Overexpression of a plasma membrane protein generated broad‐spectrum immunity in soybean

Summary Plants fight‐off pathogens and pests by manifesting an array of defence responses using their innate immunity mechanisms. Here we report the identification of a novel soybean gene encoding a plasma membrane protein, transcription of which is suppressed following infection with the fungal pathogen, Fusarium virguliforme. Overexpression of the protein led to enhanced resistance against not only against F. virguliforme, but also against spider mites (Tetranychus urticae, Koch), soybean aphids (Aphis glycines, Matsumura) and soybean cyst nematode (Heterodera glycines). We, therefore, name this protein as Glycine max disease resistance 1 (GmDR1; Glyma.10g094800). The homologues of GmDR1 have been detected only in legumes, cocoa, jute and cotton. The deduced GmDR1 protein contains 73 amino acids. GmDR1 is predicted to contain an ecto‐ and two transmembrane domains. Transient expression of the green fluorescent protein fused GmDR1 protein in soybean leaves showed that it is a plasma membrane protein. We investigated if chitin, a pathogen‐associated molecular pattern (PAMP), common to all pathogen and pests considered in this study, can significantly enhance defence pathways among the GmDR1‐overexpressed transgenic soybean lines. Chitin induces marker genes of the salicylic‐ and jasmonic acid‐mediated defence pathways, but suppresses the defence pathway regulated by ethylene. Chitin induced SA‐ and JA‐regulated defence pathways may be one of the mechanisms involved in generating broad‐spectrum resistance among the GmDR1‐overexpressed transgenic soybean lines against two serious pathogens and two pests including spider mites, against which no known resistance genes have been identified in soybean and among the most other crop species.


Introduction
Food supply is often interrupted severely by devastating plant disease and pest epidemics. One of such major plant disease outbreaks is the Irish famine of 1845-1852 caused by late blight disease in potatoes, in which one million people died from starvation (Griffith, 2007). Plant breeders constantly breed disease and pest-resistant crop varieties to secure food supply. Plant disease resistance mechanisms are highly complex (Andersen et al., 2018;Dodds and Rathjen, 2010). Pattern-triggered immunity (PTI) activated by pathogen-associated molecular patterns (PAMPs), herbivore-associated molecular patterns (HAMPs), nematode-produced ascarosides or unknown molecular patterns is the first layer of plant defences, which is overcome by pathogen effector proteins to cause effector-triggered susceptibility (ETS) (Jones and Dangl, 2006). Plants then have evolved with receptors that recognize some of these effectors and trigger a strong form of disease resistance, named effector-triggered immunity (ETI) (Jones and Dangl, 2006;Zipfel, 2014). Receptors that recognize pathogen effector proteins to induce ETI often contain nucleotide-binding leucine-rich repeat (NB-LRR) domains (Macho and Zipfel, 2014). PTI and ETI defend plants from most pathogen and pest attacks by activating one or more signalling pathways regulated by plant hormones such as salicylic acid (SA), jasmonic acid (JA), abscisic acid (ABA) and ethylene (Bigeard et al., 2015;Kunkel and Brooks, 2002).
ETI has been extensively applied in breeding disease-resistant varieties in most crop species (Gu et al., 2005). It's usually effective only against a subset of a pathogen population and is not broad-spectrum. In contrast, PTI provides broad-spectrum and low level or partial resistance, not only against all isolates of a single pathogen, but also against multiple plant pathogens (Bigeard et al., 2015).
Plants possess numerous genes that encode putative surface receptors; for example, transmembrane receptor kinases (RKs) (Zipfel, 2014). Many of these genes may have been evolved to regulate plant defences. Unfortunately, the majority of these genes are yet to be studied. Characterized plant pattern recognition receptors (PRRs) involved in PTI are classified into: (i) receptor-like kinases (RLKs) and (ii) receptor-like proteins (RLPs) (Macho and Zipfel, 2014). RLPs do not contain a kinase domain as observed in RLKs for signalling. Interaction of an RLP and a kinase with an RLK for stem and floral meristem development has been demonstrated (Bleckmann et al., 2010).
Worldwide, soybean is an economically very important crop. In the United States, soybean suffers annual yield suppression valued over $5 billion from various pathogenic diseases (Allen et al., 2017). Sudden death syndrome (SDS) is one of the most serious soybean diseases, which is caused by the fungal pathogen Fusarium virguliforme. The pathogen infects and colonizes soybean roots causing necrosis and root rot, and subsequently foliar SDS, which is characterized initially by leaf chlorosis followed by necrosis, leaf and pod drops. The pathogen remains in roots and releases phytotoxins to cause foliar SDS (Brar and Bhattacharyya, 2012;Brar et al., 2011;Pudake et al., 2013). In the U.S., recently the yield suppressions from F. virguliforme have been reported to be second to that from the most serious pathogen, soybean cyst nematode (SCN). In recent years, the total annual soybean yield suppressions caused by the two pathogens have been valued at close to $2 billion (Allen et al., 2017).
