Ethylene response pathway modulates attractiveness of plant roots to soybean cyst nematode Heterodera glycines

Plant parasitic nematodes respond to root exudates to locate their host roots. In our studies second stage juveniles of Heterodera glycines, the soybean cyst nematode (SCN), quickly migrated to soybean roots in Pluronic F-127 gel. Roots of soybean and non-host Arabidopsis treated with the ethylene (ET)-synthesis inhibitor aminoethoxyvinylglycine (AVG) were more attractive to SCN than untreated roots, and significantly more nematodes penetrated into roots. Moreover, Arabidopsis ET insensitive mutants (ein2, ein2-1, ein2-5, ein3-1, ein5-1, and ein6) were more attractive than wild-type plants. Conversely, the constitutive triple-response mutant ctr1-1, was less attractive to SCN. While ET receptor gain-of-function mutant ein4-1 attracted more SCN than the wild-type, there were no significant differences in attractiveness between another gain-of-function ET receptor mutant, etr1-3, or the loss-of-function mutants etr1-7 and ers1-3 and the wild type. Expression of the reporter construct EBS: β-glucuronidase (GUS) was detected in Arabidopsis root tips as early as 6 h post infection, indicating that ET signaling was activated in Arabidopsis early by SCN infection. These results suggest that an active ET signaling pathway reduces root attractiveness to SCN in a way similar to that reported for root-knot nematodes, but opposite to that suggested for the sugar beet cyst nematode Heterodera schachtii.

hormones including auxin, ethylene (ET), jasmonic acid (JA), salicylic acid (SA), brassinosteroids (BR), gibberellins (GA), and abscisic acid (ABA) [22][23][24] . A number of studies have indicated that the signaling pathways of SA, JA, BR, ABA, and ET are involved in the defense response of host plants to nematodes [25][26][27][28][29][30] . Auxin, ET and GA have also been reported to play a positive role in the early stages of development of nematode feeding cells [31][32][33] . ET is recognized to play a role in nematode attraction, migration, feeding site formation, and host defense during early stages of nematode infection 32,[34][35][36] , but the role may differ for the interaction of plants with cyst nematode and RKN. For example, an intact ET pathway is prerequisite for plant defense to RKN 27,37 . In contrast, the ET pathway facilitates cyst nematode parasitism as evidenced by the significant reduction of cyst nematodes inside the roots of ET-insensitive mutants or wild type plants treated with ET inhibitors 31,36 . ET is also known to play an important role in H. glycines development 38,39 .
Over the past few decades, the ET signal transduction pathway has been extensively investigated in the model plant Arabidopsis. Ethylene is perceived by five membrane receptors ETHYLENE RESPONSE1 (ETR1), ETHYLENE RESPONSE SENSOR1 (ERS1), ETR2, ERS2, and ETHYLENE INSENSITIVE4 (EIN4), which are negative regulators of ET responses; ET binding to these receptors represses the negative regulation 40-43 . In the absence of ET, the active receptors recruit another negative regulator of ET signaling, Constitutive Triple Response1 (CTR1), to phosphorylate the C-terminal domain of EIN2. and promote the binding of EIN2 to the F-box proteins EIN2 TARGETING PROTEINS, ETP1 and ETP2. This interaction leads to the ubiquitylation and degradation of EIN2 by 26S proteasome, which subsequently represses the EIN3/ ETHYLENE-INSENSITIVE3-LIKE (EIL)-dependent transcriptional cascade [44][45][46][47] . EIN2 is a central positive regulator of ethylene responses, and its null mutant ein2 is completely insensitive to ET 48 . EIN2 can directly activate the EIN3/EIL1 transcription factors, which results in EIN3/EIL1 binding to the promoters of ET response genes ERF1 or other downstream genes to activate or repress their expression 49,50 , thereby regulating ET responses in plants.
