Isoflavonoid-specific prenyltransferase gene family in soybean: GmPT01, a pterocarpan 2-dimethylallyltransferase involved in glyceollin biosynthesis Arjun Sukumaran1,2, Tim McDowell1, Ling Chen1, Justin Renaud1 and Sangeeta Dhaubhadel1,2,* 1Agriculture andAgri-FoodCanada, LondonResearch andDevelopment Centre, 1391 Sandford Street, London, ON, Canada, and 2Department of Biology, University of Western Ontario, London, ON, Canada Received 1 March 2018; revised 30 August 2018; accepted 3 September 2018; published online 3 September 2018. *For correspondence (e-mail sangeeta.dhaubhadel@canada.ca). SUMMARY Phytoalexin glyceollins are soybean-specific antimicrobial compounds that are derived from the isoflavonoid pathway. They are synthesized by soybean in response to extrinsic stress such as pathogen attack or injury, thereby conferring partial resistance if synthesized rapidly at the site of infection and at the required con- centration. Soybean produces multiple forms of glyceollins that result from the differential prenylation reac- tion catalyzed by prenyltransferases (PTs) on either the C-2 or C-4 carbon of a pterocarpan glycinol. The soybean genome contains 77 PT-encoding genes (GmPTs) where at least 11 are (iso)flavonoid-specific. Tran- script accumulation of five candidates GmPTs was increased in response to Phytophthora sojae infection, suggesting their role in phytoalexin synthesis. The induced GmPTs localize to plastids and display tissue- specific expression. We have in this study identified two additional GmPTs: an isoflavone dimethylallyltrans- ferase 3 (IDT3); and a glycinol 2-dimethylallyl transferase GmPT01. GmPT01 prenylates (�)-glycinol at the C-2 position, localizes in the plastid, and exhibits root-specific gene expression. Furthermore, its expression is induced rapidly in response to stress, and is associated with a quantitative trait loci linked with resistance to P. sojae. Based on these results, we conclude that GmPT01 are possibly one of the loci involved in confer- ring partial resistance against stem and root rot disease in soybean. Keywords: phytoalexins, glyceollin, isoflavonoids, prenyltransferase, partial resistance, soybean, gene expression. INTRODUCTION Soybean (Glycine max [L.] Merr.) is one of the most eco- nomically important and widely grown grain legumes worldwide. Soybean seeds are an excellent source of pro- teins, essential fatty acids and specialized metabolites such as isoflavonoids and saponins and, therefore, used as a component of food, animal feed and a wide range of industrial products. Soybean is Canada’s third largest prin- cipal field crop in terms of acreage, and has become a major part of the country’s economy (soycanada.ca/statis- tics/). However, there are numerous biotic and abiotic fac- tors that pose problems in soybean production. One of the major deterrents to maximizing the yield of soybean is the oomycete Phytophthora sojae. First identified in the USA in the 1950s (Kaufmann and Gerdemann, 1958), it has since spread around the globe, causing stem and root rot dis- ease in soybean, and is responsible for up to 2 billion dol- lars in annual soybean loss worldwide (Qin et al., 2017). The most effective way to manage P. sojae is the use of soybean cultivars that carry resistance to P. sojae (Rps) gene(s). Soybean Rps genes encode for nucleotide-binding site leucine-rich repeat receptor proteins, which will recog- nize the effector avirulence protein (Avr) of P. sojae, and provide race-specific resistance. A total of 27 Rps genes have been identified so far that will counteract more than 200 races of P. sojae (Sahoo et al., 2017). Although the Rps genes provide complete resistance, they recognize only a narrow race of P. sojae. Furthermore, as the pathogen con- tinues to evolve, these Rps genes become ineffective to new races of P. sojae leading to disease susceptibility. Phytoalexins are host-induced antimicrobial compounds that are released by plants in response to a variety of stress including pathogen attack, and provide partial resis- tance. This type of resistance involves the effect of multiple genes and genomic regions known as quantitative trait loci (QTL), and is more durable, broad-spectrum and non-race- specific, thus effective in environments with diverse © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology. Reproduced with the permission of the Minister of Agriculture and Agri-Food Canada. 966 The Plant Journal (2018) 96, 966–981 doi: 10.1111/tpj.14083 mailto:sangeeta.dhaubhadel@canada.ca P. sojae populations under moderate disease pressure. Furthermore, strong partial resistance did not negatively impact soybean yield (Dorrance et al., 2003). Glyceollins are isoflavonoid pterocarpan phytoalexins in soybean that are elicited as a defense mechanism against pathogens such as P. sojae, Sclerotonia sclerotiorum and Macrophomina phaseolina (Lygin et al., 2010). They are synthesized via the isoflavonoid branch of the phenyl- propanoid pathway (Figure 1a). The first step towards glyceollin production is the hydroxylation of isoflavone aglycone daidzein by isoflavone 2’-hydroxylase (Kochs and Grisebach, 1986; Akashi et al., 1998), followed by subse- quent reduction (Fischer et al., 1990a) and cyclization (Fischer et al., 1990b) to produce glycinol, the parent non- prenylated 6a-hydroxypterocarpan (Schopfer et al., 1998). Soybean predominantly produces three isomers of glyce- ollin: glyceollin I, glyceollin II and glyceollin III. However, the presence of glyceollin IV, glyceollin V, glyceollin VI and glyceofurans has also been reported in challenged soy- bean (Simons et al., 2011). The main glyceollin isomers result from the differential prenylation of glycinol at either the C-4 or C-2 position by prenyltransferases (PTs) [glycinol 4-dimethylallyltransferase (G4DT) or glycinol 2-dimethylal- lyltransferase (G2DT)] to produce a precursor, which is then cyclized by glyceollin synthase to produce glyceollins. Prenyltransferases are a class of enzymes that transfer an allylic prenyl moiety to an acceptor molecule (Figure 1b; Liang et al., 2002). The transfer of the prenyl group is an electrophilic or a Friedel–Crafts alkylation. Depending on the stereochemistry of the product these enzymes are split into two classes, trans- or cis-PT. Within trans-PTs, there are two aspartate-rich motifs that are essential for sub- strate binding and catalytic activity. These aspartate-rich motifs are not found in cis-PTs (Takahashi and Koyama, 2006). The UbiA superfamily of intramembrane aromatic PTs is involved in the prenylation of many plant bioactive compounds (Li, 2016), and contains at least one aspartate- rich motif (NDxxDxxxD) and requires the presence of a divalent cation (e.g. Mg2+) for activity (Saleh et al., 2009). Two isoflavonoid-specific PTs, G4DT (Akashi et al., 2009) and G2DT (Yoneyama et al., 2016), belonging to the UbiA superfamily have recently been identified in soybean. Soy- bean is a paleopolyploid whose genome has undergone two whole genome duplications at approximately 59 and then 13 million years ago, resulting in 75% of genes within the genome in multiple copies (Schmutz et al., 2010). Many isoflavonoid biosynthetic genes contain large gene families where spatio-temporal expression and subcellular location of specific members of the gene family determine the composition of the isoflavonoid metabolon and synthe- sis of specific metabolites (Dastmalchi et al., 2016). There- fore, it is crucial to identify all the members of the gene families involved in the glyceollin biosynthetic pathway, and determine their role at diverse circumstances. Here, we performed a genome-wide analysis of the PT genes in soybean, and identified a total of 11 putative (iso)flavonoid-specific GmPTs. Because glyceollin produc- tion is an induced response, we checked the candidate GmPTs for their stress-induced expression, and identified five candidates among which two are unique PTs, a G2DT-2, and a isoflavone dimethylallyltransferase 3 (IDT3). Elucidation of subcellular localization and spatio- temporal expression of the stress-induced GmPT revealed their isoform-specific roles. Our results suggest that GmPT01 (G2DT-2) is possibly