Exocyst components promote an incompatible interaction between Glycine max (soybean) and Heterodera glycines (the soybean cyst nematode)

Vesicle and target membrane fusion involves tethering, docking and fusion. The GTPase SECRETORY4 (SEC4) positions the exocyst complex during vesicle membrane tethering, facilitating docking and fusion. Glycine max (soybean) Sec4 functions in the root during its defense against the parasitic nematode Heterodera glycines as it attempts to develop a multinucleate nurse cell (syncytium) serving to nourish the nematode over its 30-day life cycle. Results indicate that other tethering proteins are also important for defense. The G. max exocyst is encoded by 61 genes: 5 EXOC1 (Sec3), 2 EXOC2 (Sec5), 5 EXOC3 (Sec6), 2 EXOC4 (Sec8), 2 EXOC5 (Sec10) 6 EXOC6 (Sec15), 31 EXOC7 (Exo70) and 8 EXOC8 (Exo84) genes. At least one member of each gene family is expressed within the syncytium during the defense response. Syncytium-expressed exocyst genes function in defense while some are under transcriptional regulation by mitogen-activated protein kinases (MAPKs). The exocyst component EXOC7-H4-1 is not expressed within the syncytium but functions in defense and is under MAPK regulation. The tethering stage of vesicle transport has been demonstrated to play an important role in defense in the G. max-H. glycines pathosystem, with some of the spatially and temporally regulated exocyst components under transcriptional control by MAPKs.

Each exocyst component plays an important role in secretion. Mutants of S. cerevisiae Sec3 (sec3), the primary exocyst subunit that connects vesicles with the target membrane, exhibit secretory vesicle accumulation in the cytoplasm 48,49 because vesicles are unable to tether with the target membrane 48,49 . N. benthamiana Sec5 (EXOC2) is important for secretion of the pathogenesis-related 1 (PR-1) protein and callose deposition in the process of defense against Phytophthora infestans 18 . N. benthamiana Sec5 is targeted by the P. infestans effector AVR1 RXLR, which impairs PR-1 secretion and callose deposition 18 . Notably, G. max PR-1 (Glyma.15G062400) functions in defense against H. glycines 50 . G. max PR-1 is also under regulation by mitogen-activated protein kinases (MAPKs) 50 . The work of Austin et al. 51 led to the identification of a number of G. max callose synthases (CSs) expressed within the syncytium during the defense process with functions in defense. The results demonstrated that a secreted G. max protein (PR-1) and an enzyme (CS) that generates a secreted defense molecule (callose) function during the defense response against H. glycines 51 . Therefore, G. max PR-1 and CS act in a manner that is very similar to their function in N. benthamiana during PTI 11 . In another recent work, some pathogen effectors were shown to impair the function of the exocyst structure through ubiquitination of an exocyst protein component (Exo70B1) in a manner resembling ETS 52 . Due to the importance of each exocyst component, experiments have shown that the removal of just one protein impairs the ability of the other components to function properly [52][53][54] , resulting in the impairment of biological processes [52][53][54] .
Exocyst proteins are coiled-coil proteins that share some structural homology with helical bundles 41,55 . Helical bundles facilitate exocyst component interactions, which are essential for complex formation 41,55 . The structure of the exocyst complex is rod-shaped, with N-and C-termini located at opposite poles of the structure. This structure aids in the tethering of vesicles to the plasma membrane and delivery of vesicle cargo to the apoplast 38,40,44,46,[56][57][58] . The exocyst functions by connecting vesicles through the EXOC5 and EXOC6 proteins to the plasma membrane through EXOC1 and EXOC7 38,44,49,59,60 . On the target (plasma) membrane is phosphatidylinositol 4,5-biphosphate (PI(4,5)P2), to which EXOC1 and EXOC7 bind [61][62][63] . In S. cerevisiae, the movement of vesicles is regulated by vesicle membrane-bound Sec4p, which directs the vesicle to the plasma membrane at a targeted site [64][65][66][67] . Through its interaction with EXOC6, Sec4p functions by regulating assembly of the exocyst 38,39 . These results support observations showing that G. max Sec4 functions in facilitating the defense response to H. glycines 37 .
The experiments presented here have identified the components of the G. max exocyst. At least one exocyst component of each gene family is expressed within the syncytium during the defense response of G. max against H. glycines parasitism. In some cases, these exocyst genes are under regulation by MAPKs. Experimental overexpression of exocyst genes in the H. glycines-susceptible cultivar G. max [Williams 82/PI 518671] suppresses parasitism. In contrast, experimental decreases in the expression of exocyst components through RNAi in the H. glycinesresistant cultivar G. max [Peking/Pi 548402] facilitate parasitism. The combination of suppressing H. glycines parasitism in a normally susceptible G. max cultivar and facilitating H. glycines parasitism in a normally resistant G. max cultivar successfully demonstrated the functions of the target genes in defense. These results demonstrate that the G. max exocyst plays an important role in defense against H. glycines parasitism. Furthermore, these results show the importance of the plant secretion process to defense in general.

exocyst genes were expressed within H. glycines-parasitized root cells during defense.
The observation that G. max Sec4 functions in defense against H. glycines parasitism implies a similar role for the exocyst 37 . The most recently released G. max genome annotation (Wm82.a2.v1) was examined through BLAST searches using A. thaliana exocyst component protein sequences as a query. The analysis resulted in the identification of 5 EXOC1 genes, 2 EXOC2 genes, 5 EXOC3 genes, 2 EXOC4 genes, 2 EXOC5 genes, 6 EXOC6 genes, 31 EXOC7 genes and 8 EXOC8 genes (Supplementary Table S1). These gene accessions served as the basis for subsequent analyses.