While several studies have focused on the role of hormones in the later stages of nematode infection, only a few studies have investigated their contribution to mediating nematode attraction and host-seeking behaviors. Using Arabidopsis mutant lines in ET signaling and perception, one study concluded that active ET signaling played a positive role in the attraction of the sugar beet cyst nematode (SBCN) Heterodera schachtii to Arabidoposis 32 . However, in another study, ethylene signaling was found to negatively affect attraction of the RKN, M. hapla 35 .
In the present study, we investigated the role of the ET signaling pathway in the attraction of H. glycines to soybean and the non-host plant Arabidopsis by using a Pluronic F-127 (PF-127) gel attraction assay 20,35 . We found that soybean and Arabidopsis roots treated with the ET-synthesis inhibitor aminoethoxyvinylglycine (AVG) were more attractive to J2 than untreated roots. Our studies of Arabidopsis mutants in ET perception and signaling further suggested that an active signaling pathway reduces attractiveness of plant roots to SCN in a similar way as to RKN, but opposite to the results reported for H. schachtii.

Results
Attraction of Heterodera glycines to soybean roots. Second-stage juveniles of H. glycines were observed to move toward soybean roots within 1 h post exposure in the PF-127 gel attraction assay. Nematode J2 close to root tips moved to the root surface and started to penetrate into the roots. The numbers of J2 touching the root tips of soybean cv. Dongsheng 1 at 2 h and 3 h post exposure were significantly greater than those at 1 h, 4 h, and 6 h post exposure (Fig. 1). The decline in numbers at the latter time points was due to nematode penetration into the roots. Therefore, we selected the 2-h time point for the attraction assay in the subsequent experiments.
We proposed to treat soybean roots with the ET-synthesis inhibitor AVG prior to the attraction assay as a strategy to evaluate whether ET signaling played a role in attractiveness of soybean roots to H. glycines. To Figure 1. Attraction of Heterodera glycines to soybean root tips. The number of J2 touching the terminal 5 mm of the roots of soybean seedlings was counted at 1, 2, 3, 4, and 6 h after placing root tips in 2 ml Pluronic F-127 gel containing 200 J2. Bars represent the mean ± SE of one representative experiment (n = 24). Bars with different letters indicate significant differences (P < 0.05, Tukey's t Test). The experiments were repeated three times with similar results.
confirm that AVG treatment of soybean roots under our conditions had the anticipated effect, the expression of ethylene-response marker genes GmERF5 and GmERS2 was measured by qRT-PCR after 6, 12 and 24 h of treatment with AVG. GmERF5 transcript levels were significantly reduced compared to untreated roots at 12 and 24 h after AVG treatment, whereas GmERS2 levels were reduced at all three time points (Fig. 2). These results indicated the AVG-treatment down-regulated expression of the ethylene-response pathway in soybean roots.
More nematodes were attracted to soybean root tips that had been pretreated with AVG for 24 h than to untreated roots (Fig. 3a). To exclude the possibility that the increased attractiveness of AVG-treated roots was the result of nematode chemotaxis to AVG, we tested the response of J2 to AVG and found that there was no  difference in nematode chemotaxis toward AVG compared to a water control (data not shown). We also treated soybean roots with the ET analog ethephon (ETH) 27,36,51 , but no significant difference was seen between the number of nematodes touching the ETH-treated and water-treated root tips (Fig. 3a,b). The number of nematodes inside root tips treated with AVG or ETH was counted at 6 h after assay start and following staining with acid fuchsin. Exogenous application of AVG resulted in a significant increase in nematode numbers inside soybean root, while no change was found following ETH treatment (Fig. 3c,d).