In a transcriptomic study of the soybean-F. virguliforme interaction, we observed that the steady state transcript levels of only a few soybean genes were reduced by the F. virguliforme infection (Ngaki et al., 2016;Sahu et al., 2017). One of these genes, Glycine max disease resistance 1 (GmDR1; Glyma.10g094800) encodes a novel protein with unknown function (Ngaki et al., 2016). Overexpression of GmDR1 in transgenic soybean plants enhances immunity not only against F. virguliforme, but also against SCN, spider mites and soybean aphids. GmDR1 is an integral plasma membrane protein. It is predicted to contain an ecto-and two transmembrane domains with no kinase domain. The chitin, molecular pattern present in all four pathogen and pests included in this study, induces defence signalling pathways mediated by plant hormones, SA and JA, but suppressed the one mediated by ethylene among the GmDR1overexpressed soybean plants. We therefore hypothesize that GmDR1 could be a receptor-like protein. Following ectopic overexpression, it presumably recognizes pathogen and pestassociated molecular pattern(s) including chitin to initiate the broad-spectrum disease and pest resistance in soybean.

Results
Overexpression of GmDR1 enhanced F. virguliforme resistance Earlier we have shown that GmDR1 and a few other soybean genes are down-regulated following infection with F. virguliforme (Ngaki et al., 2016). We hypothesized that F. virguliforme suppresses the transcription of GmDR1 to cause susceptibility. To test this hypothesis, we fused GmDR1 to three infection-inducible promoters and created three fusion GmDR1 genes ( Figure S1; Tables S1 and S2). A total of 30 independent transformants from these three GmDR1 fusion genes were generated, and progenies of the transgenic soybean plants were evaluated for responses to F. virguliforme infection. It was observed that approximately 40% of the segregating R 1 progenies of transgenic lines were SDSresistant; whereas only 9% of the nontransgenic, GmDR1-fusion gene recipient Williams 82 plants showed SDS resistance (Figure 1a,b; Figure S2a). Furthermore, root rot was significantly reduced among the SDS F. virguliforme R 1 progenies (Figure 1c, d; Figure S2b). GmDR1 transgenes were expressed among the F. virguliforme-resistant R 1 progenies; but not among the SDS susceptible progenies (Figure 1e; Figure S2c). No amplification was detected in nontransgenic Williams 82 that did not carry the GmDR1 transgene (Figure 1e). qPCR of a pathogen and a host gene revealed at least 2.5-to 5-folds reduction in fungal biomasses in roots of the SDS-resistant transgenic soybean lines as compared to that in nontransgenic, SDS susceptible Williams 82 line (Figure 1f). The reduced fungal biomasses among the SDS-resistant transgenic lines was associated with the 2-fold reduction in root rot symptoms in SDS-resistant transgenic plants as compared to the nontransgenic Williams 82 plants (Figure 1d).
Transgenic plants carrying the GmDR1 transgenes were also evaluated for SDS resistance under field conditions. GmDR1 transgenes enhanced SDS resistance of the transgenic soybean plants during the 2015, 2016, 2017 and 2018 growing seasons. We observed that 65 to 91% of the basta-resistant transgenic R 1 plants descended from five independent transgenic R 0 soybean plants exhibited enhanced SDS resistance under field conditions in 2015 ( Figure 1g; Figure S3). The copy number of the R 1 plants was ascertained by conducting qPCR (Ngaki et al., 2016), and seeds of at least one homozygous progeny from individual transgenic lines were planted in the 2016 field trial. Up to 91% of the R 2 progenies showed SDS resistance with little or no visible foliar SDS symptoms (Figure 1g; Figure S3). Similar results were observed in the 2017 field trial conducted for the R 3 progenies of the homozygous R 2 lines ( Figure 1g). In 2018, the R 4 progenies of R 3 lines were tested against a very high load of the F. virguliforme. R 4 progenies exhibited significantly higher levels of SDS resistance as compared to that in the nontransgenic Williams 82 line ( Figure 1g). The responses of the transgenic lines to F. virguliforme infection were determined in four independent fields during the 4-year trial. There were no obvious morphological changes among the transgenic soybean lines. The mean seed size and yield/plant of the transgenic lines were statistically not different from that of the nontransgenic Williams 82 lines ( Figure S4). Our data suggest that overexpression of GmDR1 in roots of transgenic plants enhances resistance against F. virguliforme without affecting the yield potential.
Overexpression of GmDR1 resulted in induced expression of the GmPR1-1 gene, a marker of the defence pathway regulated by SA The GmDR1-transgenes were highly expressed among the roots of transgenic lines (Figure 2a). The overall expression levels of GmDR1-transgenes were~500-folds higher than that of the constitutively expressed soybean ELF1b gene. As observed before, the expression of the endogenous GmDR1 gene was suppressed among the nontransgenic and transgenic soybean plants following F. virguliforme infection (Figure 2b; Ngaki et al., 2016). The expression of GmDR1-homeologues, GmDR2 (Glyma.02g180500.1) and GmDR3 (Glyma.19G142700.1), was not influenced by F. virguliforme infection (Figure 2c,d). The GmDR4 (Glyma.03g139900.1) transcripts were not detected among the roots of the either nontransgenic or transgenic soybean plants (data not presented).