Here, the H. glycines life cycle guided the design of gene expression experiments ( Fig. 1) 31 employing LM to isolate RNA from targeted cells. The targeted cells are involved in successful parasitism by H. glycines during a susceptible reaction and the defense response by G. max during a resistant reaction. The collected cells included pericycle cells and surrounding cells and were collected at 0 dpi. Furthermore, syncytia were collected at an early stage of parasitism (3 dpi). Syncytia formed in susceptible or resistant reactions at 3 dpi showed a similar cytological appearance. Their features included hypertrophy, the enlargement of nuclei, the development of dense cytoplasm and an increase in the endoplasmic reticulum (ER) and ribosome content. As a consequence of these similarities, 6 dpi was selected as a time point. The 6-dpi time point assisted in differentiating between a susceptible and resistant reaction. By 6 dpi, the syncytia formed during a susceptible reaction were characterized by the hypertrophy of nuclei and nucleoli, proliferation of cytoplasmic organelles, a reduction in vacuoles, the  31 . The cDNA probe made from mRNA 31 was tagged with a proprietary Affymetrix label and used for gene expression studies, leading to the identification of a pool of 1,787 candidate defense genes ( Fig. 1) 31 . This analysis was focused on examining the relationship between G. max and its defense response against H. glycines parasitism in relation to the exocyst (Fig. 1). From these data, exocyst genes expressed within the syncytium were identified (Fig. 2). Under our analytical parameters, the exocyst genes exhibited four profiles of expression in relation to their defense response to H. glycines. However, other gene expression profiles not observed here are possible. First, 11 exocyst genes were not expressed at any time point: EXOC1-3, EXOC6-3, EXOC8-3, EXOC8-5, EXOC8-8, EXOC7-A1-1, EXOC7-B1-1, EXOC7-B1-3, EXOC7-E2-1, EXOC7-F1-2 and EXOC7-H4-1 (Fig. 2). Second, 8 exocyst genes lacked measurable gene expression at 0 dpi (control) but were expressed at the 6-dpi time point: EXOC1-1, EXOC3-5, EXOC4-1, EXOC6-1, EXOC6-6, EXOC7-D1-2, EXOC7-E1-1 and EXOC7-G1-4 (Fig. 2). The third group was composed of 4 exocyst genes expressed at only 3 and 6 dpi: EXOC5-2, EXOC8-4, EXOC7-B1-2 and EXOC7-H7-1 (Fig. 2). The fourth group of 4 exocyst genes was expressed at all three time points (0, 3 and 6 dpi): EXOC2-1, EXOC7-A1-3, EXOC7-D1-1 and EXOC7-F1-1 (Fig. 2). Consequently, the results revealed that 16 different exocyst genes were expressed in samples from at least one of the time points chosen for the analysis. Furthermore, the analyses identified a component of each exocyst gene family expressed within cells parasitized by H. glycines as the root cell underwent a defense response. Last, among these 16 exocyst genes, four were expressed at the 0-dpi time point, 8 were expressed by the 3-dpi time point, and 16 were expressed by the 6-dpi time point. These results demonstrate an increase in the number of exocyst genes expressed during the course of the defense process. Notably, expression of a number of exocyst genes in root cells could not be evaluated (n/a) due to the nature of the original root cell gene expression analyses (Supplementary Table S2) 31 . While not studied in the functional analyses presented in "Functional analysis of  ; these genes were identified in syncytia isolated from roots infected at 3 and 6 days postinfection (dpi) but not pericycle control cells. These genes are the focus of the study presented here.

Scientific RepoRtS
| (2020) 10:15003 | https://doi.org/10.1038/s41598-020-72126-z www.nature.com/scientificreports/ syncytium-expressing exocyst genes demonstrates a defense role" section, the expression of those exocyst genes was examined in transcriptomic analyses of defense MAPKs, the results of which are presented in "The expression of certain exocyst genes was induced by specific defense MAPKs" section 50 . An additional examination determined whether the exocyst genes are regulated by signaling processes in the G. max defense against H. glycines and may also be of interest here 50 .   50 . RNA from these defense MAPK-OE and RNAi transgenic lines was subjected to RNA-seq analyses 68 , which led to the identification of thousands of transcripts whose relative abundances either increased or decreased 68 . Consequently, the G. max exocyst gene family as a whole was examined here via transcriptomic analyses of those MAPK-OE and MAPK-RNAi lines. Analyses of those RNA-seq data was conducted to determine if the syncytium-expressed exocyst genes was also expressed within individual MAPK-OE or MAPK-RNAi lines 50 . Further, the analysis also determined whether the exocyst genes were expressed across many lines overexpressing the defense MAPKs 50 .
The results demonstrated that the differential expression of exocyst genes was primarily found in specific transgenic MAPK lines (Supplementary Table S2). For many exocyst genes, differential expression was not observed at all (Supplementary Table S2). However, even if an exocyst gene was not differentially expressed (NDE), this does not mean that the gene is not expressed at all (lacking identified sequences in the RNA-seq studies). Among the exocyst genes inducing MAPK overexpression, EXOC7-H4-1 and EXOC7-H7-1 exhibited higher relative transcript levels in all 9 defense MAPK-OE lines. However, when their expression within the syncytium was examined, EXOC7-H4-1 lacked expression in the 0-dpi control samples as well as the 3-dpi and 6-dpi samples from syncytia during the defense response (Fig. 2). While EXOC7-E2-1 was also not expressed in the 0-dpi control samples or 3-dpi or 6-dpi syncytium samples, was expressed at almost the same level as EXOC7-H4-1in the defense MAPK-OE lines. However, EXOC7-E2-1 gene expression was observed in just 8 of the 9 defense MAPK-OE lines. EXOC7-E2-1 was not further examined via qRT-PCR or functional studies since it was not expressed in the syncytium or in all 9 of the defense MAPK-OE lines. In contrast, EXOC7-H7-1 was expressed in the 3-dpi and 6-dpi samples from syncytia during the defense response but not in the 0-dpi samples (Fig. 2). Expression of the exocyst genes that showed expression in the parasitized root cells as well as some of the transgenic MAPK-OE or MAPK-RNAi lines was confirmed in the transgenic MAPK-OE and MAPK-RNAi lines. Their RNA-seq expression data were confirmed by qRT-PCR using the RPS21 gene as a control. These 4 exocyst genes were EXOC1-1, EXOC7-B1-1, EXOC7-D1-1 and EXOC7-G1-4 ( Fig. 3). During the course of the analysis, a single exocyst gene (EXOC7-H4-1) was found not to be expressed within the syncytium. However, the increased expression of EXOC7-H4-1 in all 9 of the MAPK-OE lines was confirmed by qRT-PCR using the RPS21 gene as a control (Fig. 3). The observation that EXOC7-H4-1 expression was not measured within the syncytium but was increased in all 9 of the defense MAPK-OE lines (MAPK-all-OE) is notable. This result indicates that processes involving the G. max secretion apparatus outside the vicinity of the syncytium are important to defense. This hypothesis was examined later in the analysis.

