http://orcid.org/0000-0002-5639-9706
Figures
Abstract
Fusarium virguliforme is a soil borne root pathogen that causes sudden death syndrome (SDS) in soybean [Glycine max (L.) Merrill]. Once the fungus invades the root xylem tissues, the pathogen secretes toxins that cause chlorosis and necrosis in foliar tissues leading to defoliation, flower and pod drop and eventually death of plants. Resistance to F. virguliforme in soybean is partial and governed by over 80 quantitative trait loci (QTL). We have conducted genome-wide association study (GWAS) for a group of 254 plant introductions lines using a panel of approximately 30,000 SNPs and identified 19 single nucleotide polymorphic loci (SNPL) that are associated with 14 genomic regions encoding foliar SDS and eight SNPL associated with seven genomic regions for root rot resistance. Of the identified 27 SNPL, six SNPL for foliar SDS resistance and two SNPL for root rot resistance co-mapped to previously identified QTL for SDS resistance. This study identified 13 SNPL associated with eight novel genomic regions containing foliar SDS resistance genes and six SNPL with five novel regions for root-rot resistance. This study identified five genes carrying nonsynonymous mutations: (i) three of which mapped to previously identified QTL for foliar SDS resistance and (ii) two mapped to two novel regions containing root rot resistance genes. Of the three genes mapped to QTL for foliar SDS resistance genes, two encode LRR-receptors and third one encodes a novel protein with unknown function. Of the two genes governing root rot resistance, Glyma.01g222900.1 encodes a soybean-specific LEA protein and Glyma.10g058700.1 encodes a heparan-alpha-glucosaminide N-acetyltransferase. In the LEA protein, a conserved serine residue was substituted with asparagine; and in the heparan-alpha-glucosaminide N-acetyltransferase, a conserved histidine residue was substituted with an arginine residue. Such changes are expected to alter functions of these two proteins regulated through phosphorylation. The five genes with nonsynonymous mutations could be considered candidate SDS resistance genes and should be suitable molecular markers for breeding SDS resistance in soybean. The study also reports desirable plant introduction lines and novel genomic regions for enhancing SDS resistance in soybean.
Citation: Swaminathan S, Das A, Assefa T, Knight JM, Da Silva AF, Carvalho JPS, et al. (2019) Genome wide association study identifies novel single nucleotide polymorphic loci and candidate genes involved in soybean sudden death syndrome resistance. PLoS ONE 14(2): e0212071. https://doi.org/10.1371/journal.pone.0212071
Editor: Mark Gijzen, Agriculture and Agri-Food Canada, CANADA
Received: November 9, 2018; Accepted: January 25, 2019; Published: February 26, 2019
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: National Institute of Food and Agriculture (NIFA), United States Department of Agriculture Grant no. 2013-68004-20374 awarded to Madan K. Bhattacharyya, Xiaoqiu Huang, Leonor F. Leandro and Silvia R. Cianzio.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Soybean [Glycine max (L.) Merrill] oil and protein meal accounts for approximately 25% and 65% of the world consumption, respectively [1]. Sudden death syndrome (SDS) is the second most damaging soybean disease in the U.S. with an estimated soybean yield suppression valued to $0.6 billion in 2014 [2]. The disease has two components: (i) foliar SDS or leaf scorch and (ii) root rot. In North America as well as in South America, Fusarium virguliforme O’Donnell and T. Aoki (formerly F. solani (Mart.) Sacc. f. sp. glycines) causes SDS. It has only one mating type and is asexually propagated [3]. In South America, F. virguliforme and F. tucumaniae cause SDS. F. tucumaniae has both mating types and is sexually propagated [3]. In addition to these two major SDS pathogens, in South America F. crassistipitatum and F. brasiliense are also reported to cause SDS in soybean [4–6].
The F. virguliforme causes root rot and root necrosis [7]. Upon penetration through the cortex into the root xylem, the pathogen secretes toxins that cause chlorotic and necrotic interveinal foliar symptoms [7–11]. The name ‘sudden death syndrome’ derives from the observations that following infection and toxin translocation, leaves of normal-appearing plants suddenly develop interveinal chlorosis and necrosis and plants die prematurely [12–14].
Management options for controlling SDS are limited [15, 16]. Growing of SDS resistant varieties is the most effective method of protecting the crop from this fungal pathogen [17, 18]. Unfortunately, SDS resistance is partial and encoded by a large number genes, each contributing small effect [17, 19–31]. Furthermore, epistatic interactions among SDS resistance QTL result in even a more complex inheritance for the SDS resistance [32, 33].
Most of the SDS resistance QTL have been mapped using the progenies of bi-parental crosses [17, 19–32]. In recent years, genome wide association studies (GWAS) have identified 30 single nucleotide polymorphic loci (SNPL) linked to foliar SDS resistance [33–35]. Some of the SNPL were co-mapped to SDS resistance QTL that were previously identified through studies of bi-parental crosses [31, 33–35]. The population structures used in the GWAS have been different. Chang et al. (2016) [34] used a group of ancestral lines for mapping the SDS resistance loci. Zhang et al. (2015) [33] used a set of PI lines; and Wen et al. (2014) [35] a group of advanced experimental lines and released cultivars to conduct GWAS in identifying SNPL associated with SDS resistance.
The majority of the QTL for SDS resistance or SNPL identified are for foliar SDS. F. virguliforme is a root pathogen and causes damages to the infected roots [36]. Limited studies have identified soybean genotypes that carry both foliar SDS and root rot resistance [15, 32, 37–39]. Therefore, more soybean germplasm with both foliar SDS and root rot resistance may be identified for providing robust protection against F. virguliforme. The objectives of the study were therefore to identify (i) genetic materials with novel sources of foliar SDS and root rot resistance genes, and (ii) genomic regions and genes involved in both foliar SDS and root rot resistance. The phenotypic evaluation of 254 preselected PI lines [40, 41] for both foliar SDS and root rot was conducted under growth-chamber conditions using a mixture of F. virguliforme Mont-1 isolate and two highly aggressive isolates collected in Iowa. We have identified ten PI lines that are resistant to both foliar SDS and root rot symptoms and should be suitable resources for breeding soybean cultivars with robust SDS resistance. GWAS revealed 13 SNPL associated with eight novel genomic regions containing foliar SDS resistance genes and six SNPL with five novel regions for root-rot resistance. We have identified five candidate genes for foliar SDS and root rot resistances that can be potential molecular markers for introgressing SDS resistance into soybean cultivars.