Towards understanding the possible mechanisms of enhanced SDS resistance among the transgenic soybean lines with overexpressed GmDR1, we investigated if any of the two major plant hormones, SA-and JA-mediated defence pathways, are altered among the transgenic plants. Transcript levels of the two GmPR1 homeologues, GmPR1-1 and GmPR1-2 (Xu et al., 2016;Xu et al., 2018;Zeng et al., 2017), marker genes for the defence pathway regulated by SA, were investigated. Surprisingly, GmPR1-1, but not GmPR1-2, is constitutively induced over 120-folds more among the transgenic soybean lines as compared to that in the nontransgenic Williams 82 control (Figure 2e,f). The enhanced GmPR1-1 transcript levels of the transgenic soybean lines were however not changed following F. virguliforme infection. Transcript levels of GmPR1 genes were also not induced in nontransgenic Williams 82 following inoculation with F. virguliforme infection (Figure 2e,f).
Two soybean homologues of the positive basal resistance regulator Arabidopsis EDS1, GmEDS1a (Glyma04g34800) and GmEDS1b (Glyma.06g19920), were also investigated for their expression patterns following F. virguliforme infection (Wiermer et al., 2005). There were no significant differences in the GmEDS1a transcript levels between the nontransgenic and transgenic soybean lines (data not shown). However, the expression of GmEDS1b was significantly suppressed among the transgenic lines as compared to the control nontransgenic soybean lines (Figure 2g). The expression of GmNPR1-1 but not GmNPR1-2 was significantly reduced in one of the GmDR1overexpressed lines (Figure 2h). The expression of GmNPR1-1 was also significantly reduced in Williams 82 following F. virguliforme infection (Sandhu et al., 2009;Figure 2h). Transcript levels of the JA pathway marker, GmJAR1, were unchanged ( Figure 2i).
Overexpression of GmDR1 resulted in novel immunity against the spider mites Two-spotted spider mites (Tetranychus urticae Koch) are leaffeeding pests that cause yellow and brown leaf spots, bronze colour in the entire leaf blade and finally develop mite-webs leading to severe yield losses (Jimenez, 2014). Unfortunately, no acceptable mite resistance (http://corn.agronomy.wisc.edu/Mana  Figure S2c. The primers (Table S5) used for RT-PCR were transgene-specific; and as expected, no amplification was detected in the nontransgenic Williams 82. (f) Relative biomass of F. virguliforme in infected root tissues was calculated as relative amplification of the fungal genomic DNA for the FvTox1 gene as compared to that for the soybean gene Glyma.05G014200 in quantitative PCR experiments. Two weeks following infection of roots with F. virguliforme or treatment with only water, root tissues were collected for quantitative PCR. gement/pdfs/A3890.pdf) has been reported in soybean and many other crop species (Agut et al., 2018). SA and JA regulate mite infestation of plants (Arena et al., 2018).
In February of 2014, spider mites infested severely all R 0 transgenic plants except the ones that carried the P2-DS1 transgene, grown in the greenhouse ( Figure S5a). Some of the R 1 progenies of the transgenic plants carrying Overexpression of GmDR1 resulted in enhanced immunity against soybean aphids The soybean aphid (Aphis glycines Matsumura) is a sap-sucking pest. It is a major yield-reducing pest of soybean (Ragsdale et al., 2007). It can damage soybean plants either by feeding on tissues or through transmitting pathogenic viruses (Clark and Perry, 2002). Resistance to aphids is encoded by quantitative (Glyma.03G256200) in soybean roots following SDS infection. Data represent mean AE standard error of three independent experiments. Each experiment contains three biological replicates of 6 pooled seedlings for each genotype in each treatment. Expression values were normalized to the expression levels of the constitutively expressed Elongation factor 1-b (ELF1-b) gene (Glyma.02g44460) in respective samples. *, significantly different to the control W82; *, significantly different to non-infected plants for the same genotype. One star, P < 0.05; two stars, P ≤ 0.001. P1, P2, and P3 are promoter 1, promoter 2, and promoter 3 (Table S1) trait loci (QTL) as well as single genes (Wiarda et al., 2012). Growing aphid-resistant cultivars is the most effective method of controlling this pest (Hesler et al., 2013). We investigated if overexpression of the GmDR1 transgenes enhanced aphid resistance among the transgenic soybean plants. The progenies of six independent transgenic soybean plants, generated from three independent GmDR1 transgenes, were investigated for responses to aphid infestation in clip-cage experiments (Myers and Gratton, 2006). The number of aphids on the leaves of transgenic lines was up to 5-fold less that on the leaves of nontransgenic Williams 82 plants (Figure 4a-c; Figure S7). The expression of soybean aphid resistance was associated with the expression of GmDR1-transgenes among the transgenic lines ( Figure 4d).
Overexpression of GmDR1 resulted in enhanced immunity against the soybean cyst nematode The soybean cyst nematode (SCN; Heterodera glycines) is a rootfeeding parasite. In the U.S., it is the most serious soybean pathogen that causes annual yield suppression valued at over $1 billion (Allen et al., 2017;Mitchum et al., 2012). SCN resistance genes are deployed worldwide in breeding SCN-resistant soybean cultivars to reduce the crop losses from this serious pathogen (Guo X et al., 2015;Liu et al., 2012). We investigated if the transgenic plants overexpressing the GmDR1 transgenes can provide any enhanced SCN resistance. We observed that several R 1 and R 2 progenies of independent transformants showed enhanced SCN resistance (Figure 5a). The female indices (FI) were significantly reduced among the transgenic lines as compared to the nontransgenic Williams 82 plants (Figure 5b; Figure S8). Transgenic lines are moderately resistant or moderately susceptible (FI ranged from 11 to 60) (Adee and Johnson, 2008), whereas the transgenes recipient nontransgenic Williams 82 line is highly SCN susceptible. In the SCN-resistant A95-684043 line, the FI was < 10. Enhanced SCN resistance among the transgenic lines was associated with the expression of the GmDR1 transgenes ( Figure 5c). The numbers of adult females were significantly lower in the roots of transgenic soybean plants as compared to that in roots of nontransgenic Williams 82 plants, although similar numbers of juveniles were observed among the transgenic and nontransgenic soybean lines (Figure 5d, e; Figure S9).