Transgenic plants showed the expected effect on EXOC gene expression.
Affymetrix DCM microarray analysis showed that exocyst genes were expressed within parasitized root cells (syncytia) during the defense response. These 16 exocyst genes were EXOC1-1, EXOC3-5, EXCO4-1, EXOC6-1, EXOC6-6, EXOC7-G1-4, EXOC7-D1-2, EXOC7-E1-1, EXOC5-2, EXOC8-4, EXOC7-B1-2, EXOC7-H7-1, EXOC2-1, EXOC7-A1-3, EXOC7-D1-1 and EXOC7-F1-1. The 16 exocyst genes were cloned and functionally tested through transgenic analyses. Functional transgenic tests of the exocyst genes were conducted to determine if they play a role in defense. The genes were overexpressed in the H. glycines-susceptible G. max [Williams 82/PI 518671] cultivar, after which whether the H. glycines-susceptible cultivar became resistant to parasitism was determined (Fig. 4). In contrast, the same genes were engineered as RNAi cassettes used to decrease their expression in the H. glycinesresistant G. max [Peking/PI 548402] cultivar. These functional transgenic tests were conducted to determine whether decreased exocyst gene expression would result in H. glycines susceptibility (Fig. 4). A combination of two outcomes had to be met for a gene to meet our criteria for a defense role 33  www.nature.com/scientificreports/ expected, qRT-PCR showed that the relative transcript abundance of the exocyst components was increased in the overexpression lines and decreased in the RNAi lines in comparison to that in the RPS21 gene-expressing control (Fig. 6). The effect of expression of the transgene cassette on root mass was then analyzed. The results demonstrated that expression of the cassettes did not have a statistically significant effect on root mass when the data were compared to those in the respective controls (p < 0.05) (Fig. 7). However, effects on only root mass were considered in this analysis.
Functional analysis of syncytium-expressing exocyst genes demonstrates a defense role. The effect of altered exocyst gene expression on H. glycines parasitism was tested. The genetically engineered G. max roots were infected with H. glycines [NL1-Rhg/HG-type 7/race 3] as described in the Materials and Methods section ("Assaying the effect the genetic engineering events on nematode parasitism" section). The data from the experimental replicates were compared to those from the corresponding controls in the overexpression (pRAP15-ccdB control) and RNAi (pRAP17-ccdB control) studies.
In the first set of analyses, engineering of the pRAP15-ccdB vector in H. glycines-susceptible G. max [Williams 82/PI 518671] produced a robust level of infection. Quantification of the level of infection showed a cyst count of 203.09 ± 5.04 cysts per wr system and 49.03 ± 5.07 cysts pg of root system. All exocyst-OE transgenic lines were compared to this standard run in triplicate (please refer to Materials and Methods section "Assaying the effect the genetic engineering events on nematode parasitism" section for details of the analysis.). In contrast, engineering of the pRAP17-ccdB control vector in H. glycines-resistant G. max [Peking/PI 458402] strongly suppressed parasitism. Quantification of the level of infection showed 9.94 ± 1.3 cysts per wr system and 2.52 ± 0.43 cysts pg of root system. All exocyst-RNAi transgenic lines were compared to this standard run in triplicate (please refer to Materials and Methods section "Assaying the effect the genetic engineering events on nematode parasitism" section for details of the analysis.). In cases in which a numerical value of differential expression was obtained in the RNA-seq analyses (not necessarily ± 1.5-fold) and the p value was < 0.05 (p < 0.05), qRT-PCR was employed to confirm the relative change in transcript abundance of the exocyst gene of interest. These results were compared to the RPS21 control employing the 2 −ΔΔCT method 31,69,70 . For a change in relative transcript abundance to be considered statistically significant, a minimum cutoff of ± 1.5-fold was set, and p < 0.05. The p values for the replicate qRT-PCR analyses were calculated through Student's t test 70 . Error bars represent the standard deviation. Please refer to the Materials and Methods ("Functional analysis of a MAPK-induced exocyst genes that were not expressed in the syncytium" section) for details of the analysis.