Materials and methods
Plant material
The 254 PI lines studied here were selected from a collection of over 6,000 PI lines based on their preliminary responses to F. virguliforme Mont-1 isolate in greenhouse, recorded by Hartman lab, University of Illinois, Urbana [40, 41]. Seeds of the 254 accessions were kindly provided by the USDA-ARS-GRIN National Genetic Resources Collection lab (S1 Table). The maturity group (MG) of the accessions ranged from MG 000 to MG X.
F. virguliforme isolates
In this research the responses of the PI lines to a mixture of three F. virguliforme isolates, Mont-1, Scott F2I11a and Clinton 1B were studied in growth chambers located at the Department of Agronomy, Iowa State University. The highly aggressive F. virguliforme isolates, Clinton-1B and Scott-F2I11a, were collected from Clinton and Scott Counties, Iowa, respectively [42, 43]. The single-spore derived isolates Clinton-1B (LL0059) and Scott F2I11a (LL0063) are being stored and maintained at the Leandro lab. The F. virguliforme Mont-1 isolate was collected from Monticello, Illinois. The F. virguliforme Mont-1 (FSG1) isolate has been widely used by the soybean research community as an aggressive isolate and its genome has been sequenced [4, 15, 40, 44–48].
Inoculum preparation
The methods of inoculum preparation and infection were described earlier [49]. Each isolate was grown on potato dextrose agar (PDA) (13 g Difco PDA/L) amended with antibiotics (0.150 g/L of streptomycin sulfate, and 0.15 g/L of chlortetracycline hydrochloride) for 6 weeks at room temperature (19–23°C) under dark conditions.
Sterile white sorghum [Sorghum bicolor (L.) Moench] seeds were soaked in distilled water for 24 h in quart mason jars (500 g seeds/jar). After the water was drained off, the sorghum filled jars were autoclaved twice, 1 h in each time. Sterile sorghum kernels of each jar were inoculated with ten mycelial plugs (7 mm in diameter) of the individual F. virguliforme isolate. Jars were incubated at room temperature (21 ± 2° C) for one month under continuous fluorescent light (40 W) and shaken daily for 1 to 2 min to ensure uniform fungal growth. After a month, the infested sorghum kernels were placed on a tray under a fume hood for 24 h to dry. Then the inocula were stored at 4° C and used in all five experiments to avoid any variation resulting from the individual batches of inocula.
The DNA concentration of the F. virguliforme isolate in individual inoculum was determined by qPCR. The DNA concentration ranged from 25 to 30 ng /mg of infested dried sorghum kernels [49]. The three inocula prepared for three F. virguliforme isolates were combined in equal proportions based on the DNA concentration of individual isolate. The thoroughly mixed inoculum prepared from three isolates was then added to pasteurized soil and sand (1:2) mix at a proportion of 1:20 :: inoculum:soil and sand mix. The inoculum was mixed thoroughly with the soil and sand mix by hand.
Phenotyping
The seeds of the soybean lines were planted in inoculum mixed soil-sand mix in Styrofoam cups (240 mL). Three seeds of individual soybean genotypes were placed on the surface of inoculum mixed soil-sand mix and additional 30 mL inoculum mixed soil:sand mixture was added to cover the seeds. The cups were randomized in a growth chamber. Cups were watered once daily. The seedlings were grown at 23 ± 1 °C (day time) and 16 ± 1 °C (night time) with 16 h light and 8 h dark periods. Light intensity was maintained at 300 μmol photons m−2s−1.
Five weeks after planting, foliar disease score of each plant was recorded using a modified scale (S1 Fig) of the one originally described by Hartman et al. (2004) [44]: 1, no foliar symptoms observed; 1.5, leaves showing a few chlorotic specks (1–5% foliage affected); 2, leaves showing slight yellowing and/or chlorotic flecks or blotches (6–10% foliage affected); 2.5, leaves showing big chlorotic flecks or blotches (11–20% foliage affected); 3, leaves showing interveinal chlorosis (21–30% foliage affected); 3.5, leaves showing interveinal chlorosis spread throughout the plant (31–40% foliage affected); 4, leaves start to fold (cupping of leaves), with slight necrosis (41–50% foliage affected); 4.5, leaves with necrosis along the >2 cm sectors along the leaf margin (51–60% foliage affected); 5, necrosis along the entire margin of the leaves (61–70% foliage affected); 5.5, heavy necrosis and cupping of leaves (71–80% foliage affected); 6, most of the leaf area necrotic and the leaves being heavily rolled and/or irregular in shape (81–90% foliage affected); 6.5, most of leaf area necrotic and dry (>90% foliage affected); 7, leaf drop resulting in defoliated plants (S1 Fig). On the basis of foliar disease scores (FDS), PIs were classified as highly resistant (HR; FDS <1.51); resistant (R; FDS 1.51–2.00); moderately resistant (MR; FDS 2.01–2.50); susceptible (S; FDS 2.51–3.00); or highly susceptible (HS; FDS >3.00) [44].
Thirty-seven days after planting, plants were carefully removed from the cups and roots were washed with warm tap water. Root rot (%) was evaluated as root areas showing dark brown to black discolorations, visually assessed on a percentage scale from 0 to 100 with an increment of 5% of the total root area (S2 Fig) [14].
Statistical analysis
The model for the statistical analyses for foliar SDS scores and root rot (%) data was,
where, Yij is the observed phenotype of the ith genotype in jth block or experiment, μ is the overall population mean, gi is the genetic effect of the ith genotype, bj is the effect of the jth block, and eij is the effect of experimental error. In each growth chamber experiment, three cups carrying a total of nine plants for each of the 254 PI lines and control lines were randomized. The mean disease scores or extent of root rot (%) were calculated from the plants of the three cups of each PI line. The experiment was repeated five times.