one of the loci involved in conferring partial resistance against stem and root rot dis- ease in soybean as it is associated with the QTL linked with P. sojae resistance. RESULTS Soybean (iso)flavonoid prenyltransferase gene family contains 11 members To identify isoflavonoid-specific GmPTs, we first searched the annotated G. max Wm82.a2.v1 genome on the Phyto- zome database for all the PTs using the key words ‘dimethylallyltransferase’ and ‘prenyltransferase’. This search identified 77 putative GmPTs involved in various metabolic processes. Each identified GmPT was then used as a query for a BLASTP search to ensure no GmPTs were missed by the keyword search, but no additional GmPTs were identified. UbiA superfamily PTs are involved in multiple biosyn- thetic pathways in plants such as homogentisate, ubiqui- none, shikonin and flavonoid biosynthesis. PTs involved in a similar function or common metabolic pathway have been shown to cluster together in a phylogenetic tree (Aka- shi et al., 2009; Karamat et al., 2014; Li, 2016). Therefore, to identify GmPTs involved in isoflavonoid biosynthesis, we created a neighbor-joining tree from the amino acid sequences of 77 putative GmPTs and previously character- ized PTs from Karamat et al. (2014), with the assumption that isoflavonoid-specific GmPTs will cluster together in the (iso)flavonoid group. As shown in Figure 2, out of 77 putative GmPTs, only 11 grouped in the flavonoid clade that included previously identified G4DT (Gly- ma.10G295300) and G2DT (Glyma.20G245100) from soy- bean (Akashi et al., 2009), N8DT from Sophora flavescens (Sasaki et al., 2008) and PT1 from Lupinus albans (Shen, 2012). The UbiA superfamily of intramembrane PTs contains seven–nine transmembrane (TM) a-helices (Ohara et al., 2009). Furthermore, all UbiA PTs contain at least one aspartate-rich motif (Saleh et al., 2009), with flavonoid/ho- mogentisate PTs possessing a second conserved sequence critical for enzyme activity (Akashi et al., 2009). Therefore, we analyzed 11 candidate GmPTs that clustered in the fla- vonoid clade for TM domains and critical residues. Similar © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 967 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense to G4DT (GmPT10a; Figure 3a), all (iso)flavonoid-specific candidate GmPTs were predicted to contain eight or nine TM domains, except for GmPT08 (Glyma.08G274800) and GmPT10c (Glyma.10G070200) that contained only five. A search for the aspartate-rich motifs within candidate GmPTs identified that GmPT10c was missing an aspartate Figure 1. Biosynthesis of phytoalexin isoflavonoid glyceollin in soybean. (a) The pathway starts with phenylalanine and branches to produce diverse specialized metabolites. The isoflavonoid pathway produces three isoflavone agly- cones (highlighted in yellow), which are normally conjugated with glucose and malonylglucose sequentially and stored in vacuole. Upon exposure to stress, daidzein acts as the key substrate for further enzymatic reactions that culminate in the production of glyceollins (blue text). The dotted arrows represent multiple steps within the pathway. I2’H, isoflavone 2’-hydroxylase; 2HDR, 2’-hydroxydaidzein reductase; PTS, pterocarpan synthase; 3,9DPO, 3,9-dihydroxypterocarpan 6a-monooxygenase; G4DT, glycinol 4-dimethylallyltransferase; G2DT, glycinol 2-dimethylallyltransferase; GS, glyceollin synthase. G2DT and G4DT used in this study are shown in red text. (b) Enzymatic reaction catalyzed by G2DT and G4DT converting (�)-glycinol to glyceocarpin and 4-glyceollidin, respectively. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 968 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense residue (Figure 3b). Because GmPT08 and GmPT10c con- tain fewer TM domains to qualify as UbiA PT, and GmPT10c lack critical residue for enzymatic activity, these two candidates were eliminated for further study. Detailed characteristics of the remaining nine candidate GmPTs are shown in Table 1. Most of the candidates GmPTs were predicted to contain a single transcript, except for GmPT01, GmPT11a and GmPT11b, which were Figure 2. Phylogenetic analysis of soybean prenyltransferases (PTs). The deduced protein sequences of the soybean PTs and other known PTs were aligned using ClustalΟ, followed by construction of a neighbor-joining tree with 1000 bootstrap replications using MEGA 7 software. Colors indicate separate clades as labeled. Isoflavonoids-specific PTs are indicated in parenthesis (bold). Ap, Allium porrum; At, Arabidopsis thaliana; Cp, Cuphea pulcherrima; Cr, Chlamydomonas reinhardtii; HI, Humulus lupulus; Hv, Hordeum vulgare; La, Lupinus albus; Le, Lithospermum erythrorhizon; Os, Oryza sativa; Sf, Sophora flavescens; Ta, Triticum aestivum; Zm, Zea mays. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 969 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense predicted to contain six, three and three alternate tran- scripts, respectively. Pairwise nucleotide coding sequence comparisons between candidate GmPTs range from 73.01 to 94.7% identity, while the pairwise amino acid sequence comparisons range from 56.74 to 91.89% (Table S1). Five GmPTs are induced upon pathogen infection Isoflavonoid biosynthetic gene expression and glyceollin production are induced by stress including pathogen infec- tion (Ayers et al., 1976; Ward et al., 1979; Kimpel and Kosuge, 1985; Bhattacharyya and Ward, 1986). To deter- mine if candidate GmPT genes show stress response, G. max cv. L76-1988 stems were infected with P. sojae Race 7, and samples were collected at 24, 48 and 72 h post-infection. Expression analysis of nine candidate GmPT genes in control and P. sojae-infected samples was performed by reverse transcriptase-polymerase chain reac- tion (RT-PCR) using gene-specific primers. As shown in Figure 4(a), expression of five GmPTs including GmPT01, GmPT10a, GmPT10d, GmPT11a and GmPT20 was induced upon infection. A low level of induced expression was detected for GmPT10b, while no transcript accumulation of GmPT03, GmPT11b and GmPT11c was observed in both control and treated samples. Silver nitrate induces glyceollin production (St€ossel, 1982), and thus has been used to mimic plant response to Figure 3. Identification of catalytic site residues in isoflavonoid-specific candidate GmPTs. (a) Predicted structure of G4DT (GmPT10a) showing multiple transmembrane (TM) domains and con- served first aspartate-rich motif (FARM) and second aspartate-rich motif (SARM). (b) Multiple sequence alignment of candidate GmPTs and previously characterized prenyltrans- ferases (PTs) using ClustalO. Abridged version of alignment is shown to highlight both FARM and SARM. Red rectangle in SARM indicates missing D residue in GmPT10c. [Colour figure can be viewed at wileyonlinelibrary.com]. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 970 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense www.wileyonlinelibrary.com pathogen attack (Bhattacharyya and Ward, 1986; Moy et al., 2004; Sepiol et al., 2017). For quantitative analysis of stress-induced GmPT expression, etiolated soybean hypo- cotyls were treated with 1 mM AgNO3 or water (control), and samples were collected at 6, 12, 24, 48 and 72 h post- treatment. Treated soybean hypocotyls developed brown lesions at 12 h, which increased in size and color with time. RNA was isolated from treated and control samples, and expression of the stress-induced GmPT genes from Fig- ure 4(a) were examined in detail by quantitative (q)PCR. As shown in Figure 4(b), GmPT01 gene expression peaked rapidly and plateaued across the time course. GmPT10a, GmPT10d and GmPT11a were all early induced, and their transcript accumulation level began to decline 12 h post- treatment. No induced expression was observed for GmPT20 at the early time points; however, its expression gradually increased over the time course, with highest expression occurring at 48 h and was still present at high levels at 72 h. Because phytoalexin production is an induced response, only stress-induced GmPTs were char- acterized in detail. Candidate GmPTs show tissue-specific gene expression and localize in the plastids To determine the spatio-temporal gene