Scientific RepoRtS | (2020) 10:15003 | https://doi.org/10.1038/s41598-020-72126-z www.nature.com/scientificreports/ The second set of analyses focused on syncytium-expressing exocyst components. Calculation of the FI showed that H. glycines parasitism was significantly reduced by 58-68% in roots overexpressing each of the exocyst genes in cysts per wr system and by 50-64% in cysts pg of root system (both p values < 0.001) (Fig. 8) Functional analysis of a MAPK-induced exocyst genes that were not expressed in the syncytium. The third set of analyses focused on one exocyst component (EXOC7-H4-1) that was not expressed at any time point within the syncytium during the defense response or in control cells. However, EXOC7-H4-1 expression was confirmed to be increased in all 9 of the defense MAPK-OE lines by qRT-PCR (Fig. 9). These results indicate that aspects of plant secretion important for the defense process may occur outside of the parasitized root cells or their progenitors. This observation may explain why higher levels of suppressed H. glycines parasitism were not seen in prior experiments focusing on syncytium-expressing genes. Alternatively, overexpression of defense MAPKs may synthetically induce the expression of EXOC7-H4-1. In this case, EXOC7-H4-1 may or may not function in defense at all. To determine whether EXOC7-H4-1 functions in defense, it was cloned and used in overexpression and RNAi experiments. Transgenic EXOC7-H4-1-OE and EXOC7-H4-1-RNAi lines were assessed by qRT-PCR analyses with RPS21 used as a control, confirming their expected expression (Fig. 10). Analysis of the effect of altered EXOC7-H4-1 transgene cassette expression on root mass was performed, which demonstrated that the overexpression of EXOC7-H4-1 and RNAi cassette expression did not affect root mass (p > 0.05) (Fig. 11). In replicated functional analyses employing the same controls used in the previous study, EXOC7-H4-1-OE lines showed H. glycines parasitism that was significantly decreased by 58.8% in the wr and 58.3% pg of root system, as shown by the FI (p value < 0.001) (Fig. 12). In contrast, H. glycines  . qRT-PCR confirmation of the OE of syncytium-expressed exocyst genes and RNAi of these genes through gene cassettes in transgenic roots. The results were compared to those in the RPS21 control employing the 2 −ΔΔCT method 31,69,70 . A minimum cutoff of ± 1.5-fold was set, and p < 0.05. *Statistically significant. The p values for the replicate qRT-PCR analyses were calculated through Student's t test 70 . Error bars represent the standard deviation. Please refer to the Materials and Methods ("Functional analysis of a MAPK-induced exocyst genes that were not expressed in the syncytium" section) for details of the analysis. www.nature.com/scientificreports/ parasitism in the EXOC7-H4-1-RNAi lines was significantly increased by 3.23-fold in the wr and 4.21-fold pg of root system, as shown by the FI (p value < 0.001) (Fig. 13).

Discussion
An analysis of G. max exocyst components is presented here. This study examined whether exocyst components play a role during the defense response of G. max to the parasitic nematode H. glycines. The analysis began by the identification G. max exocyst genes from the most recent Wm82.a2.v1 genome. Then, exocyst genes expressed within the pericycle and surrounding cells prior to H. glycines infection (0 dpi) were determined. Follow-up studies then determined which exocyst genes are expressed within the syncytium during the defense response to H. glycines infection. The first of the time points selected for analysis was 3 dpi, at which point several cytological    In cases in which a numeric value indicating differential expression was obtained by the RNA-seq analyses and the p value was < 0.05 (p < 0.05), qRT-PCR is employed to confirm the relative change in transcript abundance. The results were compared to those in the RPS21 control employing the 2 −ΔΔCT method 31,50,69 . For a change in relative transcript abundance to be considered statistically significant, a minimum cutoff of ± 1.5-fold was set, and p < 0.05. *Statistically significant.
The p values for the replicate qRT-PCR analyses were calculated through Student's t test 70 . Error bars represent the standard deviation. Please refer to the Materials and Methods ("Functional analysis of a MAPK-induced exocyst genes that were not expressed in the syncytium" section) for details of the analysis.    www.nature.com/scientificreports/ features did not differ between the susceptible and resistant cultivars. The second of the time points selected for analysis was 6 dpi, at which point the cytological features between the susceptible and resistant cultivars differed, characterizing each reaction. Complimentary analyses were conducted to identify whether the expression of any of the exocyst genes is under regulation by MAPKs. This analysis is undertaken because studies have demonstrated the importance of MAPKs to the defense response of G. max against H. glycines 50 . These genes are components of both ETI and PTI 50 . The subsequent functional, transgenic studies presented here demonstrated that exocyst genes function in defense. The experiments also identified an exocyst gene that is not expressed within the syncytium during the process of defense but functions in defense. Consequently, the experiments indicated that there processes important to defense occur both locally within the syncytium and outside of the syncytium. This study began with the identification of all G. max exocyst genes through BLAST searches using the default parameters in Phytozome with A. thaliana exocyst protein sequences used as queries 72 . The analysis resulted in the identification of 61 genes that span the 8 exocyst gene families. These results are consistent with the composition of the exocyst in all eukaryotes, including plants 38,[40][41][42][43][44]73 . The G. max exocyst genes include 5 EXOC1 genes, 2 EXOC2 genes, 5 EXOC3 genes, EXOC4 genes, 2 EXOC5 genes, 6 EXOC6 genes, 31 EXOC7 genes and 8 EXOC8 genes. Consequently, each gene family contains multiple gene copies, which is consistent with the duplicated nature of the G. max genome 74 . An analysis of the nature of these gene duplication events is beyond the scope of this study. However, Cvrčková et al. 73 performed phylogenetic analyses of 10 different plant exocyst gene families and obtained important insights into the plant exocyst. The results showed that the small EXOC1, EXOC2, EXOC3, EXOC4 and EXOC5 gene families were likely amplified independently, late in the diversification of each plant lineage 73 . Furthermore, the small EXOC6 and EXOC8 gene families were likely amplified from a single ancestral gene 73 . In contrast, the very large EXOC7 gene family likely arose from early amplification of an ancestral gene in a common ancestor of land plants 73 . In each case, gene amplification leads to the diversification of paralog functions, which require further study. Therefore, the G. max EXOC7 gene family is notably expansive and consists of 31 members. The large size of the G. max EXOC7 gene family is consistent with observations in other land plants 73 .
G. max exocyst proteins show homology to those in A. thaliana ranging from a low identity of 42.61% (EXOC7-E2-2) to a high identity of 89.15% (EXOC3-5). However, two outliers in the EXOC8 gene family show a low identity of 60.96% (EXOC8-4) to a high identity of 75.30% (EXOC8-3). These outliers are EXOC8-6 (25.69% identity) and EXOC8-7 (25.66% identity). It is possible that these proteins are not EXOC8 homologs, and although they were included here, further study is required.
The main objective of this study was to understand whether the G. max exocyst plays a role in defense in the root. Toward that goal, exocyst gene accessions from the most recent G. max genome assembly (Wm82.a2.v1) were used for the identification of exocyst genes with Affymetrix GeneChip Soybean Genome Array probe set identifiers. This analysis was used to identify whether any of the exocyst genes exhibit expression specifically within the H. glycines-induced syncytium during the defense process. To the best of our knowledge, this is the only study to sue single-cell transcriptomic analyses from a multicellular organism to specifically identify exocyst genes. The analyses resulted in the identification of 27 of the 61 exocyst genes (44.26%) with an Affymetrix GeneChip Soybean Genome Array probe set identifier. Consequently, it was possible to identify whether any exocyst genes are expressed in the syncytium during the process of defense.