Promoters 2 and 3 fused GmDR1 transgenes are strongly expressed in leaves inducing up-regulation of the Aphidinducible 1 gene Promoter 2 and Promoter 3 used in generating the GmDR1 fusion genes for this study were isolated, respectively, from the Glyma.10g168900 and Glyma.20g220800 genes encoding germin-like proteins (GLPs) of the subfamily 1 member 10 that contain the cupin domain (Lanubile et al., 2015). In general, plant GLPs are differentially expressed during plant growth and development. They are responsive to biotic and abiotic stresses including bacteria, fungi, insects, nematodes, salinity, temperature, drought, nutrient, (Davidson et al., 2009;Dunwell et al., 2008;Gunadi et al., 2016;Lanubile et al., 2015;Lu et al., 2010;Ngaki et al., 2016;Wei et al., 1998). ELF-1b is the internal control. Significant differences observed between the transgenic and Williams 82 lines are shown with * for P < 0.05 and ** for P < 0.01. P1, P2 and P3 are promoter 1, promoter 2 and promoter 3, respectively (Table S1). In this study, we observed that Promoter 2 (P2) and Promoter 3 (P3) fused GmDR1 transgenes (P2-DS1 and P3-DS1) enhanced SDS and SCN resistance in roots, and spider mite and soybean aphid resistance in leaves of transgenic soybean plants (Figures 1,  3-5). The two promoters are root-specific (Gunadi et al., 2016). Promoter 3 has the highest expression level in hairy roots, three times more than that of the CaMV 35S promoter. Promoter 2 showed slightly lower activity than Promoter 3 (Hernandez-Garcia et al., 2010). The Promoter 3 is weakly active in leaves (Table S1).
We investigated the expression levels of the Glyma.10g168900 and Glyma.20g220800 genes in leaves and root ( Figure S10). The two promoters, fused to GmDR1, were active in leaves of transgenic soybean lines ( Figure S11). There are three additional GmDR1-like genes in the soybean genome (Table S2; Figure S12). Overexpression of GmDR1 transgenes ( Figure S13a) did not significantly influence the expression of endogenous GmDR1, GmDR2 or GmDR3 genes ( Figure S13b-d).
In a transcriptomic study of the soybean-soybean aphid interactions, it was reported that transcripts of only one gene (Glyma06g14090) was induced 7 days following aphid infestation (Studham and MacIntosh, 2013). We name the gene as Aphid-inducible 1 (GmAI1). Considering the induction of aphid and mite resistance among the transgenic soybean lines among the GmDR1-overexpressed transgenic plants, we investigated if the expression of the GmAI1 gene is influenced by the overexpressed GmDR1 transgenes in soybean leaves. The expression levels of GmAI1 were highly increased in both transgenic lines with overexpressed-GmDR1 ( Figure S13e).

GmDR1 an integral plasma membrane protein
GmDR1-like sequences were detected only in cotton, cocoa, jute and legumes. Phylogenetic analysis of the GmDR1 and its homoand homeologues revealed three clades (Table S3; Figure S12a) with GmDR1 and its three homeologues clustered in two subclades. Alignment of the closely related six GmDR1-like legume proteins including GmDR1 showed strong conservation of several amino acid residues, of which D35 and S36 residues could be involved in protease cleavage and phosphorylation, respectively (Figure 6a). The sizes of GmDR1 and its homologues range from 67 to 73 aa (Figure 6a).
Functional data indicate that GmDR1 could be a pattern recognition receptor (PRR) that recognizes a molecular pattern common to a variety of organisms including fungus, nematode and insects to induce broad-spectrum basal plant immunity in transgenic soybean plants (Figures 1-5). We therefore investigated if it is a plasma membrane bound protein. The 73 aa GmDR1 protein is predicted to have an N-terminal cytoplasmic domain (11 aa) followed by a transmembrane domain (23 aa), an ecto-domain (14 aa), a second transmembrane domain (23 aa), and a short cytoplasmic tail (2 aa) (Figure 6a; Figure S14a). The predicted 3D model for the GmDR1 protein revealed two helical regions that perfectly match the predicted transmembrane  Figure S14b) (Yang and Zhang, 2015).
To experimentally verify its possible plasma membrane residence, subcellular localization study was conducted in Glycine max and Nicotiana benthamiana. Transient expression of GmDR1 fusion proteins with the GFP tag at its either N-or C-terminus revealed that GmDR1 is localized to plasma membrane (Figure 6b; Figure S15). Investigation of genes co-expressed with GmDR1 revealed that many of the co-expressed genes are involved in cell wall biogenesis and are membrane bound (Table S4).