Attraction of Heterodera glycines to Arabidopsis roots. We tested whether H. glycines was attracted to roots of the non-host plant Arabidopsis. The number of SCN J2 touching root tips of Arabidopsis ecotypes Col-0, Ler and Ws was counted at 1, 2, 3, 4, 6, and 9 h post exposure. The greatest (P < 0.05) number of nematodes touching the root of Col-0 was detectable at 2 h ( Fig. 4). At 9 h post exposure, the number of nematodes touching the root declined to the level at 1 h post exposure. The three Arabidopsis ecotypes tested showed a similar pattern of attractiveness to SCN J2. This result indicates that the roots of Arabidopsis also secrete chemicals that attract H. glycines. Based on these results, the 2-h time point was chosen to evaluate attraction in subsequent experiments.
Similar to our results with soybean, AVG-treated Arabidopsis roots showed a significantly higher attractiveness to SCN J2 than the control roots, and no differences were found in the attractiveness of ETH-treated roots and non-treated roots (Fig. 5a,b). As another test of the effect of ET action on Arabidopsis root attractiveness, we treated seedlings with AgNO 3 (Ag + ), which inhibits ET perception by inactivating ET receptors 52,53 . Ag + -treated roots were observed to attract more nematodes than control roots ( Fig. 5a,b).

Attractiveness of Arabidopsis ET mutants to Heterodera glycines. Because chemical modulation
of ET in Arabidopsis and in soybean affects SCN attraction in a similar way, we tested Arabidopsis mutants in ET biosynthesis, perception, and downstream signal transduction. Root tips of Arabidopsis mutants eto1-2 and eto3, which produce more ethylene than WT 44,54 , did not show a significant difference in the number of nematodes touching the roots compared to WT (Fig. 6a,b). However, it appeared that fewer nematodes were found near the roots of eto1-2 and eto3 than of WT (Fig. 6a). To test this further, we counted J2 within a circle of 5 mm diameter centered at 2.5 mm above the root tip and found that, indeed, the numbers of J2 near the eto1-2 and eto3 root tips were lower than that in WT (Fig. 6c). This may suggest that a signal produced at a different level in the mutant plants is perceived at a distance from the roots. The phenotypes of shorter roots and more and longer root hair characteristic of eto1-2 and eto3 are known to be restored to the wild-type phenotype by treatment with Ag + 55 and we found that Ag-treated eto1-2 and eto3 mutants become more attractive to J2 than non-treated mutant or even wild type plants (Fig. 6a,b). This supports a role of ET-receptor mediated signaling in reducing attraction of the root tip to J2.
ET sensing in Arabidopsis occurs through a family of ET receptor proteins, and dominant gain-of-function mutations in the encoding genes that confer ethylene insensitivity are available 40,41,43 . We tested two gain-of-function mutants for effect on SCN J2 attraction, etr1-3 did not differ in attractiveness from WT, while ein4-1 was more attractive. Furthermore, loss-of-function ET receptor mutants etr1-7 and ers1-3, which have enhanced ET sensitivity, did not exhibit obvious differences in attractiveness from WT (Fig. 7a,b).
CTR1 is a negative regulator of the ET response, and the loss-of-function mutant ctr1-1 has the constitutive triple-response phenotype 44 . Interestingly, ctr1-1 attracted significantly fewer J2 than WT (Fig. 8), and the number of nematodes touching roots or in the vicinity of the root tip was significantly lower (P < 0.05) than that in WT (Fig. 8b,c). Further, ET-insensitive mutants ein2, ein2-1, ein2-5, ein3-1, ein5-1, and ein6, which are defective in positive regulation of ET signaling, showed significantly higher attractiveness compared with wild-type plants (Fig. 9a,b). Heterodera glycines infection activates ET signaling. Previous studies indicated that infective J2 of H. glycines could penetrate and migrate inside the Arabidopsis roots in the same manner as H. schachtii 56 . To assess whether ET signaling is activated in Arabidopsis early in the response to H. glycines infection, we examined the expression levels in transgenic Arabidopsis carrying the ET reporter construct, EBS::GUS, in which the β -glucuronidase (GUS) reporter gene is driven by a synthetic EIN3-responsive promoter 57 . No visible expression of EBS::GUS was observed in uninfected roots at tested time points (Fig. 10a,c,e and g). At 6 h post inoculation with H. glycines, we observed that J2 had penetrated and migrated inside the root and that there was a marked increase in GUS expression in the majority of root infection sites (Fig. 10b). Strong GUS expression was also observed in the infection sites and surrounding cells at 12 and 24 h after infection (Fig. 10d,f). However, the nematode-induced increase in GUS expression in the infection sites was reduced at 48 h post infection (Fig. 10h). We also noted a dramatic increase in the number and length of root hairs for infected EBS::GUS seedlings compared to uninfected roots (Fig. 10). ET plays a vital role in root hair formation and elongation 58 , and this increase further supports activation of ET signaling early during the infection process.