The mean phenotypic data of each genotype from each of the five experiments were subjected to analysis of variance (ANOVA) in a randomized block design by considering each experiment as a block, and tested for homogeneity of variances among experiments using the Levene’s test in the R package car [50]. Fisher’s protected least significant difference (LSD) test was used to compare means at p ≤ 0.05 using the R package agricolae [51]. Estimation of variance components for foliar SDS scores and root rot were determined using the R package. The histograms and fitted lines describing the distribution of foliar disease scores and extent of root rot (%) were generated using R software (package fBasics).
Broad sense heritability (H2) estimates for FDS and root rot (%) were estimated using the formula, H2 = σg2/(σg2 + σe2), where, σg2is the genotypic variance; σe2 is the error variance; and σg2 + σe2 is phenotypic variance (σp2).
Genotyping and quality control
The SNP dataset for the PIs was reported earlier by Song et al. (2013) [52], and it was downloaded from SoyBase (http://www.soybase.org/; 2018 version). Imputation of some data points was conducted using BEAGLE version 3.3.1. Of the total SNPs 42,168, accessible from SoyBase for the association panel, 59 SNPs were unanchored to the reference genome and were removed from further analysis. Individual SNP markers with missing rate larger than 10% were omitted from the analysis and only 31,506 SNPs were used in GWAS. The statistically significant SNPL were not originated from the imputed SNPL.
Genome-wide association study
The phenotypic data were analyzed using the best linear unbiased prediction (BLUP) program and the R lem4 package to reduce the effect of experimental variation [53]. GWAS was performed with the mixed linear model (MLM) using genome assessment and prediction integrated tools (GAPIT) in the R package [54, 55]. The most stringent approach between the false discovery rate (FDR) at p < 0.05 and empirical significant level at p < 0.001 was used to determine the threshold of significance for SNP-trait association, as described earlier [33]. A 1,000 permutations of genome-wide association was conducted to assess the empirical significant of SNPs [33]. SoyBase (ww.soybase.org; Glyma.Wm82.a2 Gmax2.0) was searched to find out additional information on SNPL and their associated candidate SDS resistance genes.
Results
Infection of PI lines with F. virguliforme isolates revealed novel PI lines with foliar SDS and root rot resistance
The 254 PI lines considered for this study were selected from a collection of over 6,000 PI lines based on their preliminary resistant responses to F. virguliforme Mont-1 isolate [40, 41]. The lines were evaluated for leaf and root responses against a mixture of three highly aggressive F. virguliforme isolates including Mont-1, Scott F2I11a and Clinton 1B, under growth chamber conditions. The lines were also evaluated for 31,506 single nucleotide polymorphism loci (SNPL). To study the structure of this sub-population of 254 PI lines, a neighbor joining (NJ) tree and principal component analysis (PCA) of the lines were conducted using Tassel 5.2.33 program on 31,506 single nucleotide polymorphisms (SNPs) distributed across the soybean genome [52]. The NJ tree of SNPs revealed six groups (S3 Fig). Both the NJ tree and PCA indicated a possible weak association between genotypes and geographic origin (S1 Table, S4A Fig). The PCA was also conducted for maturity groups (MGs) of the PI lines. No apparent association between MGs and genotypes of the PI lines was observed (S4B Fig). Similarly, foliar as well as root responses of the lines collected using a published protocol [49] to a mixture of three aggressive F. virguliforme isolates, Mont-1, Scott F2I11a and Clinton 1B were used to study the population structure. Both foliar and root responses of the 254 lines were randomly distributed across the population (S4C and S4D Fig). Although the foliar responses of the lines to F. virguliforme Mont-1 isolate was not normal (S5 Fig), the responses of the lines to the mixture of three F. virguliforme isolates including Mont-1 exhibited a normal distribution (p < 0.05) (Figs 1 and 2; S6 Fig). The Levene’s test of homogeneity of variance suggested that variation among the five experiments were homogenous for foliar as well as for root rot symptoms (p > 0.05). ANOVA revealed significant differences (p < 0.05) for both foliar SDS and root rot symptoms among the genotypes.
(A) The foliar disease symptoms were scored 35 days following infection with F. virguliforme. Arrows indicate the disease scores of the SDS resistant cultivars, MN1606 and Ripley, and susceptible cultivar, Spencer. Phenotypic evaluation was conducted five times, each with three replications. The values are means of fifteen biological replications. (B) Distribution of foliar SDS scores among the 254 PI that are lines presented in (A).
(A) The root rot symptoms were scored 37 days following infection with F. virguliforme. Arrows indicate the disease scores of the SDS resistant cultivars, MN1606 and Ripley, and susceptible cultivar, Spencer. Phenotypic evaluation was conducted five times, each with three replications. The values are means of fifteen biological replications. (B) Distribution of root rot symptoms among the 254 PI lines that are presented in (A).
Mean foliar SDS score of the resistant check MN1606 was 1.53, significantly different (p < 0.05), from that (4.5) of the susceptible check, Spencer (Fig 1). For root rot scores, the resistant check MN1606 exhibited 15% root rot, which is significantly (p < 0.01) less than that (34%) of the susceptible check, Spencer (Fig 2). Among the 254 PI lines, seven were highly foliar SDS resistant with foliar SDS scores of <1.5, while 71 were resistant with scores ranging from 1.50–2.00, and 61 PIs were highly susceptible with foliar SDS scores of >3.00 (S1 Table). Among the 254 PI lines, 23 lines showed <10% root rot and six showed >40% root rot (S1 Table).
A weak but significant association was observed between foliar SDS scores and root rot (r = 0.19; p < 0.01). The 25 most foliar SDS resistant PI lines are presented in Fig 3A to show that the root rot resistances of these lines are variable ranging from 5 to 56% root rot. Similarly, 23 PI lines with root rot scores less than 10% exhibited variation in the foliar SDS scores, which ranged from 1.26 to 3.42 (Fig 3B). We observed that 10 lines showed foliar disease scores < 2 and 10% or less root rot (S1 Table). These lines could be suitable for breeding SDS resistant soybean cultivars.