Chitin induces SA-and JA-mediated defence pathways among the GmDR1 transgenic lines Chitin, a well-known PAMP or MAMP, is present in F. virguliforme, SCN, aphids and spider mites (Bos et al., 2010;Chen and Peng, 2019;S anchez-Vallet et al., 2015;Zhou et al., 2017). Considering GmDR1's plasma membrane residence, we hypothesize that following either direct or indirect interaction with chitin, GmDR1 triggers broad-spectrum immunity mechanisms against all four pathogen and pests considered in this study. To test this hypothesis, stem-cuts of 2-week-old (i) transgenic plants carrying either GmDR1 or GUS transgene, and ii) transgene-recipient Williams 82 plants were treated with chitin (Khan et al., 2003). qRT-PCR was conducted to monitor the expression of eight defence genes representing markers of the SA-, JA-and ethylene-mediated defence signalling pathways. The genes considered were two GmPR1 genes, GmPR2, GmEDS1, stress-induced NAC transcription factor 6 (GmNAC6) gene (Glyma.12G022700) and the isochorismate synthase gene GmICS1 (Glyma01g25690) as markers for the SA pathway, GmJAR1 as the JA pathway marker and the aminocyclopropane-1-carboxylate synthase encoded by GmACS1k as a marker for the ethylene pathway (Glyma.16G032200) (Garcion et al., 2008;Lin et al., 2013;Melo et al., 2018;Pimenta et al., 2016;Xu et al., 2018). Chitin significantly induced the expression of the two GmDR1 transgenes but not the endogenous GmDR1 and GmDR3 genes (Figure 7a,b,d). The expression of GmDR4 was not detectible either in the control or in the chitin treated plants (data not presented). The expression of the endogenous GmDR2 gene was however significantly reduced among the GmDR1 transgenic lines 12 h following chitin treatment (Figure 7c). Expression of the SA and JA pathway markers were induced at least in one of the two transgenic lines carrying GmDR1 transgenes (Figure 7e-k), whereas the expression of the marker for the ethylene pathway was suppressed in one of the two transgenic lines (Figure 7l). No effect of the chitin treatment was observed in the expression of the selected marker genes among the transgenic line carrying GUS gene and transgene-recipient, nontransgenic Williams 82 line.

Discussion
Four classes of soybean genes including GmDR1 are downregulated following F. virguliforme infection (Ngaki et al., 2016). F. virguliforme presumably manipulates the expression of a few putative defence-related soybean genes to cause SDS. Exchange of promoters of GmDR1 and one member each from two other classes of soybean genes with infection-inducible and strong rootspecific promoters enhanced resistance of transgenic soybean lines to the fungal pathogen F. virguliforme (Figure 1; Ngaki et al., 2016;M. Ngaki and M.K. Bhattacharyya, unpublished). These results suggest that F. virguliforme somehow down-regulates expression of a few soybean genes to suppress the active defence mechanisms. It could be possible that F. virguliforme pathogenicity factors directly bind promoters of defence genes to induce susceptibility. Binding of plant promoters by pathogen effector proteins has been shown to induce susceptibility. It has been concluded that the CRN effector PsCRN108 of the soybean pathogen Phytophthora sojae containing a putative DNA-binding helix-hairpin-helix (HhH) motif could be involved in suppressing the expression of plant defence-related genes by directly targeting specific plant promoters (Song et al., 2015).
We have identified three homeologues of GmDR1; viz., GmDR2, GmDR3, and GmDR4 ( Figure S12). The qRT-PCR results showed that the suppression of transcripts following F. virguiforme infection was statistically significant only for the endogenous GmDR1 gene, but not for the GmDR1 homologues (Figure 2b-d). Overexpression of the GmDR1 gene through swapping its promoter with those of the two root-specific and F. virguliforme infectioninducible genes led to enhanced resistance of transgenic soybean plants against not only the fungal pathogen F. virguliforme, but also against a nematode pathogen, SCN, and two pests, spider mites and soybean aphids, all of which are major deterrents of soybean production (Figures 1, 3-5; Tables S1-2; Allen et al., 2017;Ngaki et al., 2016;Sahu et al., 2017;Brandenburg and Kennedy, 1987;Costamagna et al., 2007).
Williams 82 is an SDS susceptible soybean cultivar. We failed to detect expression of PR1 genes in roots of Williams 82 following infection with F. virguliforme (Figure 2e,f). Earlier, we had failed to observe expression of GmPR1 gene 3-and 5-day following inoculation of etiolated Williams 82 seedlings with F. virguliforme (Ngaki et al., 2016). Induction was observed only after 10 days following inoculation. In soybean, transcript levels of GmPR1-like and GmPR genes may either decrease or increase in the susceptible soybean lines following infection (Abdelsamad et al., 2019;Kim et al., 2011).
The transcript levels of GmEDS1b were down-regulated in both GmDR1-overexpressed plants (Figure 2g). EDS1 is a positive regulator of basal resistance in Arabidopsis. Pathogen effectors alters its interaction with receptors that regulate immunity (Bhattacharjee et al., 2011). In soybean, EDS1 homologues function however differently as compared to that by Arabidopsis EDS1. Interaction of GmEDS1a/GmEDS1b proteins with the cognate bacterial effector protein is required for virulence function of a bacterial pathogen in soybean (Wang et al., 2014). Thus, EDS1b expression may have a similar role in pathogenicity function of F. virguliforme, and down-regulation of EDS1b in GmDR1-overexpressed plants therefore may contribute towards enhancing SDS resistance.