Discussion
In this study, the increased attractiveness of SCN to the roots of the host soybean and non-host Arabidopsis in the presence of the ET biosynthesis inhibitor AVG indicated that ET levels or signaling modulated the root attractiveness to SCN. We explored available Arabidopsis mutants in ET synthesis and signaling to gain additional information on whether ET levels or signaling were responsible. Arabidopsis mutants that had reduced ET signaling (ein4-1, ein2, ein2-1, ein2-5, ein3-1, ein5, ein6) were more attractive compared to wild type, indicating that reduced ET signaling in the plant played a positive role in the attraction of SCN to roots. Of the two GOF ET-receptor mutants tested, ein4-1 was significantly more attractive to J2, but etr1-1 was not. The lack of phenotype for etr1-1 is consistent with observations for other ET response phenotypes where it has been attributed to the redundant functions of ET receptors 59 . For example, previous studies reported that an etr1, etr2, and ein4 triple loss-of-function mutant was required for dramatic ET response phenotypes 60 . In addition, the significant attractiveness of plant roots treated with ET-perception inhibitor AgNO 3 suggests that ET perception by receptors negatively affects the attractiveness of Arabidopsis to SCN J2. Taken together, our results indicate that reducing ethylene pathway signaling increases attractiveness of SCN nematode to both host and non-host roots.   The number of nematodes touching the root of the ET-overproducing mutants eto1-2 and eto3 was similar to that of WT suggesting that increased ET production of these mutants did not reduce attraction. However, fewer J2 were found in the areas around the root tips of eto1-2 and eto3 than of WT, suggesting that increased ET levels or ET signaling in the mutant plants is perceived at a distance from the roots of the non-host plant Arabidopsis. However, the constitutive ET-response mutant ctr1-1 also showed less attraction than WT indicating that a strong, constitutive signaling of ethylene response results in lower attraction. Together these results indicate that strong, constitutive signaling of the ET response results in reduced attraction, but the ET-overproduction of mutants eto1-2 and eto3 is not sufficient to be detected in our touching assay, although an effect can be observed at a distance from the root. A possible explanation is that a repellent is produced that diffuses from the root.
Our findings are similar to those obtained using similar assays for the northern root-knot nematode M. hapla 35 . The roots of AVG-treated Arabidopsis and ET-insensitive mutants displayed increased attraction for M. hapla. However, for M. hapla, ET-overproducing mutants attracted fewer M. hapla, and the roots of GOF-ET receptor mutants etr1-3 and ein4-1 both significantly increased the attraction of M. hapla compared to WT while our results for SCN found that only ein4-1 roots were significantly more attractive to SCN J2 than those of the WT. Although etr1-3 and ein4-1 identified in genetic screens were all dominant and exhibited partial ET insensitivity 59,61 , there was a difference in their insensitivity to ET. Therefore, different attraction between SCN and RKN suggested that RKN are more sensitive than SCN to some consequences of host ethylene signaling.