(A) Root responses of 25 most foliar SDS resistant PI lines to F. virguliforme. (B) Foliar responses of 25 most root rot resistant PI lines to F. virguliforme. The values are means and standard errors calculated from five independent experiments. The Scott-Knott statistical significance difference of the PI lines are provided in S1 Table.
Genome wide association study (GWAS) revealed novel SNPL and candidate genes for SDS resistance
The broad sense heritability values calculated were 0.90 and 0.77 for foliar SDS and root rot symptoms, respectively. The high heritability values suggest that the variation observed in the phenotypic data among the PI lines was less influenced by experimental variation and phenotypic data collected in the growth chamber were suitable for conducting GWAS to detect the genetic loci associated with the SDS resistance. GWAS was conducted using the mixed linear model (MLM) [54, 55]. The threshold of significance for SNP-trait association was determined by the levels of significance for false discovery rate (FDR) at p < 0.05 and empirical significant level at p < 0.001 as described in Materials and Methods [33].
GWAS identified 19 SNP loci (SNPL) that significantly associated with 14 genomic regions encoding foliar SDS resistance (Fig 4, Table 1) and eight SNPL associated with seven genomic regions for root rot resistance (Fig 5, Table 2). Of these 27 SNPL, six SNPL for foliar SDS resistance and two SNPL for root rot resistance co-mapped to previously identified QTL for SDS resistance (Fig 6). This study identified 13 SNPL associated with eight novel genomic regions containing foliar SDS resistance genes and six SNPL with five novel regions for root-rot resistance. Of the 27 SNPL identified, five generated nonsynonymous mutations in five genes: (i) three of which mapped to previously identified QTL for foliar SDS resistance and (ii) two mapped to two novel regions containing root rot resistance genes (Table 3).
The − log10 p-values from a genome-wide scan are plotted against the positions of each of the SNPs on 20 chromosomes. The horizontal blue line indicates the genome-wide significance threshold (FDR < 0.05). SNPs in Manhattan plot were placed in kb unit.
The − log10 p-values from a genome-wide scan are plotted against the positions of each of the SNPs on 20 chromosomes. The horizontal blue line indicates the genome-wide significance threshold (FDR < 0.05). SNPs in Manhattan plot were placed in kb unit.
Green checked box, foliar SDS SNPL; Pink checked box, root rot SNPL. SNPL are shown with numerical numbers (Tables 1 and 2). Previously identified foliar SDS and root rot resistance QTL are shown with solid green and pink boxes, respectively (S2 Table). The genetic map is generated in centi-Morgan (cM) unit.
Of the three genes containing nonsynonymous mutations and mapped to genomic regions containing previously reported foliar SDS resistance QTL, two encode LRR-receptor proteins and one encodes an unknown protein. Two nonsynonymous mutations (ss715613738, and ss715615487) resulted in Ser322Arg and Glu261Lys mutations in the two LRR-receptor proteins encoded by the Glyma.13g035700 and Glyma.13g079100 genes, respectively (Tables 1 and 3). The non-synonymous mutations were detected in conserved residues of the LRR- proteins (Table 3, S7 and S8 Figs). The nonsynonymous mutation (ss715618125) in Glyma.14g035000 gene encoding an unknown protein caused the Pro70Leu mutation (Tables 1 and 3).
Of the two nonsynonymous mutations localized to two genes associated with root rot resistance, the nonsynonymous mutation (ss715608261) in Glyma.10g058700 encoding a putative heparan-alpha-glucosaminide N-acetyltransferase enzyme resulted in the His201Arg mutation (Tables 2 and 3). The nonsynonymous mutation (ss715580538) in Glyma.01g222900 gene encoding a LEA hydroxyproline-rich glycoprotein resulted in the Ser14Asn mutation (Tables 2 and 3). In this study we also identified 22 additional SNPL, which were mapped to non-coding regions (intron, 3’ UTR and 5’ UTR) or intergenic regions (Tables 1 and 2).
Discussion
The genetics of resistance to SDS is highly complex. It was reported that a single dominant gene, Rfs controls SDS resistance in the soybean cultivar, Ripley under greenhouse conditions [29]. Subsequent studies revealed that SDS resistance is conditioned by additional QTL [17, 19–30]. Recently, it has been reported that SDS is controlled by over 80 QTL and epistasis gene interactions [31–33].
We evaluated a collection of 254 PI lines, pre-selected as putative foliar SDS resistant lines from a collection of over 6,000 PI lines, for foliar and root responses to a mixture of three F. virguliforme aggressive isolates under growth-chamber conditions. Earlier, preliminary study conducted in the Hartman Lab demonstrated that most of the 254 lines were highly resistant to F. virguliforme Mont-1 isolate (S5 Fig). Previous studies had showed that different soybean genotypes had varying levels of resistance to different F. virguliforme isolates [72, 73]. We therefore mixed three aggressive F. virguliforme isolates, Mont-1, Scott-F2I11a and Clinton-1B, to identify the most SDS resistant soybean lines.
In our study we identified seven highly foliar SDS resistant PIs with foliar disease scores <1.5 and 23 PIs with highly root rot resistant lines with <10% of the total root volume showing rotting. Only one PI line showed foliar SDS score of <1.5 and <10% root rot (S1 Table; Fig 3B). We however identified 10 lines that showed foliar disease scores < 2 and root rot ≤ 10% (S1 Table), which could be ideal resources for breeding SDS resistant lines.
We phenotyped the 254 lines in growth chambers. Disease phenotypes of individual lines were very consistent. We observed broad sense heritabilities of 0.90 and 0.77 were for responses of leaves and roots to the infection by the three isolate mixture. Furthermore, the leaf and root responses of the 254 lines to the F. virguliforme isolate mixture showed normal distributions (Figs 1 and 2; S6 Fig). Therefore, the phenotypic disease data were suitable for conducting GWAS to identify genetic loci for SDS resistance.