Like the expression of GmEDS1b, the expression of GmNPR1-1 was also significantly reduced in one of the GmDR1-overexpressed lines (Figure 2h). The expression of GmNPR1-1 was significantly reduced in Williams 82 following F. virguliforme infection. It appears that as in GmEDS1, GmNPR1 may also act differently in soybean as compared to that by NPR1 in Arabidopsis. In wheat, constitutive expression of the positive regulator of immunity Arabidopsis NPR1 resulted in increased susceptibility to Fusarium asiaticum that causes fusarium seedling and head blights (Gao et al., 2013). It is becoming apparent that the knowledge gained in the model plant Arabidopsis may not always be translated to all crop species. The resources created in this study should be useful in dissecting immunity signalling pathways in soybean. Our results suggest that SA-mediated defence signalling pathway could be one of the mechanisms used by GmDR1 in enhancing immunity of transgenic soybean plants against F. virguliforme.
Transcript levels of the constitutively expressed soybean ELF1b gene were used to standardize the expression levels of all genes in our quantitative RT-PCR experiments. Therefore, the results of two experiments can be comparable. In our study, we observed that the expression levels of Glyma.10g168900 and Glyma.20g220800 genes containing Promoter 2 (P2) and Promoter 3 (P3), respectively, in roots around 10-fold of the transcript levels of the ELF1b gene ( Figure S10). Surprisingly, the transcript levels of the P2-GmDR1 and P3-GmDR1 transgenes were over 500-fold higher than that of the ELF1b gene (Figure 2a; Figure S10). Thus, a 50-fold increases in activities of the two promoters were observed among the transgenic lines carrying the two transgenes ( Figure 2a). We evaluated 15 independent transformants for P2-GmDR1, and eight for P3-GmDR1 transgene. From these transgenic lines, we identified the most SDS-resistant transgenic lines for molecular analyses. Enhanced activities of Promoters 2 and 3, measured by GmDR1 transcript level, among the transgenic lines  Reuter and Spierer, 1992). The enhancer and silencer elements as well methylation activities of the T-DNA insertion sites can influence the expression of transgenes. Position effect-induced variegation is often observed in transgenic studies (Bhattacharyya et al., 1994;Wakimoto, 1998;Williams et al., 2008). Study of a large number of transgenic events allowed us to identify the desirable transgenic Figure 7 Regulation of defence-related genes 12 h following treatment with chitin. Relative transcript abundance of (a) GmDR1 transgenes, (b) endogenous GmDR1 gene, (c) GmDR2 (Glyma.02g180500.1), (d) GmDR3 (Glyma.03g139900.1), (e) GmPR1-1 (Glyma.15g062500), (f) GmPR1-2 (Glyma.13G251600), (g) GmPR2 (Glyma.03G132700), (h) GmEDS1b (Glyma.06g19920), (i) GmNAC6 (Glyma.12G022700), (j) GmICS1 (Glyma01g25690), (k) GmJAR1 (Glyma.19G254000), (l) GmACS1k (Glyma.16G032200) 12 h following chitin treatment. Data are mean expression values normalized to the expression levels of the soybean Elongation factor 1-b (ELF1-b; Glyma.02g44460) in respective samples and AE standard errors calculated from three independent experiments. In each experiment, three pools of six seedlings for each genotype was considered. *, significantly different to the control W82; *, significantly different due to chitin treatment in a genotype. * or *, P < 0.05; ** or **, P ≤ 0.001. P2, and P3 are promoter 2 and promoter 3 (Table S1), respectively. W82, Williams 82; GUS, a transgenic line harbouring the GUS transgene. plants with enhanced SDS resistance resulting from very strong overexpression of GmDR1.
Overexpression of GmDR1 however did not result any noticeable undesirable effect on the soybean plants. Both plant height and seed yield were not affected among the transgenic plants ( Figure S4). Overexpression of GmDR1 led to constitutive induction of a novel aphid resistance-related novel gene GmAI1 in addition to defence mechanisms including SA-mediated defence pathway (Figure 2; Figure S13e; Studham and MacIntosh, 2013). A transcriptomic study of the transgenic lines is warranted to gain a comprehensive understanding of the defence pathways constitutively induced by overexpressed-GmDR1 among the transgenic lines.
We have demonstrated that GmDR1 is an integral plasma membrane protein ( Figure 6). GmDR1 with 73 amino acid residues contains an N-terminal cytoplasmic domain (11 aa), two predicted transmembrane domains (23 aa), an ecto-domain (14 aa), and a short cytoplasmic tail (2 aa) (Figure 6a; Figure S14b). GmDR1 is a member of the Panther family PTHR33659:SF7 that contains twenty-five uncharacterized genes including four from soybean (GmDR1, À2, À3, À4). Plasma membrane residence and priming of multiple defence mechanisms to enhance broad-spectrum pathogen and pest resistance indicate that most likely GmDR1 is a receptor that recognizes pathogen and pest-associated molecular pattern(s). One pathogen-associated molecular pattern (PAMP) common to all four organisms, two pathogens and two pests, is chitin (Bos et al., 2010;Chen and Peng, 2019;S anchez-Vallet et al., 2015;Zhou et al., 2017). Chitin application to intact soybean plants significantly enhanced the accumulation transcripts of marker genes of the SA and JA-regulated defence pathways, whereas suppressed the transcription of a marker gene of the defence pathway mediated by ethylene (Figure 7). We failed to see induction in any of the SA or JA marker genes in response to chitin treatment among the transgenic line carrying the GUS gene or nontransgenic Williams 82 line. Lack of endogenous GmDR1 genemediated induction of SA and JA markers among these lines could be attributed to very low expression levels of this receptor protein gene (Figure 7a). Because of overexpression, the GmDR1 transgenes were able to show the chitin responses as well as broad-spectrum disease and pest resistance among the transgenic lines (Figure 7a).