In contrast to our findings, previous reports found that the roots of ET-overproducing mutant eto3 32 and ETH-treated Arabidopsis plants 36   H. glycines, are closely related and exhibit the same parasitic behaviors in penetration, migration, and development throughout their life cycle 62 , but differ in host range. There are several possible reasons for the differences regarding the influence of ET signaling between the previous studies and our own work. For example, the designs of the attraction assays differ substantially; we recorded attraction to living root tips after 2 h, whereas Wubben et al. 32 , recorded attraction to root-free exudate plugs after 20 h. Thus, our assay reflects a rapid nematode response to an intact root tip, and the other assay may reflect nematode accumulation in response to a host secreted metabolite rather than attraction. It is also possible that these two cyst nematode species recognize or respond to different compounds in the root exudates.
Cyst nematodes cause considerably more wounding during the infection process than do RKN, and the ET response that we see using the ET reporter system may be due to this wounding. Strong GUS expression was observed in the infection sites and its surrounding cells up to 24 h after infection indicating an increase in ET signaling at early stages of the Arabidopsis -H. glycines interaction. In addition, ET is a positive regulator of root hair development, and ET-overproducing mutants 44,54 and ETH-treated Arabidopsis plants exhibit the increased number and length of root hairs (Fig. 5a). We also observed root hair proliferation in EBS::GUS transgenic seedlings when exposed to SCN infection, further suggesting that ET signaling pathway might be involved in the response of Arabidopsis to early SCN infection. In addition, significant increase in the concentration of the ET precursor ACC concentration was found in SCN-infected soybean roots 39,63 . These studies suggested that ET pathway plays important roles in host or non-host plant at the early infection and later parasitic stages.

Materials and Methods
Nematode culture. A culture of H. glycines was initiated with a single cyst isolated from a soybean field in Taian, Shandong Province, China. The single cyst was used to infect the susceptible commercial soybean cultivar Dongsheng1. After 35-40 days, a single cyst collected from the roots was used to initiate the second generation; this process was repeated for 10 generations to produce the strain used for this study. The strain was determined to be SCN race 5 using the SCN differential host test 64 in a greenhouse in the Northeast Institute of Geography Plant materials and culture. For nematode attraction assays, seeds of soybean cv. DongSheng1 were germinated in pots containing vermiculite in a greenhouse at 22-28 °C with a photoperiod of 16/8 h light/dark cycle, and watered every 3 days with Hoagland's nutrient solution. The 12-day-old seedling roots were washed with water and lateral roots with intact root tips were used for attraction assays since primary roots were too big to be observed under the microscope.
Nematode attraction. Nematode attraction assays were conducted as previously described 35,68 . For soybean assays, 2 ml of 23% (w/v) Pluronic F-127 (NF Prill Poloxamer 407, BASF, Mt Olive, NJ, USA) containing 200 J2s was added into each well of 12-well tissue culture plates at 4 °C; then a 1-cm root piece containing an intact tip was placed into each well. For attraction assays with Arabidopsis seedlings, 1 ml 23% PF-127 containing 300 J2 and one root was added into each well of a 12-well tissue culture plate. The plates were transferred to room temperature to allow the gel to solidify.
To optimize the attraction assay, the attractiveness of roots to H. glycines was observed microscopically, and the number of J2 touching the root surface up to 5 mm from the root tip was counted at 1 h, 2 h, 3 h, 4 h, and 6 h for soybean or at 1 h, 2 h, 3 h, 4 h, 6 h, and 9 h for Arabidopsis. For Arabidopsis mutants ctr1-1, eto1-2, and eto3, the nematodes within the circle area of 5 mm diameter centered at 2.5 mm above the root tips were also counted. Roots and nematodes were photographed with an OLYMPUS SZX-16 dissecting microscope using Cellsens Standard image software (Olympus Corporation, Japan). The experiment was repeated three times with 24 replicates each time.