Over 20,000 PI lines including the 254 lines have been genotyped using the Infinium SoySNP50K BeadChip [52]. We used this published genotypic data with our phenotypic data to conduct the GWAS and identified 27 genetic loci (SNPL) for SDS resistance (Tables 1 and 2). Of the 27 SNPL, 19 govern foliar SDS resistance and eight root rot resistance (Tables 1 and 2). We observed that six of these SNPL for foliar SDS resistance and two SNPL for root rot resistance mapped to genomic regions that already have been reported to contain SDS resistance QTL [17, 22, 30, 32, 74–76] (Tables 1 and 2; Fig 6; S2 Table). This study identified 19 SNPL associated with eight novel genomic regions containing foliar SDS resistance genes and five novel regions for root-rot resistance, which will however require validation through genetic mapping studies.
Five of the 27 SNPL resulted changes in amino acid residues in five proteins (Table 3). We investigated if the nonsynonymous mutations affected any conserved amino acid residues that may be involved in gene functions. Comparison of the target proteins with their respective homologous protein sequences revealed that in each of the five proteins, the mutations affected the conserved amino acid residues that may lead to changes in either protein structure or post-translational modification, which could be important for expression of SDS resistance (Table 3, S7–S11 Figs). In three of the five proteins, the conserved amino acid residues, serine or histidine for phosphorylation, were altered by the mutations. Thus, altered functions for these three proteins could be possible leading to SDS susceptibility. The three proteins are: LRR–receptor protein, a LEA hydroxyproline-rich glycoprotein and a heparan-alpha-glucosaminide N-acetyltransferase. In an LRR-receptor and a novel protein with unknown function, lysine was mutated to glutamate and proline was mutated to leucine, respectively. These mutations can change the structures and functions of the two proteins. We however need to validate these genes for their role in SDS resistance through genetic mapping studies and/or study of the mutants generated for these candidate SDS resistance genes.
In our study, we identified two genomics regions to which more than one SNPL mapped. For example, SNPL ss715608333 and ss715608329 are only 26.8 kb apart and located in a 2 cM interval for SDS resistance. SNPL ss715617107, ss715617111, ss715617189 and ss715617218 are located in a 75.5 kb genomic region, one the average with one SNLP in every 25 kb region. We did not identify any non-significant SNPs in these two SNPL-rich regions indicating linkage disequilibrium in these regions associated with QTL for foliar SDS resistance.
In our study, we used only 31,506 SNPs for a genome of 1,150 Mb DNA carrying over 46,000 genes [77, 78]. If the 31,506 SNPs are randomly distributed, we expect to have only 1 SNP in every 36.5 kb soybean genome sequence. In that range, more than one gene is not unusual in the soybean genome, especially in the gene-rich regions; and therefore, causative allele for SDS resistance may not have been detected by the SNP panel used in this study. Thus, some or all of the five candidate genes for SDS resistance identified in this study may not encode SDS resistance. However, these five genes and other SNPL linked to the 13 novel genomic regions can be considered potential molecular markers for enhancing SDS resistance in soybean because linkage disequilibrium in soybean is relatively large, 90 to 574 kb [79].
In summary, the GWAS conducted on foliar SDS and root rot phenotypes of 254 PI lines collected in growth chambers led to identification of 27 SNPL mapped to 21 genomic regions for SDS resistance. Eight of the 21 genomic regions for SDS resistance were previously reported QTL validating the outcomes of our GWAS. Thirteen novel genomic regions containing putative SDS resistance genes and 10 PI lines with both foliar SDS and root rot resistance identified in this investigation will significantly contribute towards breeding soybean cultivars for SDS resistance. Five candidate genes for SDS resistance identified in this study will facilitate further studies for advancing our knowledge of SDS resistance mechanisms in soybean.
Supporting information
S2 Table. QTL containing foliar SDS resistance and root rot resistance genes.
https://doi.org/10.1371/journal.pone.0212071.s002
(PDF)
S1 Fig. SDS scoring scheme.
The severity of SDS foliar symptoms were scored with an increment of 0.5. The foliar SDS score of each plant was recorded 4 to 5 weeks after planting.
https://doi.org/10.1371/journal.pone.0212071.s003
(PDF)
S2 Fig. Root rot scoring scheme.
Root rot symptoms were scored in a percentage scale from 0 to 100 with an increment of 5 of the total root area.
https://doi.org/10.1371/journal.pone.0212071.s004
(PDF)
S3 Fig. Neighbor joining tree.
Neighbor joining tree showing the relatedness of 254 soybean PI lines was constructed from 31,506 SNPs using Tassel 5.2.33 program. Subgroups are color coded.
https://doi.org/10.1371/journal.pone.0212071.s005
(PDF)
S4 Fig. Principal components analysis (PCA).
(A), PCA showing the extent of relatedness among 254 PI lines with the countries of their origin. (B) PCA showing the extent of relatedness among 254 PI lines based on their maturity groups. (C), PCA showing the extent of relatedness among 254 PI lines based on their foliar SDS score. (D) PCA showing the extent of relatedness among 254 PI lines generated based on root rot.
https://doi.org/10.1371/journal.pone.0212071.s006
(PDF)
S5 Fig. Distribution of the foliar SDS disease scores among the 254 PI lines identified from a collection of over 6,000 soybean PI lines.
(A), The frequency distribution of the 254 selected lines for foliar SDS scores. (B), The Q-Q plot of foliar SDS scores of the 254 PI lines. This experiment was conduct earlier in a greenhouse located in the University of Illinois, Champaign by the Hartman Lab.
https://doi.org/10.1371/journal.pone.0212071.s007
(PDF)
S6 Fig. Histogram of the residuals and Q-Q plots of the severities of foliar disease scores or root rots (%) among the 254 selected PI lines.
(A), The residual plot of the foliar SDS scores among the 254 PI lines. (B), The Q-Q plot of the foliar SDS scores among 254 PI lines. (C), The residual plot of the root rot (%) among the 254 PI lines. (D), The Q-Q plot of the root rot (%) among the 254 PI lines.
https://doi.org/10.1371/journal.pone.0212071.s008
(PDF)
S7 Fig. The Ser322Arg mutation in a putative LRR-receptor protein encoded by Glyma.13g035700.1 is localized to a conserved amino acid residue (highlighted with yellow).