expression of the induced GmPTs under normal growth and development, RNA was extracted from vegetative and reproductive tis- sues at different developmental stages in soybean, and qPCR analysis of five stress-induced GmPTs (GmPT01, GmPT10a, GmPT10d, GmPT11a and GmPT20) was conducted. The relative expression of GmPTs in different tissues is shown in Figure 5. Compared with other tissues, the highest level of GmPT01 transcript accumulation was found in leaf tissue followed by flower and root, while no transcript for GmPT01 was detected in developing embryo. GmPT10a was highly expressed in mature embryos [70 days after pollination (DAP)] before they start approaching desiccation. Transcript accumulation of GmPT10a was also detected in root, leaf and flower. GmPT10d displayed the largest distribution of expression. It was expressed in all the tissues under the study except in stems, but showed highest expression in the late embryo, flower bud and seed coat. GmPT11a transcript level was highest in flower among the other tissues; however, the relative expression level was low in all. GmPT20 was mostly expressed in the leaf and roots. The tissue-specific expression profile of GmPT01 and GmPT20 displayed close resemblance, but for the most part the other GmPTs had unique expression profiles. All five candidate GmPTs are predicted to possess a plastid transit peptide and localize in the plastid. To con- firm their subcellular localization, we created a transla- tional fusion of the full-length candidate GmPTs with yellow fluorescent protein (YFP). These constructs were transiently expressed in the leaf epidermal cells of Nico- tiana benthamiana, and fluorescence was visualized by confocal microscopy. Confirmation of the plastid localiza- tion was performed by co-expressing the GmPT-YFP con- struct with a plastid localization signal fused to cyan fluorescent protein (CFP). All five induced GmPTs show Table 1 Characteristics of soybean (iso)flavonoid-specific prenyltrasferase gene family Gene name Locus name Locus range Predicted molecular weight (kDa) Coding sequence length (bp) Splice variant (s) Predicted number of TM domains Function GmPT01 Glyma.01G134600 Chr01:45619439..45625938 reverse 43.66 1182 6 9 G2DTa GmPT03 Glyma.03G033100 Chr03:3833718..3843653 reverse 46.72 1239 1 9 nd GmPT10a Glyma.10G295300 Chr10:51250135..51256351 forward 46.02 1230 1 9 G4DT GmPT10b Glyma.10G070100 Chr10:6923409..6931780 forward 41.07 1080 1 9 nd GmPT10d Glyma.10G070300 Chr10:7023173..7029710 forward 45.5 1209 1 8 IDT3a GmPT11a Glyma.11G210300 Chr11:30238668..30248588 reverse 46.24 1266 3 9 IDT1 GmPT11b Glyma.11G210400 Chr11:30278128..30291380 reverse 47.51 1233 3 9 IDT2 GmPT11c Glyma.11G210500 Chr11:30310328..30320581 reverse 46.32 1227 1 9 nd GmPT20 Glyma.20G245100 Chr20:47561270..47568860 forward 45.84 1227 1 9 G2DT nd, not determined; TM, transmembrane. aExperimentally characterized in this study. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 971 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense plastid localization (Figure 6), with GmPT10a replicating the previous finding (Akashi et al., 2009). GmPT20, uniquely, was localized to both the cytoplasm and the chloroplast. When the predicted transit peptide for plastid targeting of GmPTs was removed, and their subcellular localization was determined, they were not confined to cytoplasm but also observed in other subcellular compart- ments (Figure S1). Functional analysis of GmPTs To determine the enzymatic function of candidate GmPTs, truncated GmPT without the predicted transit peptide was cloned in a yeast expression vector. Yeast microsomal frac- tions containing GmPTs were prepared to perform an in vitro enzyme assay for their ability to transfer a prenyl group to glycinol. Using semi-preparative high-perfor- mance liquid chromatography (HPLC), (�)-glycinol was iso- lated from AgNO3-treated soybean seeds (Figure S2). The purity of the isolated compound was measured by HPLC coupled to an evaporative light-scattering detector, and the chemical structure of (�)-glycinol was confirmed by NMR spectroscopy (Figure S3). 1H and 13C chemical shifts and coupling constants (J; Table S2) agree with those reported in the literature (Akashi et al., 2009; Yoneyama et al., 2016). In vitro enzyme assay was performed by incubating microsomal fractions containing N-terminal truncated GmPTs with glycinol, in the presence of DMAPP and MgCl2, and the reaction products were analyzed by HPLC (Figure 7). Previously characterized G4DT (GmPT10a) was used as a positive control (Akashi et al., 2009). The reaction product of GmPT10a was confirmed by LC-MS with the mass-to-charge ratio (m/z) of 323.1270 corresponding to that of 4-glyceollidin (Figure 7c). Yoneyama et al. (2016) recently found that GmPT20 prenylates glycinol on the C-2 position, thereby naming this enzyme as G2DT. Further- more, it was reported that GmPT11a was incapable of prenylating glycinol, but instead was found to prenylate daidzein and was, thereby, named isoflavone dimethylallyl- transferase 1 (IDT1). Of the two remaining candidate GmPTs characterized in our study, GmPT10d was unable to use glycinol as a substrate, however, it prenylated daid- zein (Figure 7g) and genistein (Figure 7h). No product was formed when yeast microsomes expressing GmPT10d Figure 4. Expression analysis of GmPT genes in response to stress. (a) Transcript analysis of isoflavonoid-specific candidate GmPTs in response to Phytophthora sojae infection. Total RNA (1 lg) extracted from P. sojae- infected (T) or control (C) stems of soybean cv. L76-1988 for 24, 48 or 72 h was used for reverse transcriptase-polymerase chain reaction (RT-PCR). CONS4 was used as a loading control. (b) Detail transcript analysis of stress-induced GmPTs in response to AgNO3 treatment. Total RNA (1 lg) of soybean cv. Harosoy63 was used for cDNA synthesis and quantitative (q)PCR analysis using gene-specific primers. Error bars denote standard error of the mean (SEM) of three biological and three technical replicates per biological replicate. Values were normalized against the reference gene, CONS4. The asterisk (*) denotes statistically sig- nificant expression (one-tail t-test, P < 0.05). [Colour figure can be viewed at wileyonlinelibrary.com]. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 972 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense www.wileyonlinelibrary.com were incubated with glycitein. The products of the GmPT10d reactions were analyzed by LC-MS with the m/z of 323.1268 corresponding to that of prenylated daidzein and m/z of 339.1224 corresponding to prenylated genistein (Figure 7g,h). The results demonstrate that GmPT10d can transfer prenyl moiety to isoflavone aglycones daidzein and genistein; therefore, it can be named as IDT3. GmPT01 was previously reported as having no PT activ- ity (Yoneyama et al., 2016). On the contrary, when micro- somal fraction containing N-terminal truncated GmPT01 was incubated with glycinol in the presence of DMAPP and Mg2+, a product was formed. LC-MS analysis confirmed the m/z of the product as a prenylated glycinol (Figure 7d). The reaction product was recovered from a large-scale experiment for its chemical structure analysis by 1H NMR. The results revealed no signal for H-2 in the 1H proton spectra for the unknown product, while a signal at 6.35 ppm was present corresponding to H-4, suggesting that the dimethylallyl group was introduced at position 2 in the enzymatic assay, thereby identifying GmPT01 to be a new G2DT (Table S2). GmPT01 is located within a QTL linked to P. sojae resistance, and is expressed at higher levels in resistant soybean cultivars Recently, we conducted a survey of QTLs linked with P. so- jae resistance in soybean and identified 77 QTLs (Sepiol et al., 2017). These QTLs were searched to determine if they contain any of the stress-induced GmPT loci. The results revealed that a QTL, Phytoph 13-3, located on chromosome 1 (Wang et al., 2012), contains the GmPT01 locus (Chr 01:45,619,439..45,625,938) with a LOD score of 3.7. Phytoph 13-3 spans a 435-kb pair region (42,423,585..46,773,166) that contains a total of 166 genes of which 42 are not annotated (Table