Chemical treatments. To evaluate whether ET plays a role in H. glycines attraction, 12-day-old soybean plants and 7-day-old Arabidopsis plants were treated with the ET analog ethephon (ETH, Sigma-Aldrich), AVG (Sigma-Aldrich) and/or Ag + (Sigma-Aldrich) before starting the attraction assay. The seedling roots were dipped into solutions of 50 μ M AVG, and 200 μ M ETH and/or 10 μ M AgNO 3 . After 24 h, seedling roots of soybean or Arabidopsis were washed three times with sterilized water and used for the attraction assays. At 6 h after the attraction assay, ETH and AVG-treated roots were stained with acid fuchsin 69 and the number of nematodes inside the root was counted. All experiments were repeated at least three times.
Histochemical analyses. Histochemical staining for GUS activity was performed as described by Jefferson et al. 70 with minor modifications. Seeds of EB3::GUS lines were sown on modified Knop medium (pH 6.1) containing 2% (w/v) sucrose, 0.8% (w/v) Daishin agar (Research Products International Corp., USA) 71 . For inoculation, hatching J2s were sterilized in 0.01% mercuric chloride and 0.002% sodium azide for 10 min and immediately washed three times in sterilized water; 50 surface-sterilized J2s were then applied on each twelve-day-old EB3::GUS plant. Seedlings were collected at 6, 12, 24 and 48 h after inoculation and then incubated in GUS-staining buffer containing 1 mM X-Gluc, 100 mM sodium phosphate (pH 7.5), 0.5 mM potassium ferrocyanide, 10 mM EDTA and 0.1% Triton X-100 for 12 h at 37 °C in the dark. Individual seedlings were mounted in 50% glycerol on microscopic glass slides and photographed under an Olympus compound microscope using Cellsens Standard image software (Olympus Corporation, Japan). At least 25 replicate plants were analyzed and the experiments were repeated three times.
Analysis of root hair phenotyping. The number of root hairs was counted in the 500-μ m region starting at the primary root tip. The length of root hairs was analyzed by NIH Image software (ImageJ, version 1.43). The average length of root hairs was determined upon measuring 40 hairs for each root. The experiments were repeated at least three times.
Quantitative reverse transcription PCR (qRT-PCR) analysis. After grinding soybean roots to a fine powder in liquid nitrogen, total RNA was isolated with Trizol (Invitrogen, Carlsbad, CA, USA). RNA samples were digested using RNase-free DNase I (Invitrogen) to eliminate any contaminating genomic DNA. For qRT-PCR analysis, first-strand cDNA was synthesized with 2 μ g of total RNA using PrimeScript RT reagent Kit (Thermo Fisher, USA). PCR reactions were performed in the LightCycler ® 480 System with FastStart Universal SYBR Green Master (ROX) (Roche) according to the procedure described by the manufacturer. The gene-specific primers were as follows: forward 5′ -GGGAAGGGGATGCACACAACCAAGG-3′ and reverse 5′ -GTTGGCCAT TCCATCCTTCCACCACCT-3′ for GmERF5; forward 5′ -CAGATTGAGCTTCAGCATTT-3′ and reverse 5′ -AAGTGTCATGCTTTGAGGAA-3′ for GmERS2; forward 5′ -GTGTAATGTTGGATGTGTTCCC-3′ and reverse 5′ -ACACAATTGAGTTCAACACAAACCG-3′ for GmUBQ3. All PCR cycles began with 10 min at 95 °C, Scientific RepoRts | 7:41282 | DOI: 10.1038/srep41282 followed by 40 two-step cycles comprising 10 s at 95 °C and 1 min at 60 °C. The relative expression of specific genes was calculated by the 2 −∆∆Ct method using GmUBQ3 as a reference. All experiments were conducted with three independent biological replicates and three technical repetitions.

Statistical analysis.
Data were subjected to one-way analysis of variance (one-way ANOVA) using SPSS software (SAS Institute, Cary, NC, USA). Results are reported as significant or non-significant in Tukey's t Test (Tukey HSD) Test (P < 0.05).