The Glyma.13G035700.1 (XP_003543967.1) protein identified in our study is highlighted in green. The list of highly homologous proteins to Glyma.13G035700.1 are: KHN15520.1 (Glycine soja); XP_014622059.1 (Glycine max); XP_025980869.1 (Glycine max); XP_006596249.2 (Glycine max); RDX78236.1 (Mucuna pruriens); XP_020225386.1 (Cajanus cajan); XP_014495638.2 (Vigna radiata var. radiata); XP_007162178.1 (Phaseolus vulgaris); XP_003543967.1 (Glyma.13G035700.1) (Glycine max); KHN09050.1 (Glycine soja); XP_014505177.1 (Vigna radiata var. radiata); XP_014505176.1 (Vigna radiata var. radiata); BAT82169.1 (Vigna angularis var. angularis); XP_017428871.1 (Vigna angularis); KOM48560.1 (Vigna angularis); KOM48561.1 (Vigna angularis); BAT82170.1 (Vigna angularis var. angularis); XP_022637426.1 (Vigna radiata var. radiata); XP_022637938.1 (Vigna radiata var. radiata); XP_007161408.1 (Phaseolus vulgaris); XP_007161410.1 Phaseolus vulgaris); XP_003545672.1 (Glycine max); KHN35232.1 (Glycine soja); RDX65315.1 (Mucuna pruriens); XP_020225379.1 (Cajanus cajan).
https://doi.org/10.1371/journal.pone.0212071.s009
(PDF)
S8 Fig. The Glu261Lys mutation in a putative LRR-receptor protein encoded by Glyma.13g079100.1 is localized to a conserved amino acid residue (highlighted with yellow).
The Glyma.13G079100.1 (KRH18726.1) protein identified in our study is highlighted in green. The list of highly homologous proteins to Glyma.13G079100.1 are: XP_007161165.1 (Phaseolus vulgaris); XP_007161182.1 (Phaseolus vulgaris); RDX99908.1 (Mucuna pruriens); KRH18016.1 (Glycine max); KRH18017.1 (Glycine max); NP_001238112.1 (Glycine max); KRH18018.1 (Glycine max); RDX92217.1 (Mucuna pruriens); KRH18726.1 (Glyma.13G079100.1) (Glycine max); KRH15824.1 (Glycine max); XP_014622088.1 (Glycine max); NP_001237957.1 (Glycine max); XP_025982075.1 (Glycine max); KRH05215.1 (Glycine max); XP_006600115.2 (Glycine max); KRH05216.1 (Glycine max); KRH05221.1 (Glycine max); KRH05219.1 (Glycine max); XP_006601168.1 (Glycine max); XP_014625496.1 (Glycine max); KRH05237.1 (Glycine max); KRH05235.1 (Glycine max); XP_014625601.1 (Glycine max); XP_014625602.1 (Glycine max).
https://doi.org/10.1371/journal.pone.0212071.s010
(PDF)
S9 Fig. The Pro70Leu mutation in an unknown protein encoded by Glyma.14g035000 is localized to a conserved amino acid residue (highlighted with yellow).
The Glyma.14g035000 (KRH14580.1) protein identified in our study is highlighted in green. The list of highly homologous proteins to Glyma.14g035000 are: XP_007141497.1 (Phaseolus vulgaris); KOM46644.1 (Vigna angularis); RDX61322.1 (Mucuna pruriens); KRH14580.1 (Glyma.14g035000) (Glycine max); KRH73546.1 (Glycine max); AET00215.1 (Medicago truncatula); PNX72299.1 (Trifolium pratense); GAU22154.1 (Trifolium subterraneum); PKI57765.1 (Punica granatum); OWM69137.1 (Punica granatum); ONI14172.1 (Prunus persica); PQP95689.1 (Prunus yedoensis var. nudiflora); PON97614.1 (Trema orientale); POE81190.1 (Quercus suber); OIW04185.1 (Lupinus angustifolius); OIW02814.1 (Lupinus angustifolius); XP_010651515.1 (Vitis vinifera); XP_007146115.1 (Phaseolus vulgaris); KOM26446.1 (Vigna angularis); KYP52351.1 (Cajanus cajan); RDY02770.1 (Mucuna pruriens); KHN28152.1 (Glycine soja); KRH50598.1 (Glycine max); KHN14693.1 (Glycine soja); KRG89674.1 (Glycine max).
https://doi.org/10.1371/journal.pone.0212071.s011
(PDF)
S10 Fig. The Ser14Asn mutation in a late embryogenesis abundant (LEA) hydroxyproline-rich glycoprotein encoded by Glyma.01g222900.1 is localized to a conserved amino acid residue (highlighted with yellow).
The Glyma.01G222900.1 protein identified in our study is highlighted in green. The list of highly homologous proteins to Glyma.01G222900.1 are: XP_006574309.1 (Glycine max); XP_006591523.1 (Glycine max).
https://doi.org/10.1371/journal.pone.0212071.s012
(PDF)
S11 Fig. The His201Arg mutation in an heparan-alpha-glucosaminide N-acetyltransferase encoded by Glyma.10g058700 is localized to a conserved amino acid residue (highlighted with yellow).
The Glyma.10G058700.1 (KRH32551.1) protein identified in our study is highlighted in green. The list of highly homologous proteins to Glyma.10G058700.1 are: XP_016651648.1 (Prunus mume); XP_018814810.1 (Juglans regia); XP_008241451.1 (Prunus mume); XP_024022989.1 (Morus notabilis); POO01767.1 (Trema orientale); PON49032.1 (Parasponia andersonii); XP_017251878.1 (Daucus carota subsp. sativus); PKI57622.1 (Punica granatum); XP_010039059.1 (Eucalyptus grandis); KRH32551.1 (Glycine max); XP_025979781.1 (Glycine max); XP_017413005.1 (Vigna angularis); XP_014511918.1 (Vigna radiata var. radiata); XP_006595362.1 (Glycine max); RCW19085.1 (Glycine max); XP_007144374.1 (Phaseolus vulgaris); KHN48245.1 (Glycine soja); XP_020239895.1 (Cajanus cajan); KYP41914.1 (Cajanus cajan); XP_004515935.1 (Cicer arietinum); XP_024633997.1 (Medicago truncatula); XP_004494902.1 (Cicer arietinum); XP_007162785.1 (Phaseolus vulgaris); XP_020211654.1 (Cajanus cajan); KYP70913.1 (Cajanus cajan).
https://doi.org/10.1371/journal.pone.0212071.s013
(PDF)
Acknowledgments
This work was supported by the National Institute of Food and Agriculture (NIFA), United States Department of Agriculture (Grant no. 2013-68004-20374). We thank USDA-ARS- GRIN National Genetic Resources Program for sending us the seeds of soybean lines. We also thank Alexander Luckew, Peter Lundeen, Gregory Gebhart, Vijitha Silva and Sharon Eggenberger for their assistance during the course of the work.