S3). Among the annotated genes, there are some defense-related genes such as those involved in the synthe- sis of ethylene, abscisic acid, baicalein and flavonoids. One interesting gene located adjacent to GmPT01 within the QTL Phytoph 13-3 is Glyma.01G134700, annotated as ‘trihy- droxypterocarpan dimethylallyltransferase/glyceollin syn- thase’. Amino acid sequence comparison of the protein encoded by Glyma.01G134700 and GmPT01 revealed that they share strong sequence identity within the first 50 amino acids. Furthermore, Glyma.01G134700 encodes for a protein with a low molecular mass (7.39 kD) that does not contain any amino acid residue essential for PT/cytochrome P450 activity. To investigate if the expression level of GmPT01 differs in resistant and susceptible soybean cultivars, we mined the publicly available GEO datasets with the accession GSE7124 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi? Figure 5. Tissue-specific expression of stress-induced isoflavonoid-specific candidate GmPTs during normal growth and development. Total RNA (1 lg) was extracted from various soybean tissues, including root, stem, leaf, flower bud, flower, embryo (30, 40, 50, 60 and 70 DAP), seed coat and pod wall, and used for quantitative polymerase chain reac- tion (qPCR) analysis using gene-specific primers. Error bars denote standard error of the mean (SEM) of two biological and three technical replicates. Values were normalized against the reference gene, CONS4. [Colour figure can be viewed at wileyonlinelibrary.com]. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 973 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense http://www.ncbi.nlm.nih.gov/nuccore/GSE7124 https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE7124 www.wileyonlinelibrary.com acc=GSE7124). This dataset contains differential expres- sion of genes during P. sojae infection of soybean cultivars differing in quantitative resistance using Affymetrix Soy- bean Genome Array. The experiment was conducted using RNA samples from P. sojae inoculated or mock-treated (control) eight soybean cultivars at two time points (72 and 120 h) with four experimental replicates. The analysis of GmPT01 transcript abundance in soybean cultivars revealed that P. sojae-resistant soybean cultivars accumu- lated a higher level of GmPT01 transcript compared with medium and lowly resistant (susceptible) cultivars in control samples after 120 h mock-treatment (Figure 8a). Similar transcript patterns were observed at 72 h mock- treatment of highly and medium resistant cultivars, although GmPT01 transcript abundance in susceptible cul- tivars was comparable to that of resistant cultivars at this time point. It is possible that mock-treated plants were also stressed to some extent until 72 h, but later became stabi- lized during 120 h. No difference in the levels of GmPT01 transcript among these soybean cultivars was observed in Figure 6. Subcellular localization of the candidate GmPTs. A translational fusion of GmPT-YFP was transiently expressed in the leaf epidermal cells of Nicotiana benthamiana and visualized by confocal microscopy. Confirmation of localization was performed through co-localization of GmPT-YFP fusion with a plastid marker fused to CFP. Merged signal was obtained by sequential scanning of the two channels. Scale bar: 10 lM. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 974 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE7124 pathogen-treated samples. The experiment did not include early time points. We also analyzed GmPT01 gene expression in root tis- sues of 2-week-old seedlings of several soybean cultivars that differ in P. sojae resistance. As illustrated in Fig- ure 8(b), expression of GmPT01 tended to be higher in P. sojae-resistant cultivars (Conrad, AC Colombe, Haro- soy63) compared with most susceptible cultivars (AC Glengary, TNS, Maple Ridge Brown, AC Albatros, and OX760-6). These results suggest that soybean cultivars with higher resistance against P. sojae consistently accu- mulate higher basal level of GmPT01 transcript. Coding sequence comparison of GmPT01 from the cultivars shown in Figure 8(b) did not show any allelic variation specific to the resistant or the susceptible lines. It is possi- ble that the differential expression of GmPT01 in these cultivars (Figure 8b) was due to variation in its promoter and/or intron regions. Figure 7. Functional analysis of candidate GmPTs. Microsomal fraction containing N-terminal truncated GmPT was incubated with substrate, DMAPP and MgCl2. Ethyl acetate extract was analyzed by high-performance liquid chromatography (HPLC). (a) Glycinol substrate only. (b) Microsomal protein containing GmPT01 was used for enzymatic reaction without the substrate glycinol. (c) GmPT10a (G4DT) was used a positive control. (d) GmPT01 was able to prenylate glycinol. S, substrate glycinol; P, product. MS spectra of the reaction products are shown. (e) and (f) substrates daidzein and genistein, respectively. (g) GmPT10d was able to prenylate daidzein. (h) GmPT10d was able to prenylate genistein. LC-MS spectra of the reaction products of GmPTs depict the parent ion, the diagnostic loss of the prenyl group, and the prenyl cation. [Colour figure can be viewed at wileyonlinelibrary.com]. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 975 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense www.wileyonlinelibrary.com DISCUSSION Soybean is a paleopolyploid with a considerable number of large gene families (Schmutz et al., 2010). For example, several isoflavonoid biosynthetic genes including chalcone synthase (CHS; Tuteja and Vodkin, 2008), chalcone reduc- tase (Sepiol et al., 2017), chalcone isomerase (Dastmalchi and Dhaubhadel, 2015) and isoflavone synthase (Jung et al., 2000) exist as multi-gene families. They are pre- dicted to be the result of whole genome, tandem and segmental duplications in soybean. Following duplication event, stress-responsive genes have a higher probability of retention (Hanada et al., 2008) as it provides genetic robustness by preserving the functional compensations for severe phenotypic effects in case of null mutation (Hanada et al., 2009). An essential part of the soybean defense response, isoflavonoid biosynthesis culminates in glyce- ollin production. After the hydroxylation of daidzein, the pathway follows a linear progression until the prenylation of glycinol, leading to the production of the different Figure 8. Accumulation of GmPT01 in soybean cultivars varying in partial resistance against Phytophthora sojae. (a) Box plots showing the comparison of GmPT01 in soybean cultivars with varied levels of quantitative resistance, by using Affimetrix Soybean Gen- ome Array analysis. GmPT01 transcript levels were calculated from the RNA samples of mock-inoculated (control) soybean tissues at 72 and 120 h post- inoculation. High, medium and low indicates degree of resistance. Treat- ments with the same letter are not significantly different as determined by one-way ANOVA followed by Tukey’s post hoc test for multiple comparisons (P < 0.05). (b) Expression analysis of GmPT01 in soybean roots collected from 2-week- old seedlings from different soybean cultivars using quantitative poly- merase chain reaction (qPCR). Error bars indicate SEM of three biological replicates where three technical replicates per biological replicate were used. [Colour figure can be viewed at wileyonlinelibrary.com]. Figure 9. Isoflavonoid-specific GmPT distribution in soybean. The proposed model suggests that isoflavonoid-specific PTs are spatially distributed, and respond according to the proximity of the stress. The model highlights 12 soybean tissues: root, stem, leaf, flower bud, flower, embryo 30–70 days after pollination (DAP), seed coat and pod wall. Isoflavonoid-specific GmPT genes are expressed predominantly in the designated sites. The presence and relative abundance is highlighted by the corresponding color associ- ated in the legend. Stem and embryo 30–50 DAP remain uncolored, indicat- ing no basal-level expression of any GmPT. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 976 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense www.wileyonlinelibrary.com glyceollin isomers that are prenylated either at the C-2 or C-4 position (Figure 1). Two PTs, G4DT and G2DT, which function in the production of precursors for glyceollin iso- mers, have been reported previously (Akashi et al., 2009; Yoneyama et al., 2016). However, the presence of addi- tional isoflavonoid-specific GmPTs cannot be ruled out. Through extensive mining of the annotated soybean genome, we identified a total of 77 putative PT genes in soybean where nine of them clustered together in the same branch of phylogenetic tree with two previously identified isoflavonoid-specific GmPTs from soybean and flavonoid PTs from other plant species (Figure 2), sug- gesting there are at least 11 putative (iso)flavonoid-speci- fic PTs. Our approach provides more robustness as we first identified all the putative GmPTs present in soybean genome, and then categorized them into different groups. This study not only identified (iso)flavonoid-specific GmPTs, but also PTs involved in other metabolic pro- cesses in soybean (Figure 2). After analysis of critical amino acid residues for enzymatic function, number of TM domains to qualify as UbiA PT, and stress-induced expression of 11 GmPT genes, we selected five candi- dates for detail characterization (Figures 3 and 4). Though six eliminated GmPTs either did not respond to stress or are structurally different from other UbiA PT, it does not rule out the possibility of having enzymatic activity or performing a different function. Among the five stress- induced GmPTs, GmPT10a and GmPT20 catalyze the reactions that transfer prenyl moiety to C-4 (Akashi et al., 2009) and C-2 (Yoneyama et al., 2016) position of (�)-gly- cinol, respectively, while GmPT11a is an IDT2 (Yoneyama et al., 2016). Here, we report the identification of two additional GmPTs, GmPT01 as a G2DT, and GmPT10d as an IDT3. Many isoflavonoid biosynthetic genes are upregulated in response to pathogen attack (Moy et al., 2004; Vega- S�anchez et al., 2005). Among the five isoflavonoid-specific GmPT genes (GmPT01, GmPT10a, GmPT10d, GmPT11a and GmPT20) that are upregulated upon stress (Figure 4), three are involved in glyceollin biosynthesis. While GmPT10a catalyzes the reaction to produce 4-glyceollidin, a precursor for glyceollin I and glyceollin VI, GmPT20 and GmPT01 are involved in the production of glyceocarpin, a precursor for glyceollin II to glyceollin V and glyceofuran (Figure 1). The GmPTs belonging to (iso)flavonoid branch formed two distinct clusters: G2DT/G4DT and IDT (Fig- ure 2). Glyma.08G274800 is present in G2DT/G4DT clade and is closely related to GmPT10a. Despite that it contains only five TM domains, it is possible that Glyma.08G274800 may possess G4DT activity or may have lost its function during evolution. Furthermore, accumulation of both prenylated daidzein and prenylated genistein has been reported when soybean was stressed with lactofen (Cheng et al., 2011), therefore, corroborating the finding seen in Figure 4 where GmPT10d and GmPT11a showed induced expression in response to stress. The stress-induced GmPTs contain predicted transit pep- tide for their localization in plastid. As previously shown for GmPT10a (G4DT; Akashi et al., 2009), other GmPTs also localize in the plastids, except for GmPT20, which was found in both the plastid and the cytoplasm (Figure 6). ATP sulfury- lase 2 involved with sulfur assimilation in Arabidopsis thali- ana was found to encode both plastidic and cytosolic isoforms. This differential localization resulted from transla- tional initiation at different start codons within its transit pep- tide (Bohrer et al., 2015). Furthermore, an evidence for the existence of parallel cytosolic and plastid pathways was found for tyrosine metabolism in soybean (Schenck et al., 2015). For the functional characterization of GmPTs, we removed the predicted transit peptide at the N-terminus, and produced truncated proteins for the enzyme assay. These truncated proteins were also evaluated for their localization in cellular compartments. In contrast to the full-length GmPTs, the truncated GmPTs localized in the cytoplasm as well as other subcellular compartments (Figure S1). The truncated GmPT01 lacking N-terminus 21 amino acids was able to prenylate (�)-glycinol at the C-2 position (Figure 7; Table S2), while deletion of 26 amino acids at the N-terminus of GmPT01 resulted in a non-functional protein (Yoneyama et al., 2016), suggesting the amino acid residue(s) within 22ASKA26S may be critical for GmPT01 activity. Even prior to glyceollin accumulation, precursor mole- cules were found to accumulate at the site of infection (Hahn et al., 1985; Graham et al., 1990). The transcript accumulation of stress-induced GmPTs was also detected under normal developmental conditions in different soy- bean tissues (Figure 5). This basal level of expression may lend to a mechanism to potentially trigger rapid phy- toalexin production in the event of pathogen infection or upon exposure to any other stress. Subfunctionalization of two members of grapevine cation/proton antiporter 1 gene family in response to osmotic and salt stress was discov- ered where they function in different tissues and in differ- ent stages of stress (Ma et al., 2015). Similar subfunctionalization by their unique tissue-specific expres- sion profiles was observed for CHS gene family in soybean (Tuteja et al., 2004). Following this model, subfunctional- ization of an ancestral isoflavonoid PT would have resulted in the formation of the present GmPT gene family, which contains GmPTs with altered tissue expression. Differential localization of GmPT expression across various tissues would eliminate the necessity to express all GmPTs in all tissues. This distribution may be a mechanism by which the plant selectively responds to stress. Because isoflavo- noids accumulate in many tissues in soybean (Dhaubhadel et al., 2003), the isoflavonoid metabolon is expected to be localized ubiquitously throughout the plant. However, the GmPTs that comprise these metabolons may differ © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 977 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense according to the tissue and time of infection. For example, GmPT01 expression was induced during the early period of infection, while an increased level of GmPT20 was noticed later on (Figure 4). Figure 9 illustrates a visual depiction suggesting that isoflavonoid-specific PTs are spatially distributed under normal conditions. Upon expo- sure to stress, they respond according to the proximity of the stress; however, the rapidity of stress response and accumulation of phytoalexin glyceollin will lead into dis- ease resistance. Partial resistance to sudden death syn- drome caused by Fusarium solani is defined by the ability of soybean plant to rapidly produce sufficient levels of glyceollin at the site of infection (Lozovaya et al., 2004). Induction of glyceollin synthesis was found in both resis- tant and susceptible soybean genotypes following infec- tion; however, a faster and greater response was observed in the partially resistant genotype. There are several other evidences indicating a critical role of phytoalexins in dis- ease resistance only when their synthesis and accumula- tion are rapidly induced upon infection (Ahuja et al., 2012). Screening for QTL and QTL markers has been a large component of breeding for resistance in plants. A detailed search of the soybean database (Grant et al., 2010) and lit- erature identified 77 QTLs associated with P. sojae resis- tance in soybean (Sepiol et al., 2017), where GmPT01 was found within a QTL Phytoph 13-3 for resistance to P. sojae suggesting a high probability of genetic linkage or pheno- typic association. Despite that QTL regions generally cover several megabase pairs and contain hundreds to thou- sands of genes (Dupuis and Siegmund, 1999), Phytoph 13- 3 only spans 435 kb pairs with 166 genes. Even though there are some other defense-related genes such as ethy- lene- and abscisic acid-related genes within the region, GmPT01 is the most likely candidate linked with the trait. Furthermore, the basal expression level of GmPT01 is higher in