We, therefore, hypothesize that one of the genetic mechanisms involved in generating broad-spectrum resistance among the transgenic soybean lines with overexpressed GmDR1 transgenes could be through activation of defence pathways mediated by chitin, a well-recognized PAMP (Wan et al., 2008b).
In our study, we infected roots with F. virguliforme and treated leaves with chitin. We observed variations in the expression patterns of genes including GmDR1, GmEDS1b, GmJAR1 and GmPR1s following infection with F. virguliforme and treatment with chitin. These differences may be arisen because of variation in organ types studied and complexity of F. virguliforme infection as compared to the chitin treatment (Figures 2,7). Pathogenicity factors of F. virguliforme are involved in developing SDS and interfere the host defence mechanisms. These results may also indicate that GmDR1-induced broad-spectrum resistance is governed by multiple genetic mechanisms including the ones induced by chitin.
Transgenic studies showed that chitin induced defence responses are specific to constitutively overexpressed GmDR1 and are absent in nontransgenic Williams 82. We, therefore, hypothesize that one of the mechanisms involved in generating broad-spectrum resistance among the transgenic lines could be through activation of multiple defence pathways mediated by chitin, a well-recognized PAMP, through a possible interaction with the overexpressed GmDR1 protein (Wan et al., 2008b). GmDR1, therefore, could be a PAMP recognition receptor. It does not seem to contain a typical LysM domain for binding chitin and peptidoglycans found in LysM-containing receptor-like kinases (Buist et al., 2008;Petutschnig et a., 2010;Tanaka et al., 2013). Further studies warranted to confirm the possible interaction of GmDR1 with chitin and establish its PRR role in generating broad-spectrum disease and pest resistance in soybean.
Natural spider mite resistance has not yet been identified in most crop plants including soybean (http://corn.agronomy. wisc.edu/Management/pdfs/A3890.pdf; Agut et al., 2018). Biological and chemical controls are the major methods of managing this serous pest (Agut et al., 2018). The generated spider mite resistance through overexpression GmDR1 is novel and has the potentiality in breeding spider mite-resistant legumes, cotton, cocoa and jute, in which GmDR1 homologues are detected ( Figure S12). It will also be important to investigate the potentiality of GmDR1 in creating novel spider mite resistance in other crop species that do not carry any GmDR1 homologs.

Binary vector constructions and generation of soybean transgenic lines
GmDR1 was fused to three promoters as follows. The pTF102 binary vector (Frame et al., 2002) was modified by replacing the CaMV 35S promoter with any one of the three selected promoters: promoter (prom) 1 (Glyma18g47390), prom 2 (Gly-ma10g31210) and prom 3 (Glyma20g36300) ( Figure S1a; Table S1; Ngaki et al., 2016). Primers used for amplifying and cloning these new promoters are listed in Table S5. The GUS gene in the modified pTF102 vector was replaced with the genomic sequence of the GmDR1 gene ( Figure S1a). The created plasmids therefore contained the GmDR1 fusion genes generated by fusing GmDR1 to any of the above three promoters. The CaMV 35S polyA signal was fused at the 3'-end of GmDR1 for polyadenylation of the transgene transcripts. The GmDR1-Fw and GmDR1-Rev primers were used to amplify GmDR1. The created binary plasmid vectors were introduced into Agrobacterium tumefaciens strain EH101 through electroporation.
Seven independent transformants for P1-DS1, 15 for P2-DS1 and eight for P3-DS1 transgenes were generated by conducting Agrobacterium-mediated transformation of the Williams 82 soybean cultivar at the Plant Transformation Facility, Iowa State University (Paz et al., 2004). Transgenic R 0 plants resistant to the glufosinate-ammonium (Liberty 280 SL, Bayer CropScience, Research Triangle Park, NC, USA) were grown in individual pots containing the standard soil fertilized with Osmocote (Scotts, Marysville, OH, USA) in a greenhouse under 16 h light and 8 h dark photoperiod. To confirm the transgene insertion, genomic DNA was extracted from young leaves of R 0 plants (Ngaki et al., 2016). PCR analysis was conducted to determine the integration of GmDR1 and bar (bialaphos resistance) genes into the soybean genome ( Figure S1b). The R 1 seeds from individual R 0 plants carrying the GmDR1 and bar genes were harvested.