References
- 1.
Thoenes P. Soybean international commodity profile. In: Competitive commercial agriculture in sub–saharan Africa (CCAA) study, FAO, UN. 2015:1–25.
- 2.
Bradley C, Allen T. Estimates of soybean yield reductions caused by diseases in the United States. University of Illinois Urbana-Champaign, Department of Crop Science extension and outreach. http://extensioncropsciencesillinoisedu/fieldcrops/diseases/yield_reductionsphp. 2014.
- 3. Hughes TJ, O’Donnell K, Sink S, Rooney AP, Scandiani MM, Luque A, et al. Genetic architecture and evolution of the mating type locus in fusaria that cause soybean sudden death syndrome and bean root rot. Mycologia. 2014;106(4):686–97. pmid:24891421
- 4. Li S, Hartman GL. Molecular detection of Fusarium solani f. sp. glycines in soybean roots and soil. Plant Pathol. 2003;52(1):74–83.
- 5. O’Donnell K, Sink S, Scandiani MM, Luque A, Colletto A, Biasoli M, et al. Soybean sudden death syndrome species diversity within North and South America revealed by multilocus genotyping. Phytopathology. 2009;100(1):58–71. pmid:19968550
- 6. Aoki T, Scandiani MM, O’Donnell K. Phenotypic, molecular phylogenetic, and pathogenetic characterization of Fusarium crassistipitatum sp. nov., a novel soybean sudden death syndrome pathogen from Argentina and Brazil. Mycoscience. 2012;53(3):167–86. https://doi.org/10.1007/S10267-011-0150-3.
- 7. Li S, Hartman GL, Widholm JM. Viability staining of soybean suspension cultured cells and a stem-cutting assay to evaluate phytotoxicity of Fusarium solani culture filtrates. Plant Cell Rep. 1999;18:375–80.
- 8. Abeysekara NS, Bhattacharyya MK. Analyses of the xylem sap proteomes identified candidate Fusarium virguliforme proteinacious toxins. PLos One. 2014;9(5). pmid:24845418
- 9. Brar H, Swaminathan S, Bhattacharyya M. The Fusarium virguliforme toxin FvTox1 causes foliar sudden death syndrome-like symptoms in soybean. Mol Plant-Microbe Interact. 2011;24(10):1179–88. pmid:21635141
- 10.
Navi SS, Yang XB. Foliar symptom expression in association with early infection and xylem colonization by Fusarium virguliforme (formerly F. solani f. sp. glycines), the causal agent of soybean sudden death syndrome. Plant Health Prog. 2008;(February):0222–01. CABI:20083261184.
- 11. Pudake R, Swaminathan S, Sahu B, Leandro L, Bhattacharyya M. Investigation of the Fusarium virguliforme fvtox1 mutants revealed that the FvTox1 toxin is involved in foliar sudden death syndrome development in soybean. Current Genet. 2013;59(3):107–17.
- 12.
Hershman DE. Soybean sudden death syndrome in Kentucky. In Inform Bull (Univ Kentucky Coop Ext Serv), pp. 37. 1990.
- 13.
Hirrel MC. Sudden death syndrome of soybean: New insight into its development. In Am Seed Trade Assoc, 16th Soybean Res Conf pp. 95–104. 1987.
- 14. Roy KW, Hershman DE, Rupe JC, Abney TS. Sudden death syndrome of soybean. Plant Dis. 1997;81(10):1100–11.
- 15. Mueller DS, Nelson RL, Hartman GL, Pedersen WL. Response of commercially developed soybean cultivars and the ancestral soybean lines to Fusarium solani f. sp. glycines. Plant Dis. 2003;87(7):827–31.
- 16.
Robertson A, Leandro L. Answers to questions about soybean sudden death syndrome in Iowa 2010. Integrated Crop Management News and Iowa State University Extension http://wwwextensioniastateedu/CropNews/2010/0907robertsonleandrohtml. 2010.
- 17. de Farias-Neto AL, Hashmi R, Schmidt M, Carlson S, Hartman GL, Li S, et al. Mapping and confirmation of a new sudden death syndrome resistance QTL on linkage group D2 from the soybean genotypes PI 567374 and ‘Ripley’. Mol Breed. 2007;20(1):53–62.
- 18. Kandel YR, Bradley CA, Wise KA, Chilvers MI, Tenuta AU, Davis VM, et al. Effect of glyphosate application on sudden death syndrome of glyphosate-resistant soybean under field conditions. Plant Dis. 2015;99(3):347–54. pmid:30699703
- 19. Chang SJC, Doubler TW, Kilo V, Suttner R, Klein J. Two additional loci underlying durable field resistance to soybean sudden death syndrome (SDS). Crop Sci. 1996;36(6):1684–8.
- 20. Hnetkovsky N, Chang SJC, Doubler TW, Gibson PT, Lightfoot DA. Genetic mapping of loci underlying field resistance to soybean sudden death syndrome (SDS). Crop Sci. 1996;36(2):393–400.
- 21. Kassem MA, Meksem K, Wood AJ, Lightfoot DA. Loci underlying SDS and SCN resistance mapped in the ‘Essex’ by ‘Forrest’ soybean recombinant inbred lines. Rev Biol Biotechnol. 2007;6(1):2–10.