P. sojae-resistant cultivars compared with sus- ceptible ones (Figure 8). Previously, we reported GmCHR2A with a higher basal expression level in P. sojae- resistant cultivars compared with susceptible cultivars (Sepiol et al., 2017). Because partial resistance is governed by the effect of multiple genes, identification of GmPT01 provides an additional insight into the disease resistance mechanism. Because GmPT01 is expressed in roots, is most rapidly induced in response to stress, and lies in the chromosomal region linked to P. sojae resistance, we con- clude that GmPT01 is possibly one of the loci involved in the partial resistance to root rot disease and can be utilized in resistance breeding. EXPERIMENTAL PROCEDURES Plant materials and growth conditions Nicotiana benthamiana was grown in a growth chamber set to 16 h light at 25°C and 8 h dark at 20°C cycle, with 60–70% relative humidity and light intensity of 100–150 lmol m�2 sec�1. For etio- lated hypocotyl, soybean (G. max L. Merr.) cv. Harosoy63 seeds were grown in a growth chamber set to 25°C with no light and 60– 70% relative humidity. In silico and phylogenetic analysis Putative PTs were identified from the annotated G. max Wm82.a2.v1 genome assembly on the Phytozome database (https://phytozome.jgi.doe.gov/pz/portal.html). A keyword search using ‘prenyltransferase’ and ‘dimethylallyltransferase’ was per- formed in the following protein databases: Panther, Pfam, GO, KOG and KEGG. To ensure complete coverage of the genome and to identify any unannotated GmPTs, each putative GmPT identi- fied by the keyword search was used as a query for a Protein BLAST (BLASTP) analysis against the G. max Wm82.a2.v1 gen- ome assembly. For phylogenetic analysis, amino acid sequences from all puta- tive PTs were first aligned with ClustalO, prior to constructing a neighbor-joining tree with 1000 bootstrap replications using MEGA 7 (Kumar et al., 2016). The presence of TM domains was predicted using TMHMM (Krogh et al., 2001), and subcellular localization was predicted using WoLF PSORT (Horton et al., 2007). Chloroplast signal peptide presence and cleavage site pre- diction was performed using ChloroP (Emanuelsson et al., 1999). Pairwise nucleotide and amino acid comparison was performed using the percent identity function of ClustalO. Pathogen infection and AgNO3 treatment Seven-day-old G. max cv. L76-1988 plants were inoculated with P. sojae Race 7 or water for control as described for the virulence assay (Shrestha et al., 2016). The inoculated stems of these plants were collected at 24, 48 and 72 h post-inoculation. Because AgNO3 can elicit glyceollin production (Kube�s et al., 2014), to mimic pathogen attack, soybeans were treated with AgNO3 according to the protocol used in Ward et al. (1979). Glycine max cv. Harosoy63 was grown in water-soaked vermi- culite at 25°C in the dark to obtain etiolated hypocotyls. After 6 days, five droplets (10 ll each) of either water (control) or 1 mM AgNO3 were placed onto the hypocotyl of each seedling and incu- bated in the dark at 25°C, and hypocotyl tissue was collected at 6, 12, 24, 48 and 72 h post-treatment. RNA extraction, RT-PCR and qPCR Total RNA was extracted from 100 mg of stem/hypocotyl tissue using the RNeasy Plant Mini kit (Qiagen). On-column DNA diges- tion was performed using DNase I (Promega, https://www.prome ga.ca). Total RNA from field-grown soybean tissues [root, stem, leaf, flower bud, flower, embryo at different stages of develop- ment (30, 40, 50, 60 and 70 DAP), seed coat and pod wall] was extracted following a protocol adapted from Wang and Vodkin (1994). Potential DNA contaminants were removed using the TURBO DNA-freeTM kit (Life Technologies, https://www.thermof isher.com/ca/en/home.html). For RT-PCR, DNase-treated RNA (1 lg) was reverse-transcribed to cDNA using ThermoScriptTM RT-PCR System (Life Technolo- gies), and cDNA was used as template for amplification with gene-specific primers (Table S4). CONS4 was used as a loading control. For qPCR, SsoFast EvaGreen Supermix (BioRad, http:// www.bio-rad.com/) and gene-specific qPCR primers were used (Table S4). Reactions were analyzed in a Bio-Rad C1000 Thermal Cycler with the CFX96TM Real-Time PCR System. All reactions were performed as three technical replicates, and expression was © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 978 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://phytozome.jgi.doe.gov/pz/portal.html https://www.promega.ca https://www.promega.ca https://www.thermofisher.com/ca/en/home.html https://www.thermofisher.com/ca/en/home.html http://www.bio-rad.com/ http://www.bio-rad.com/ normalized against the reference gene, CONS4. The data were analyzed using CFX manager (BioRad). Statistical analysis Significant differences in gene expression were calculated using a one-tail t-test or one-way ANOVA followed by Tukey’s post hoc test for multiple comparisons. Statistical significance was set at P < 0.05. Cloning and plasmid construction For subcellular localization, GmPTs were amplified using gate- way-compatible gene-specific primers (Table S4). Full-length GmPTs were initially recombined into pDONR/Zeo using Gateway BP Clonase� II Enzyme mix (Invitrogen, https://www.thermofishe r.com/ca/en/home/brands/invitrogen.html) and transformed into Escherichia coli DH5a via electroporation, PCR screened, and sequence confirmed. For transient expression in N. benthamiana, plasmids were recombined into destination vector pEarleyGate101 using Gateway LR Clonase� II Enzyme mix (Invitrogen) and transformed into Agrobacterium tumefaciens GV3101 via electroporation. For protein expression in yeast, truncated GmPTs were ampli- fied lacking their signal peptides using primers listed in Table S4. The truncated GmPTs were ligated into yeast expression vector, pYES2.1/V5-His-TOPO� (Invitrogen), and transformed into E. coli TOP10F’ cells (Invitrogen) using the pYES2.1 TOPO� TA Expres- sion Kit (Invitrogen). Transformation of E. coli TOP10F’ was car- ried out according to the manufacturer’s instructions with slight modifications. Upon sequence confirmation, recombinant plasmid was transformed into Saccharomyces cerevisiae BJ2168 and screened on minimal synthetic dextrose (SD)/-ura plates (Clontech, https://www.takarabio.com/). Transient expression in Nicotiana benthamiana and subcellular localization Constructs in A. tumefaciens GV3101 were infiltrated into leaves of 4–6-week-old N. benthamiana plants. The epidermal cell layers of the infiltrated N. benthamiana leaves were visualized using a Leica TCS SP2 inverted confocal microscope using a 63 9 water immersion objective lens. To validate the subcellular localization, the pEG101-GmPT constructs were co-infiltrated with a plastid organelle marker translationally fused to CFP at a 1:1 ratio. For YFP visualization, the excitation wavelength was set to 514 nm, and emission was collected at 525–545 nm. For CFP visualization, the excitation wavelength was set to 434 nm, and emission was collected at 460–490 nm. Co-localization of the YFP and CFP sig- nals was visualized by sequentially scanning for both fluorescent proteins. Yeast microsome preparation Protein induction for microsomal preparation was performed according to Akashi et al. (2009) with some modifications. Saccha- romyces cerevisiae BJ2168 containing the pYES2.1-GmPT con- struct was grown in SD/-ura media until bacteria reached their growth phase, whereupon the cells were harvested by centrifuga- tion. Protein induction was performed by incubating the cells in SD/-ura media containing galactose for 30°C for 24 h. Yeast cells were harvested and then resuspended in 0.1 M Tris- HCl, pH 7.5 containing 1 mM EDTA and 14 mM 2-mercaptoethanol. Cells were lysed using a French Press set to 1000–1500 psi. Cell lysates were then centrifuged at 10 000 g to remove cell debris. The supernatant was subjected to ultracentrifugation at 100 000 g for microsome preparation. Microsomes were resuspended in the same buffer as above and quantified by the Bradford assay (Brad- ford, 1976). Glycinol production and enzyme assay Production of glycinol was performed according to Boue et al. (2009) with some modification. Glycine max