Within that list of 27 exocyst genes, 4 were expressed in the control population of root cells (pericycle and surrounding cells) that were sampled prior to H. glycines infection (0 dpi). After H. glycines infection, exocyst transcriptomic measurements were performed to detect the expression of 8 genes at 3 dpi during the defense response. Furthermore, 16 exocyst genes were identified as expressed at the 6-dpi time point during the defense response. The results showed an increase in the number of exocyst genes expressed during the course of H. glycines parasitism. This expression occurred specifically in G. max root cells during defense reactions (syncytium). Increases in the expression of all defense genes comprising a large gene family during the course of the defense response were seen in studies of the G. max-H. glycines pathosystem 34 . This study 34 examined a family of 22 β-glucosidases predicted to have signal peptides and identified the secreted PEN2 homolog α-hydroxynitrile glucosidase (βg-4), which functions in G. max during its defense response against H. glycines 34 .
The results presented here indicate that the relative transcript abundance of these exocyst components may increase as a consequence of the G. max defense response. Furthermore, diverse cellular processes involving the G. max exocyst appear to lead to a successful defense response. For example, two different EXOC6 genes (EXOC6-1, EXOC6-6) and 8 different EXOC7 genes (EXOC7-A1-3, EXOC7-B1-2, EXOC7-D1-1, EXOC7-D1-2, EXOC7-E1-1, EXOC7-F1-1, EXOC7-G1-4 and EXOC7-H7-1) were seen here to function in the defense process (please refer to "Functional analysis of syncytium-expressing exocyst genes demonstrates a defense role" section: functional analysis of syncytium-expressed exocyst genes). The results indicated that different types of vesicles containing different cargos may function at specific times or in specific ways during the defense process.
Prior analyses have demonstrated the importance of the MAPK signaling platform to the G. max defense process against H. glycines 50 . MAPKs are part of a central, three-tier signal transduction platform shared by all eukaryotes. MAPKs permit cells to transduce signals into meaningful output for a variety of physiological and developmental purposes [75][76][77] . Recently, an analysis of the entire 32-member G. max MAPK gene family as it relates to defense against H. glycines was conducted 50 . The analysis led to the identification of 9 MAPKs that function in defense 50  www.nature.com/scientificreports/ The examination of syncytium-expressing exocyst genes shown here revealed that a number of these genes were differentially expressed in 1 or more transgenic MAPK lines. For example, EXOC7-B1-1 (an Exo70B1 homolog) had a lower relative transcript abundance in the MAPK2-RNAi line, in which the G. max defense response against H. glycines was suppressed 50 . These results are consistent with observations made in A. thaliana showing that some pathogen effectors impair Exo70B1 protein function through ubiquitination, leading to ETS 52 . In contrast, the relative transcript abundance of EXOC7-D1-1 was shown here to be higher in the MAPK3-2-OE line than in the corresponding control line. Furthermore, EXOC7-H7-1, which was expressed in the syncytium at the 3-dpi and 6-dpi time points, showed higher relative transcript abundance in all of the defense MAPK-OE lines (denoted MAPK-all-OE). The similar expression profile of EXOC7-H4-1 was shown here through its examination in functional experiments. However, EXOC7-H4-1 was not expressed in syncytial cells analyzed in the Affymetrix DCM study 31 . These results indicate that secretion processes outside of the vicinity of the syncytium are important for defense. If signaling processes outside of the syncytium function in defense, systemic processes such as systemic acquired resistance (SAR) may occur during the G. max defense process against H. glycines. G. max genes known to function in SAR, including the transcription factor NONEXPRESSOR of PR1 (NPR1), the lipase ENHANCED DISEASE SUSCEPTIBILITY 1 (EDS1) and the coiled-coil nucleotide-binding leucine rich repeat (CC-NB-LRR) resistance (R) protein NONRACE SPECIFIC DISEASE RESISTANCE1 (NDR1), have all been shown to function during the defense against H. glycines parasitism 33,50,[78][79][80][81][82][83][84] . The observation that NDR1 functions in the G. max defense process against H. glycines is particularly noteworthy since NDR1 is required for ETI. In A. thaliana, NDR1 binds three R proteins: the CC-NB-LRR protein RESISTANCE TO PSEUDOMONAS SYRINGAE PV MACULICOLA1 (RPM1); RPM1-interacting 4 (RIN4); and the NB adaptor shared by APAF-1, certain R gene products and the CED-4 (ARC)-LRR (NB-ARC-LRR) gene RESISTANT TO P. SYRINGAE 2 (RPS2) 79,80,82,[85][86][87][88][89][90][91][92] . NDR1 induces MAPK gene expression in the G. max-H. glycines pathosystem 50 .
Exocyst proteins functioning upstream of vesicle docking act to deliver callose to infection sites formed by pathogens 18,19 . This docking process employs the 20 S particle, which incorporates syntaxin (SYP)-containing SNARE. In A. thaliana, SYP121 and callose are delivered to defense sites in plants during resistance to Botrytis graminis f. sp. hordei by the ADP ribosylation factor (ARF)-GTP exchange factor GNOM 93 . Consistent with those observations, our prior analyses showed that overexpression of G. max syntaxin 121 resulted in increased callose deposition surrounding the H. glycines syncytium during the defense response 34 . In contrast, RNAi of G. max syntaxin 121 decreased callose deposition 34 . Subsequent follow-up studies identified the expression of 4 different CS genes in the syncytium during the defense process against H. glycines 51 . Furthermore, overexpression of the different CS genes resulted in a decrease in H. glycines parasitism, while RNAi of those same genes increased nematode parasitism 51 . These results provide evidence of processes requiring the exocyst to function in cells that are beyond the boundary of the syncytium during the defense response. A. thaliana Exo70H4 plays a role in callose deposition in trichomes, consistent with our observations and hypothesis 94 . The synthetic defense processes due to defense MAPK overexpression, as it relates to EXOC7-H4-1, are also under examination. Both qRT-PCR and RNA-seq analyses of controls that did not overexpress defense MAPKs detected EXOC7-H4-1 expression. These results provide evidence that EXOC7-H4-1 expression occurs at least in uninfected tissues that lack syncytia during the defense process.
