary field experiment. This discrepancy may be attributed to distinct signaling pathways mediated by HRF2 and other harpin proteins. The influence of expression of hrf2 on the agronomic performance of transgenic soybean needs to be further conducted larger field experiments.

Compared to exogenous application of harpins which can independently activate HR-related PCD and the SA-mediated SAR pathway in plants (Choi et al. 2013; Peng et al. 2003), endogenous expression of harpins may elicit more complicated responses to pathogens. Thus, transgenic plants expressing hrp genes show stronger reactivity to pathogens manifested by substantial increase in the expression of defense-related genes mediated by multiple signaling pathways (Pavli et al. 2011; Peng et al. 2004; Shao et al. 2008; Sohn et al. 2007), however, it should be noted that gene expression patterns might be distinct for different harpin proteins (Du et al. 2018). For example, although NPR1 is thought to be a key transcriptional regulator of multiple signaling pathways involved in plant defense (Peng et al. 2004), no changes in npr1 gene expression were detected in hrpN-transgenic plants inoculated with B. cinerea (Sohn et al. 2007). In contrast, in the present study, we observed a significant increase of npr1 expression in transgenic soybean after P. sojae infection. Similarly, transcriptional upregulation of the npr1 gene in response to pathogen infection was reported in transgenic tobacco plants expressing hpa1Xoo but not in those expressing hpaGEP (Peng et al. 2004; Sohn et al. 2007). This difference in npr1 regulation among harpin-expressing plants may be due to distinct receptors targeted by harpins of various origins (Miao et al. 2010).

In conclusion, our study suggests that the expression of the hrf2 gene encoding harpinXooc in soybean enhances plant resistance to P. sojae infection and protects against PRR through transcriptional upregulation of multiple defense-related genes. However, it should also be noted that transgenic soybean event expressing hrf2 are not entirely immune to P. sojae under our experimental conditions. For sustainability of PRR resistance, it is important to rely on multiple components of defense by pyramiding complementary resistance genes thorough breeding or genetic transformation.