cv. Harosoy63 seeds were surface-sterilized with 70% EtOH and soaked in water for 4– 5 h. Soaked seeds were halved and treated with 10 mM AgNO3. Samples were incubated at 25°C in the dark for 48 h (Figure S1). Total isoflavonoid extraction was performed in methanol, and separated using an Agilent HPLC 1100 attached to 1260 analytical scale fraction collector with a semi-preparative Gemini� C18 110 �A column (150 9 4.6 mm, 5 lm; Phenomenex, http://www.phenome nex.com/). The injection volume was 100 ll at a flow rate of 5 ml min�1. The mobile phase (A) consisted of water and mobile phase (B) consisted of acetonitrile. The gradient system was as follows: 0–2 min held at 10% B; 2–7.5 min increasing to 20% B; 7.5–17 min increasing to 40% B; 17–17.5 min increasing to 100% B; 17.5–19.5 min 100% B; 19.5–20 min returning to 10% B; 20–22.5 min 10% B prior to the next injection. The column was maintained at 35°C. Spectra were detected at 250, 257 and 280 nm. The glycinol peak was collected, and its identity was confirmed by accurate mass LC-MS/MS (Thermo� Q-Exactive Orbitrap) and NMR spectroscopy. For GmPT enzyme assay, 80 lg of recombinant yeast microso- mal protein was incubated with 400 lM (�)-glycinol, 400 lM DMAPP and 10 mM MgCl2 in a total volume of 250 ll at 30°C for 2 h, followed by an ethyl acetate extraction, and analysis by an Agi- lent 1100 HPLC system equipped with a multi-wavelength diode array detector using a Poroshell 120, EC-C18 column (4.6 9 100 mm, 2.7 lm; Agilent). The injection volume was 10 ll at a flow rate of 1 ml min�1. Mobile phase (A) consisted of 0.1% formic acid in water and mobile phase (B) consisted of 0.1% formic acid in ace- tonitrile. Samples were run under the following gradient system: 0–2 min held at 10% B; 2–7.5 min increasing to 20% B; 7.5–17 min increasing to 40% B; 17–17.5 min 100% B; 17.5–19.5 min increasing to 100% B; 19.5–20 min returning to 10% B, 20–22.5 min 10% B prior to the next injection. Product was isolated following the method used for glycinol isolation and was confirmed with LC- MS/MS (Thermo� Q-Exactive Orbitrap) and NMR. All NMR data were collected at 25°C on a Varian INOVA 600-MHz spectrometer equipped with a pulsed field gradient triple resonance probe and referenced using the residual protons in the acetone-d6 solvent (2.05 pmm) at the Biomolecular Nuclear Magnetic Resonance Facility, London Regional Proteomics Centre, Western University. ACKNOWLEDGEMENTS The authors thank Alex Molnar, Arun Kumaran Anguraj Vadivel, Gang Tian in AAFC-London, and Liliana Santamaria-Kisiel in Wes- tern University for technical assistance. This research was sup- ported by the Natural Sciences and Engineering Research Council of Canada’s Discovery Grant 385922-2011 RGPIN, and Agriculture and Agri-Food Canada’s Genomics Research and Development Ini- tiative Grant J-000151 to SD. CONFLICT OF INTEREST The authors declare no competing financial interests. SUPPORTING INFORMATION Additional Supporting Information may be found in the online ver- sion of this article. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 979 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense https://www.thermofisher.com/ca/en/home/brands/invitrogen.html https://www.thermofisher.com/ca/en/home/brands/invitrogen.html https://www.takarabio.com/ http://www.phenomenex.com/ http://www.phenomenex.com/ Figure S1. Subcellular localization of truncated candidate GmPTs. Translational fusion of truncated GmPTs was transiently expressed in Nicotiana benthamiana as described in (Figure 6), and visualized using confocal microscopy. Figure S2. Production of glycinol in soybean seeds using AgNO3- treatment. Figure S3. Isolation of glycinol from AgNO3-treated soybean seeds. Table S1. Pairwise coding DNA and amino acid sequence compar- ison (%) of isoflavonoid-specific candidate prenyltransferase gene family in soybean. Table S2. Chemical structure and comparison of chemical shifts of GmPT01 reaction product with G2DT and G4DT. Table S3. List of genes within QTL Phytoph 13-3 with their anno- tated function. Table S4. List of primers used in the study for subcellular localiza- tion, qPCR and protein expression. REFERENCES Ahuja, I., Kissen, R. and Bones, A.M. (2012) Phytoalexins in defense against pathogens. Trends Plant Sci. 17, 73–90. Akashi, T., Aoki, T. and Ayabe, S.I. (1998) Identification of a cytochrome P450 cDNA encoding (2S)-flavanone 2-hydroxylase of licorice (Gly- cyrrhiza echinata L.; Fabaceae) which represents licodione synthase and flavone synthase II. FEBS Lett. 431, 287–290. Akashi, T., Sasaki, K., Aoki, T., Ayabe, S.-I. and Yazaki, K. (2009) Molecular cloning and characterization of a cDNA for pterocarpan 4-dimethylallyl- transferase catalyzing the key prenylation step in the biosynthesis of glyceollin, a soybean phytoalexin. Plant Physiol. 149, 683–693. Ayers, R.A., J€urgen, E., Finelli, R., Berger, N. and Peter, A. (1976) Host-patho- gen interactions: IX. Quantitative assays of elicitor activity and characteri- zation of the elicitor present in the extracellular medium of cultures of Phytophthora megasperma var. sojae. Plant Physiol. 57, 751–759. Bhattacharyya, M.K. and Ward, E.W.B. (1986) Resistance, susceptibility and accumulation of glyceollins I–III in soybean organs inoculated with Phy- tophthora megasperma f. sp. glycinea. Physiol. Mol. Plant Pathol. 29, 227–237. Bohrer, A.S., Yoshimoto, N., Sekiguchi, A., Rykulski, N., Saito, K. and Taka- hashi, H. (2015) Alternative translational initiation of ATP sulfurylase underlying dual localization of sulfate assimilation pathways in plastids and cytosol in Arabidopsis thaliana. Front. Plant Sci. 5, 750. Boue, S.M., Tilghman, S.L., Elliott, S. et al. (2009) Identification of the potent phytoestrogen glycinol in elicited soybean (glycine max). Endocrinology, 150, 2446–2453. Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Cheng, J., Yuan, C. and Graham, T.L. (2011) Potential defense-related preny- lated isoflavones in lactofen-induced soybean. Phytochemistry, 72, 875– 881. Dastmalchi, M. and Dhaubhadel, S. (2015) Soybean chalcone isomerase: evolution of the fold, and the differential expression and localization of the gene family. Planta, 241, 507–523. Dastmalchi, M., Bernards, M.A. and Dhaubhadel, S. (2016) Twin anchors of the soybean isoflavonoid metabolon: evidence for tethering of the com- plex to the endoplasmic reticulum by IFS and C4H. Plant J. 85, 689–706. Dhaubhadel, S., McGarvey, B.D., Williams, R. and Gijzen, M. (2003) Isoflavo- noid biosynthesis and accumulation in developing soybean seeds. Plant Mol. Biol. 53, 733–743. Dorrance, A.E., McClure, S.A. and St. Martin, S.K. (2003) Effect of partial resistance on Phytophthora stem rot incidence and yield of soybean in Ohio. Plant Dis. 87, 308–312. Dupuis, J. and Siegmund, D. (1999) Statistical methods for mapping quanti- tative trait loci from a dense set of markers. Genetics, 151, 373–386. Emanuelsson, O., Nielsen, H. and Von Heijne, G. (1999) ChloroP, a neural network-based method for predicting chloroplast transit peptides and their cleavage sites. Protein Sci. 8, 978–984. Fischer, D., Ebenau-Jehle, C. and Grisebach, H. (1990a) Phytoalexin synthe- sis in soybean: purification and characterization of NADPH:20-hydroxy- daidzein oxidoreductase from elicitor-challenged soybean cell cultures. Arch. Biochem. Biophys. 276, 390–395. Fischer, D., Ebenau-Jehle, C. and Grisebach, H. (1990b) Purification and characterization of pterocarpan synthase from elicitor-challenged soy- bean cell cultures. Phytochemistry, 29, 2879–2882. Graham, T.L., Kim, J.E. and Graham, M.Y. (1990) Role of constitutive isofla- vone conjugates in the accumulation of glyceollin in Soybean infected with phytophthora-megasperma. Mol. Plant-Microbe Interact. 3, 157–166. Grant, D., Nelson, R.T., Cannon, S.B. and Shoemaker, R.C. (2010) SoyBase, the USDA-ARS soybean genetics and genomics database. Nucleic Acids Res. 38, D843–D846. Hahn, M.G., Bonhoff, A. and Grisebach, H. (1985) Quantitative localization of the phytoalexin glyceollin in relation to fungal hyphae in soybean roots infected with Phytophthora megasperma f. sp. glycinea. Plant Physiol. 