In contrast to these results, the expression of a number of exocyst genes could not be analyzed by DCM because probe sets for these genes were not included in the Affymetrix GeneChip Soybean Genome Array. However, analysis of their expression was carried out through the use of data made available by a transgenic MAPK RNA-seq study 68 . The analysis identified a number of exocysts of genes that were differentially expressed in one or more transgenic MAPK-OE lines. For example, the EXOC7-E2-1 relative transcript abundance was higher in 8 of the 9 transgenic defense MAPK-OE lines. These G. max exocyst genes, however, were beyond the scope of the functional analysis presented here because their syncytium expression was not seen (M or NM) by the analytical methods used, referred to as n/a. These genes will be the focus of future analyses of the G. max exocyst.
Infection of genetically engineered exocyst-OE lines with H. glycines resulted in a 58-68% decrease in cysts per wr system and a 50-64% decrease in cysts pg of root system, depending on the exocyst component under study. Consequently, the G. max overexpression lines exhibited higher susceptibility to H. glycines. In contrast, RNAi studies revealed an increase in H. glycines parasitism of 3.16-3.77-fold in the wr system and a 4.13-5.68fold increase in the cysts pg of root system when compared to those in the control. Consequently, the RNAi lines showed an increased susceptibility to H. glycines. These results are consistent with observations that syncytiumexpressing genes function in the process of defense [32][33][34]37,50,51,83,84 . Furthermore, these results are consistent with observations showing that the vesicle transport apparatus that functions in vesicle docking and membrane fusion also functions in defense [32][33][34]37,50,51,83,84 .
Many of the genes under study in relation to G. max defense against H. glycines showed expression within the syncytium. However, the experiments did not rule out whether gene expression outside of the syncytium is important to the defense process. A recent study examining the entire G. max MAPK gene family demonstrated that 9 out of 9 MAPKs lacking expression within the syncytium have no role in defense 50 . These results argue against the functions of non-syncytium-expressed genes in defense. In contrast, 7 out of 12 syncytium-expressed MAPKs function in defense 50 . These results demonstrate that most of the MAPKs that function in defense are expressed within the syncytium. Affymetrix probe sets for the two other MAPKs that function in defense were lacking from the GeneChip Soybean Genome Array, so the syncytium expression of these MAPKs could not be determined 50 . Results demonstrating a defense role for genes expressed outside of the vicinity of the syncytium have yet to be determined. The results presented here reveal that the expression of certain exocyst genes (EXOC7-H4-1) occurs outside of the boundary of the parasitized root cells. In A. thaliana, the vesicle transport machinery involving the exocyst acts at some level to facilitate callose deposition 18,19,52,85,[95][96][97][98][99][100][101][102] . Systemic processes outside of the vicinity of infection as well as structural modifications of G. max roots, including callose deposition and cell wall modification, have been observed 33,34,51,103 . These results are consistent with the observation that these Scientific RepoRtS | (2020) 10:15003 | https://doi.org/10.1038/s41598-020-72126-z www.nature.com/scientificreports/ components are coregulated and/or part of a feedback loop that further facilitates the expression of genes that function in the defense process 33,34 .
Over the past few years, a model of how the process of defense occurs in G. max as it reacts to H. glycines parasitism including the vesicle transport apparatus has been proposed [31][32][33][34]37,50,51 . This model has its origins in the demonstration that the 20 S component α-SNAP is specifically expressed within the syncytium during the defense process 31 . A number of studies have expanded on this theme, including analyses of the α-SNAP-binding protein syntaxin 31 and the other 20 S particle components, including SNAP-25, synaptobrevin, synaptotagmin, NSF and MUNC18 34 . These studies demonstrate the importance of the docking and membrane fusion steps in defense 34 . These vesicle transport steps are preceded by a tethering process performed by thee exocyst that is essential for membrane fusion to occur (Fig. 12). Early analyses indicated the importance of vesicle tethering during the G. max defense response to H. glycines 37 . Furthermore, and earlier analysis demonstrated that G. max Sec4, a protein that is known to function in tethering by binding EXOC6, plays a defensive role against H. glycines parasitism 37 . Related experiments have demonstrated that the mechanism by which vesicles are delivered to the cell periphery, which functions through myosin XI, is also important during defense 51 . These results all point toward the function of the exocyst at a crucial point in the delivery of vesicles to the site of membrane fusion and plant secretion in the defense process of G. max against H. glycines parasitism. Furthermore, it appears possible that specific exocyst genes, which are likely the products of duplication events, may function in specific ways to increase the breadth of the defense response or general health of the plant 73 . With regard to root-organism interactions, the conserved nature of the exocyst indicates that its function is not limited to the G. max-H. glycines pathosystem, indicating the broad importance of this study.

Methods
Candidate gene selection. The Phytozome portal (https ://phyto zome.jgi.doe.gov) houses the G. max genome sequence and information about its assembly and annotation, making acquisition of the protein sequences of the entire exocyst gene family possible 72 . G. max exocyst protein accessions were identified based on comparisons to A. thaliana protein sequences in Phytozome using the Basic Local Alignment Search Tool (BLAST) with the default settings 72 . These default settings were as follows: target type: proteome; program: BLASTP (protein query to protein database); expect (E) threshold: − 1; comparison matrix: BLOSUM62; word (W) length: default = 3; number of alignments to show: 100 allowing for gaps and filter query.