F. virguliforme infection assay in growth chambers
F. virguliforme Mont-1 was grown on 1/3 potato dextrose agar (PDA) plates for 3 weeks. The inoculum was prepared by growing the pathogen in sorghum meals (Luckew et al., 2012). The inoculum was well mixed at a concentration of 1: 20: inoculum: sand and soil mixture in equal proportion and placed in 237-ml Styrofoam cups. Three independent experiments were carried out in the growth chambers maintained at 22.5°C, 16 h light with 350 µE/m2/s intensity and 8 h dark photoperiod. The SDS susceptible Williams 82 and the SDS-resistant MN1606 were grown along with the transgenic lines as controls. Plants were watered daily and SDS symptoms were scored as follows. Foliar and root rot symptoms were evaluated 4 weeks after planting according to the published methods (Hartman et al., 1997;Huang and Hartman, 1998;Li et al., 2009). The plants showing a score of 1 (no symptoms) or 2 (slight yellowing) were considered SDS-resistant plants, whereas plants with scores 3 to 7 (browning, interveinal chlorosis and necrosis) were classified as the susceptible plants (Table S6). Thirty-seven days after planting (DAP), plants were carefully removed from the cups and the roots were washed in warm tap water. The root showing dark brown to black discoloration (Roy et al., 1997) was visually assessed in a percentage scale from 0 to 100% of the total root area with an increment of 5%; 0% means healthy roots with light brown colour to 100% means rotten roots with dark black colour. Root tissues of resistant and susceptible progeny plants of each transgenic line were collected and immediately frozen in liquid nitrogen for molecular analysis.

Field trials for responses of transgenic soybean lines to F. virguliforme
In the summer of 2015 (June 11 to October 30, 2015), seeds of the R 1 progenies of independent transformants, Williams 82 and MN1606 were hand-planted along with F. virguliforme inoculum to evaluate their responses to the pathogen in a completely randomized block design with two replications. The experiment was conducted at the Hinds Research Farm, Iowa State University located in north of Ames, Iowa. The inoculum of F. virguliforme isolate NE305S was prepared the same way as for Mont-1 used in the growth chamber experiments. For each transgenic event, twenty-five seeds were mixed with 10 ml dry F. virguliforme inoculum and planted using a push planter. To eliminate the segregants with no transgenes, transgenic plants were sprayed with the Liberty 280L solution (glufosinate-ammonium at a final concentration of 250 mg/L and 0.1% Tween 20) 2 weeks following germination (Ngaki et al., 2016). The spray was repeated once more after two days of the first application. Twelve herbicide-resistant plants were randomly selected from each transgenic event for collecting young leaves to prepare DNA for transgene copy number study as reported earlier (Ngaki et al., 2016). SDS symptoms appeared in August and were scored as described for growth chamber experiments; and plants were classified into resistant with scores 0 to 2 and susceptible with scores 3 to 7 (Table S6).
In the summer of 2016 (June 7 to October 30, 2016), we conducted the field trial for R 2 transgenic plants. The method was similar to the 2015 field experiment, but this time we planted R 2 seeds of selected putative homozygous R 1 plants carrying a single transgene copy (Ngaki et al., 2016) in four blocks. We sprayed Liberty herbicide to eliminate any possible heterogeneous R 1 families.
In the summer of 2017 (May 31 to October 31, 2017), the third field trial was conducted for the R 3 generation of transgenic soybean plants. The methods followed for the field trial were similar to that of the field trials conducted in the previous years. All progenies of the homozygous R 3 lines were found to be resistant to the Liberty herbicide.
For the summer 2018 (May 31 to October 31, 2018), we tested the R 4 generation of transgenic soybean plants. We followed the same methods as in the previous years, except that the experiment was conducted at the Iowa State University Horticulture Research Station located on 55519 170 th St. in the north of Ames, IA 50010. As expected, all progenies of the homozygous R 4 lines were found to be resistant to Liberty herbicide.
(v) measurements of chlorophyll; (vi) DNA extraction and PCR analysis; (vii) RNA extraction and RT-PCR analysis; (viii) quantitative PCR (qPCR); (ix) subcellular localization of the GmDR1 protein; and (x) bioinformatics and statistical analyses can be found in the Supplementary Experimental Procedures.

Figure S1
Binary vector plasmids and PCR confirmation of transgenic soybean plants carrying the GmDR1 transgenes. Figure S2 Overexpression of GmDR1 transgenes enhances SDS resistance under growth chamber conditions. Figure S3 Transgenic lines carrying GmDR1 transgenes showed enhanced foliar SDS resistance under field conditions. Figure S4 Transgenic lines carrying GmDR1 transgenes exhibited similar plant height and seed size and seeds per plant as in nontransgenic Williams 82. Figure S5 Expression of GmDR1 conferred immunity to twospotted spider mites.

Figure S6
Transgenic soybean lines carrying GmDR1 transgenes expressed resistance to two-spotted spider mites. Figure S7 Transgenic soybean lines carrying GmDR1 transgenes expressed resistance to soybean aphids. Figure S8 Transgenic soybean lines overexpressing GmDR1 showed enhanced SCN resistance. Figure S9 Responses of transgenic soybean lines overexpressing GmDR1 to SCN. Figure S10 Expression levels of the genes containing Promoter 2 and Promoter 3. Figure S11 Expression of GmDR1 transgenes in leaves of transgenic soybean plants. Figure S12 Phylogenetic tree and alignment of the GmDR1 and its closely related homo-and homeologues. Figure S13 GmAI1 is constitutively up-regulated in GmDR1 transgenic lines. Figure S14 Putative structure of GmDR1. Figure S15 Sub-cellular localization of GmDR1. Table S1 Description of the three promoters used in generating the GmDR1 fusion genes.