77, 591–601. Hanada, K., Zou, C., Lehti-Shiu, M.D., Shinozaki, K. and Shiu, S.H. (2008) Importance of lineage-specific expansion of plant tandem duplicates in the adaptive response to environmental stimuli. Plant Physiol. 148, 993– 1003. Hanada, K., Kuromori, T., Myoga, F., Toyoda, T., Li, W.-H. and Shinozaki, K. (2009) Evolutionary persistence of functional compensation by duplicate genes in Arabidopsis. Genome Biol. Evol. 1, 409–414. Horton, P., Park, K.J., Obayashi, T., Fujita, N., Harada, H., Adams-Collier, C.J. and Nakai, K. (2007) WoLF PSORT: protein localization predictor. Nucleic Acids Res. 35, W585–W587. Jung, W., Yu, O., Lau, S.M.C., O’Keefe, D.P., Odell, J., Fader, G. and McGo- nigle, B. (2000) Identification and expression of isoflavone synthase, the key enzyme for biosynthesis of isoflavones in legumes. Nat. Biotechnol. 18, 559. Karamat, F., Olry, A., Munakata, R., Koeduka, T., Sugiyama, A., Paris, C., Hehn, A., Bourgaud, F. and Yazaki, K. (2014) A coumarin-specific prenyl- transferase catalyzes the crucial biosynthetic reaction for furanocoumarin formation in parsley. Plant J. 77, 627–638. Kaufmann, M.J. and Gerdemann, J.W. (1958) Root and stem rot of soybean caused by Phytophthora sojae n. sp. Phytopathology, 48, 201–208. Kimpel, J.A. and Kosuge, T. (1985) Metabolic regulation during glyceollin biosynthesis in green soybean hypocotyls. Plant Physiol. 77, 1–7. Kochs, G. and Grisebach, H. (1986) Enzymic synthesis of isoflavones. Eur. J. Biochem. 155, 311–318. Krogh, A., Larsson, B., Von Heijne, G. and Sonnhammer, E.L.L. (2001) Pre- dicting transmembrane protein topology with a hidden Markov model: application to complete genomes. J. Mol. Biol. 305, 567–580. Kube�s, J., T�umov�a, L., Martin, J., Vildov�a, A., Hendrychov�a, H. and Sojkov�a, K. (2014) The production of isoflavonoids in Genista tinctoria L. cell sus- pension culture after abiotic stressors treatment. Nat. Prod. Res. 28, 2253–2263. Kumar, S., Stecher, G. and Tamura, K. (2016) MEGA7: molecular evolution- ary genetics analysis version 7.0 for bigger datasets. Mol. Biol. Evol. 33, 1870–1874. Li, W. (2016) Bringing bioactive compounds into membranes: the UbiA superfamily of intramembrane aromatic prenyltransferases. Trends Bio- chem. Sci. 41, 356–370. Liang, P.-H., Ko, T.-P. and Wang, A.H.J. (2002) Structure, mechanism and function of prenyltransferases. Eur. J. Biochem. 269, 3339–3354. Lozovaya, V.V., Lygin, A.V., Zernova, O.V., Li, S., Hartman, G.L. and Wid- holm, J.M. (2004) Isoflavonoid accumulation in soybean hairy roots upon treatment with Fusarium solani. Plant Physiol. Biochem. 42, 671–679. Lygin, A.V., Hill, C.B., Zernova, O.V., Crull, L., Widholm, J.M., Hartman, G.L. and Lozovaya, V.V. (2010) Response of soybean pathogens to glyceollin. Phytopathology, 100, 897–903. Ma, Y., Wang, J., Zhong, Y., Geng, F., Cramer, G.R. and Cheng, Z.-M. (2015) Subfunctionalization of cation/proton antiporter 1 genes in grapevine in response to salt stress in different organs. Hortic. Res. 2, 15 031. Moy, P., Qutob, D., Chapman, P., Atkinson, I. and Gijzen, M. (2004) Patterns of gene expression upon infection of soybean plants by Phytophthora sojae. Mol. Plant-Microbe Interact. 17, 1051–1062. Ohara, K., Muroya, A., Fukushima, N. and Yazaki, K. (2009) Functional char- acterization of LePGT1, a membrane-bound prenyltransferase involved in the geranylation of p-hydroxybenzoic acid. Biochem. J. 421, 231–241. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 980 Arjun Sukumaran et al. 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense Qin, J., Song, Q., Shi, A., Li, S., Zhang, M. and Zhang, B. (2017) Genome- wide association mapping of resistance to Phytophthora sojae in a soy- bean [Glycine max (L.) Merr.] germplasm panel from maturity groups IV and V. PLoS One, 12, e0184613. Sahoo, D.K., Abeysekara, N.S., Cianzio, S.R., Robertson, A.E. and Bhat- tacharyya, M.K. (2017) A novel Phytophthora sojae resistance Rps12 gene mapped to a genomic region that contains several Rps genes. PLoS One, 12, e0169950. Saleh, O., Haagen, Y., Seeger, K. and Heide, L. (2009) Prenyl transfer to aro- matic substrates in the biosynthesis of aminocoumarins, meroterpenoids and phenazines: the ABBA prenyltransferase family. Phytochemistry, 70, 1728–1738. Sasaki, K., Mito, K., Ohara, K., Yamamoto, H. and Yazaki, K. (2008) Cloning and characterization of naringenin 8-prenyltransferase, a flavonoid-specific prenyltransferase of Sophora flavescens. Plant Physiol. 146, 1075–1084. Schenck, C.A., Chen, S., Siehl, D.L. and Maeda, H.A. (2015) Non-plastidic, tyrosine-insensitive prephenate dehydrogenases from legumes. Nat. Chem. Biol. 11, 52–57. Schmutz, J., Cannon, S.B., Schlueter, J. et al. (2010) Genome sequence of the palaeopolyploid soybean. Nature, 463, 178–183. Schopfer, C.R., Kochs, G., Lottspeich, F. and Ebel, J. (1998) Molecular characterization and functional expression of dihydroxyptero- carpan 6a-hydroxylase, an enzyme specific for pterocarpanoid phytoalexin biosynthesis in soybean (Glycine max L.). FEBS Lett. 432, 182–186. Sepiol, C.J., Yu, J. and Dhaubhadel, S. (2017) Genome-wide identification of chalcone reductase gene family in soybean: insight into root-specific GmCHRs and Phytophthora sojae resistance. Front. Plant Sci. 8, 2073. Shen, G., Huhman, D., Lei, Z., Snyder, J., Sumner, L.W. and Dixon, R.A. (2012) Characterization of an isoflavonoid-specific prenyltransferase from Lupinus albus. Plant Physiology, 159, 70–80. Shrestha, S.D., Chapman, P., Zhang, Y. and Gijzen, M. (2016) Strain specific factors control effector gene silencing in Phytophthora sojae. PLoS One, 11, e0150530. Simons, R., Vincken, J.-P., Roidos, N., Bovee, T.F.H., van Iersel, M., Ver- bruggen, M.A. and Gruppen, H. (2011) Increasing soy isoflavonoid con- tent and diversity by simultaneous malting and challenging by a fungus to modulate estrogenicity. J. Agric. Food Chem. 59, 6748–6758. St€ossel, P. (1982) Glyceollin production in soybean. J. Phytopathol. 105, 109–119. Takahashi, S. and Koyama, T. (2006) Structure and function of cis-prenyl chain elongating enzymes. Chem. Rec. 6, 194–205. Tuteja, J.H. and Vodkin, L.Q. (2008) Structural features of the endogenous CHS silencing and target loci in the soybean genome. Crop Sci. 48, S49– S68. Tuteja, J.H., Clough, S.J., Chan, W.C. and Vodkin, L.O. (2004) Tissue-specific gene silencing mediated by a naturally occurring chalcone synthase gene cluster in Glycine max. Plant Cell, 16, 819–835. Vega-S�anchez, M.E., Redinbaugh, M.G., Costanzo, S. and Dorrance, A.E. (2005) Spatial and temporal expression analysis of defense-related genes in soybean cultivars with different levels of partial resistance to Phytoph- thora sojae. Physiol. Mol. Plant Pathol. 66, 175–182. Wang, C.S. and Vodkin, L.O. (1994) Extraction of RNA from tissues contain- ing high levels of procyanidins that bind RNA. Plant Mol. Biol. Rep. 12, 132–145. Wang, H., Wijeratne, A., Wijeratne, S., Lee, S., Taylor, C.G., St Martin, S.K., McHale, L. and Dorrance, A.E. (2012) Dissection of two soybean QTL con- ferring partial resistance to Phytophthora sojae through sequence and gene expression analysis. BMC Genomics, 13, 428. Ward, E., Lazarovits, G., Unwin, C. and Buzzell, R. (1979) Hypocotyl reac- tions and glyceollin in soybeans inoculated with zoospores of Phytoph- thora megasperma var. sojae. Phytopathology, 69, 951–955. Yoneyama, K., Akashi, T. and Aoki, T. (2016) Molecular characterization of soybean pterocarpan 2-dimethylallyltransferase in glyceollin biosynthe- sis: local gene and whole-genome duplications of prenyltransferase genes led to the structural diversity of soybean prenylated isoflavonoids. Plant Cell Physiol. 57, 2497–2509. © 2018 Her Majesty the Queen in Right of Canada. The Plant Journal © 2018 John Wiley & Sons Ltd and Society for Experimental Biology., The Plant Journal, (2018), 96, 966–981 GmPT01 and P. sojae resistance 981 1365313x, 2018, 5, D ow nloaded from https://onlinelibrary.w iley.com /doi/10.1111/tpj.14083 by C ochrane C anada Provision, W iley O nline L ibrary on [20/04/2023]. See the T erm s and C onditions (https://onlinelibrary.w iley.com /term s-and-conditions) on W iley O nline L ibrary for rules of use; O A articles are governed by the applicable C reative C om m ons L icense