Identification and selection of the G. max exocyst defense genes for use in functional transgenic studies were carried out by using the gene expression data from Matsye et al. 31 . These data were obtained through microarray analyses using the GeneChip Soybean Genome Array (Affymetrix). In the study, Matsye et al. 31 31,104 . For our purposes, a gene was considered to be measured [M] when the probe signal was detectable above a threshold (p < 0.05) on all 6 arrays (the 3 arrays each from G. max [Peking/PI 548402] and G. max [PI 88788] ) 31,104 . For the analysis presented here, the expression of an exocyst gene was considered not measured (NM) if the probe signal was not detected at a statistically significant level (p ≥ 0.05) on any of the 6 arrays. For some genes, no corresponding probe set was fabricated onto the microarray. In these cases, gene expression was not determined and considered not applicable (n/a). For this part of the analysis, the Affymetrix annotations were mapped to the original G. max genome release (Wm82.a1.v1.1) since only that annotation was available at the time of the analysis 31 . Here, these older annotations are compared to the updated, most recent G. max Wm82.a2.v1 genome assembly and annotation. RNA sequencing. RNA sequencing (RNA-seq) data were obtained from the experiments of McNeece et al. 50 and Alshehri et al. 68 . These studies examined the G. max MAPK gene family as it relates to the defense of G. max against H. glycines parasitism and showed that 9 of the 32 G. max MAPKs function during the defense reaction against H. glycines parasitism, naming these MAPKs defense MAPKs 50 . Single replicate RNA-seq experiments examining the 9 defense MAPKs were performed using RNA isolated from transgenic lines in which the targeted MAPK genes were either overexpressed (OE) or inhibited via RNAi 50,68 . The defense MAPKs examined were MAPK2 (Glyma.06G029700), MAPK3-1 (Glyma.U021800), MAPK 3-2 (Glyma.12G073000), MAPK 4-1 (Glyma.07G066800), MAPK 5-3 (Glyma.08G017400), MAPK6-2 (Glyma.02G138800), MAPK 13-1 (Glyma.12G073700), MAPK16-4 (Glyma07g38510) and MAPK20-2 (Glyma.14G028100), and pRAP15-ccdB and pRAP17-ccdB served as corresponding controls 50,68 . RNA was isolated from the 9 defense MAPK-OE and MAPK-RNAi lines and their respective controls, and the RNA sequences were deposited and made publicly available 68 . For the experimental purposes presented here, an additional goal was the identification of exocyst genes whose expression was induced or suppressed by the different studied MAPKs. Expression of the exocyst genes that met the differential expression criterion in the RNA-seq experiments (± 1.5-fold change in expression, p < 0.05) was confirmed by quantitative real-time PCR (qRT-PCR) as described in the Materials and Methods

Scientific RepoRtS
| (2020) 10:15003 | https://doi.org/10.1038/s41598-020-72126-z www.nature.com/scientificreports/ section ("cDNA synthesis" and "qRT-PCR assessment of gene expression" sections). Expression of the remaining exocyst genes that did not meet the differential expression criterion in the RNA-seq analyses (NDE) was not confirmed by qRT-PCR. The G. max genome accessions were used to mine exocyst RNA-seq gene expression data from the defense MAPK RNA-seq study and are shown in the analysis 50,68 . The G. max accession numbers whose RNA-seq data are presented were derived from the most recent Glycine max Wm82.a2.v1 annotation. The exocyst accession numbers were further manually confirmed with Phytozome to confirm their accuracy (as of February 15, 2020) 72 Table S3). Candidate G. max exocyst defense gene amplicons were synthesized by PCR using the AccuPrime Taq Polymerase System (Invitrogen) according to the manufacturer's instructions with an Eppendorf AG Mastercycler Pro S model 6,325 PCR gradient PCR thermocycler. The reaction conditions were dependent on the nucleotide composition of the amplicon and PCR primer. In general, DNA melting was carried out at 95 °C for 2 min, followed by another 30-s melt at 95 °C. Primer annealing conditions were empirically determined through gradient PCR for 30 s. Primer extension was carried out at 68 °C for 1 min per 1,000 base pairs of the sequence. This process was carried out for 35 cycles, followed by a final step at 68 °C for 10 min, with the reaction completed at 4 °C. The PCR product was run on a 1% agarose gel. The amplicons were removed from the gel and purified using the Wizard SV Gel and PCR Clean-Up System (Promega) according to the manufacturer's instructions. Subsequently, the amplicon was ligated into the pENTR/D-TOPO entry vector using the pENTR/D-TOPO Cloning Kit (Invitrogen) according to the manufacturer's instructions. The reaction contents were transformed into One Shot TOP10 chemically competent E. coli (TOP 10) (Invitrogen) cells according to the manufacturer's instructions as described. Cells were selected on Luria-Bertani (LB) agar plates containing 50 μg/ml kanamycin. Plasmid DNA was isolated from selected colonies using the Wizard Plus SV Minipreps DNA Purification System (Promega) according to the manufacturer's instructions. The DNA sequences were confirmed by Sanger sequencing. Subsequently, the exocyst amplicons were ligated to Gateway-compatible overexpression (pRAP15) or RNAi (pRAP17) destination vectors using Gateway LR Clonase Enzyme Mix (Invitrogen) according to their instructions to transfer the candidate G. max exocyst resistance gene amplicon into the respective destination vectors. Nonengineered pRAP15 and pRAP17 vectors served as experimental controls; these vectors contained the ccdB gene where the candidate G. max exocyst defense gene amplicon would otherwise be following directional insertion during the LR clonase reaction. Based on this feature, the nonengineered pRAP15-ccdB (overexpression control) and pRAP17-ccdB (RNAi control) vectors were suitable controls to account for any nonspecific effects of gene overexpression or RNAi 50 . The reaction contents were transformed into chemically competent E. coli TOP 10 cells according to the manufacturer's instructions as described. Cells were selected on LB agar plates containing 5 μg/ ml tetracycline. E. coli colonies containing the gene of interest (GOI) after transformation with pRAP15/17-GOI were grown in 3 ml of liquid LB medium and chemically selected with 5 μg/ml tetracycline overnight at 37 °C. Plasmid preps (Promega) of these liquid cultures were carried out according to the manufacturer's instructions. Gene-specific primers were used to confirm the presence of each exocyst gene (Supplementary Table S3). The pRAP15/17 destination vectors confirmed to have the exocyst gene amplicon were transformed into chemically competent Agrobacterium rhizogenes K599 (K599) 50 via freeze-thaw transformation 50 . In this procedure, 250 μl of K599 cells was thawed on ice. A sufficient amount of plasmid DNA (0.1-1 μg) was added to K599 cells and gently mixed. The mixture of K599 cells and plasmid DNA was incubated on ice for 5 min and then subsequently transferred to liquid N 2 for 5 min. The mixture was transferred to a 37 °C water bath for 5 min. The contents were then transferred to a culture tube with 1 ml of LB medium, placed in a shaking incubator, and incubated at 28 °C for 2 h. The cells were then collected by centrifugation for 2 min at 5,000 rpm, resuspended in 200 μl of LB medium and spread on LB agar plates containing 5 μg/ml tetracycline for chemical selection at 28 °C. After 2 days, K599 colonies that underwent genetic transformation were picked for confirmation of the DNA cassette with G. max exocyst gene-specific primers. Colonies harboring the appropriate plasmid were then grown in 250 ml of LB medium containing 5 μg/ml tetracycline at 28 °C in a shaking incubator 50 .

Production of transgenic plants for functional experiments. A solution of K599 cells transformed
with the appropriate vector construct was pelleted by centrifugation in a Sorvall RC6 Plus Superspeed Centrifuge at 4 °C for 20 min. The resulting pellet of K599 cells was resuspended in Murashige and Skoog medium containing vitamins (MS) (Duchefa Biochemie) and 3.0% sucrose at pH 5.7 (MS medium) 105 . Transgenic G. max production began when the root of each 1-week-old plant was cut off at the hypocotyl with a new, sterile razor blade that had been immersed in the K599 cell solution in a Petri dish. This procedure allowed the transformed K599 cells access to the wound made by removal of the root. Subsequently   50 . Roots exhibiting eGFP reporter expression also possessed the candidate defense gene expression cassette, and each had its own promoter and terminator sequences. Gene transfer occurred because K599 cells transported the DNA cassettes between the left and right borders of the pRAP15 and pRAP17 destination vectors into the root cell chromosomal DNA. The result was a stable transformation event in the root somatic cell, even though the DNA cassette had not been incorporated into the germline. Roots subsequently developed from the transgenic cell over a period of a few weeks. The resultant genetically mosaic plants had a nontransgenic shoot with a transgenic root system. Therefore, in the experiments presented here, each individual transgenic root system is an independent transformant line. The transgenic plants were each planted in a Ray Leach "Conetainer" (SC10) (Stuewe and Sons, Inc.) secured in a Ray Leach Tray (RL98) (Stuewe and Sons, Inc.) in sandy (93.00% sand, 5.75% silt, and 1.25% clay) soil and allowed to recover for two weeks prior to the start of the experiment 50 Table S3) 50 . Fold changes in gene expression caused by the genetic engineering events were calculated using the 2 −ΔΔCT method 50,69 . Student's t-test was used to calculate the p values for the replicate qRT-PCR experiments 70 . The procedures followed those presented by McNeece et al. 50 .
Assaying the effect the genetic engineering events on nematode parasitism. Infection of the transgenic plants with H. glycines was performed according to the procedures described by Sharma et al. 34 . H. glycines eggs were obtained from cysts collected from 60-day-old, greenhouse-grown G. max stock cultures. The cultures were maintained in 500-cm 3 polystyrene pots. Stock H. glycines cysts were purified by sucrose flotation 106 . G. max roots that contained H. glycines cysts were washed through nested sieves with pore sizes of 850 μm and 250 μm. The H. glycines cysts were collected from the 250-μm sieve after this procedure and ground with a mortar and pestle to release the eggs. The H. glycines eggs were obtained after gravitational sieving followed by sucrose centrifugation. The H. glycines eggs were recovered with a 75-μm sieve nested over a 25-μm sieve. H. glycines J2s were collected from hatched eggs in a modified Baermann funnel placed on a Slide Warmer (model 77) (Marshall Scientific) at 28 °C. H. glycines eggs hatched from days 4 to 7. H. glycines J2s were collected on a sieve with a 25-μm pore sire and placed in 1.5-ml tubes. The tube and its contents were centrifuged at 10,000 rpm for 1 min, washed with distilled sterile water and centrifuged again. The J2s were concentrated by centrifugation in an IEC clinical centrifuge for 30 s at 1,720 rpm to a final optimized concentration of 2,000 pi-J2/ml. Each root was inoculated with one ml of H. glycines at a concentration of 2,000 J2s/ml per root system (per plant). Infection was allowed to proceed for 30 days. At the end of the experiment, the cysts were collected over nested 20-and 100-mesh sieves 34 . Furthermore, the soil was washed several times, and the rinse water was sieved to assure collection of all cysts for enumeration of the female index (FI) 34 . The FI, the community-accepted standard representation of the obtained data 24 , was calculated by a procedure originally described by Golden et al. 24 as follows and employed for functional transgenic experiments 50 : FI = (Nx/ Ns) × 100. In the procedure employed here, Nx was the pRAP-exocyst gene-transformed (experimental) line. Ns was the pRAP-ccdB (control) line 34 . The FI was calculated as the number of cysts per whole root (wr) system grown within 100 cc of soil and the number of cysts per gram (pg) of root system 50 . Historically, analysis by wr system has been the method of choice for data presentation 24 . Analysis of the number of cysts pg of root system, however, accounts for possible changes in root growth caused by the influence of the overexpression or RNAi of the candidate G. max exocyst defense gene. Three biological replicates consisting of 10-20 individual transgenic plants each were made for each construct. The results were statistically examined using the Mann-Whitney-Wilcoxon (MWW) rank-sum test, a nonparametric test of the null hypothesis that does not require the assumption of a normal distribution, with a cutoff of p < 0.05 50,71 .

Data availability
Data relevant to the study is presented here as supplemental data.