- 22. Kassem MA, Ramos L, Leandro L, Mbofung G, Hyten DL, Kantartzi SK, et al. The ‘PI 438489B’ by ‘Hamilton’ SNP-based genetic linkage map of soybean [Glycine max (L.) Merr.] identified quantitative trait loci that underlie seedling SDS resistance. J Plant Genome Sci. 2012;1(1):18–30.
- 23. Kassem MA, Shultz J, Meksem K, Cho Y, Wood AJ, Iqbal MJ, et al. An updated ‘Essex’ by ‘Forrest’ linkage map and first composite interval map of QTL underlying six soybean traits. Theor Appl Genet. 2006;113(6):1015–26. pmid:16953420
- 24. Njiti VN, Meksem K, Iqbal MJ, Johnson JE, Kassem MA, Zobrist AF, et al. Common loci underlie field resistance to soybean sudden death syndrome in Forrest, Pyramid, Essex, and Douglas. Theor Appl Genet. 2002;104:294–300. pmid:12582700
- 25. Kazi S, Njiti VN, Doubler TW, Yuan J, Iqbal JM, Cianzio S, et al. Registration of the Flyer × Hartwig recombinant inbred line mapping population. J Plant Regist. 2007;1(2):175–8.
- 26. Kazi S, Shultz J, Afzal J, Johnson J, Njiti VN, Lightfoot DA. Separate loci underlie resistance to root infection and leaf scorch during soybean sudden death syndrome. Theor Appl Genet. 2008;116(7):967–77. pmid:18324383
- 27. Njiti VN, Lightfoot DA. Genetic analysis infers Dt loci underlie resistance to Fusarium solani f. sp glycines in indeterminate soybeans. Can J Plant Sci. 2006;86(1):83–90.
- 28. Prabhu RR, Njiti VN, Bell Johnson B, Johnson JE, Schmidt ME. Selecting soybean cultivars for dual resistance to soybean cyst nematode and sudden death syndrome using two DNA markers. Crop Sci. 1999;39(4):982–7.
- 29. Stephens PA, Nickell CD, Kolb FL. Genetic analysis of resistance to Fusarium solani in soybean Crop Sci. 1993;33:929–30.
- 30. Yuan J, Bashir R, Salas G, Sharma H, Srour A, Lightfoot DA. New approaches to selecting resistance or tolerance to SDS and Fusarium root rot. J Plant Genome Sci. 2012;1(1):10–7.
- 31. Chang H-X, Roth MG, Wang D, Cianzio SR, Lightfoot DA, Hartman GL, et al. Integration of sudden death syndrome resistance loci in the soybean genome. Theor Appl Genet. 2018;131(4):757–73. pmid:29435603
- 32. Luckew AS, Swaminathan S, Leandro LF, Orf JH, Cianzio SR. 'MN1606SP’ by ‘Spencer’ filial soybean population reveals novel quantitative trait loci and interactions among loci conditioning SDS resistance. Theor Appl Genet. 2017;130(10):2139–49. pmid:28748252
- 33. Zhang J, Singh A, Mueller DS, Singh AK. Genome-wide association and epistasis studies unravel the genetic architecture of sudden death syndrome resistance in soybean. Plant J. 2015;84(6):1124–36. pmid:26561232
- 34. Chang H-X, Lipka AE, Domier LL, Hartman GL. Characterization of disease resistance loci in the USDA soybean germplasm collection using genome-wide association studies. Phytopathology. 2016;106(10):1139–51. pmid:27135674
- 35. Wen ZX, Tan RJ, Yuan JZ, Bales C, Du WY, Zhang SC, et al. Genome-wide association mapping of quantitative resistance to sudden death syndrome in soybean. BMC Genomics. 2014;15.
- 36. Rupe JC, Sabbe WE, Robbins RT, Gbur EE. Soil and plant factors associated with sudden death syndrome of soybean. J Prod Agric. 1993;6(2):218–21.
- 37. Hartwig EE, Epps JM. Registration of Forrest soybeans. Crop Sci. 1973;13(2):287-.
- 38. Njiti VN, Suttner RJ, Gray LE, Gibson PT, Lightfoot DA. Rate-reducing resistance to Fusarium solani f. sp. phaseoli underlies field resistance to soybean sudden death syndrome. Crop Sci. 1997;37(1):132–8.
- 39. Njiti VN, Johnson JE, Torto TA, Gray LE, Lightfoot DA. Inoculum rate influences selection for field resistance to soybean sudden death syndrome in the greenhouse. Crop Sci. 2001;41(6):1726–31.
- 40. Hartman GL, Huang YH, Nelson RL, Noel GR. Germplasm evaluation of Glycine max for resistance to Fusarium solani, the causal organism of sudden death syndrome. Plant Dis. 1997;81:515–8.
- 41. Mueller DS, Hartman GL, Nelson RL, Pedersen WL. Evaluation of Glycine max germ plasm for resistance to Fusarium solani f. sp. glycines. Plant Dis. 2002;86(7):741–6.
- 42. Sanogo S, Yang XB, Scherm H. Effects of herbicides on Fusarium solani f. sp. glycines and development of sudden death syndrome in glyphosate-tolerant soybean. Phytopathology. 2000;90(1):57–66. pmid:18944572
- 43. Scherm H, Yang XB, Lundeen P. Soil variables associated with sudden death syndrome in soybean fields in Iowa. Plant Dis. 1998;82(10):1152–7.
- 44. Hartman GL, Huang YH, Li S. Phytotoxicity of Fusarium solani culture filtrates from soybeans and other hosts assayed by stem cuttings. Australas Plant Pathol. 2004;33(1):9–15.
- 45. Lozovaya VV, Lygin AV, Zernova OV, Li S, Widholm JM, Hartman GL. Lignin degradation by Fusarium solani f. sp glycines. Plant Dis. 2006;90(1):77–82.
- 46. Achenbach LA, Patrick J, Gray L. Use of RAPD markers as a diagnostic tool for the identification of Fusarium solani isolates that cause soybean sudden death syndrome. Plant Dis. 1996;80(11):1228–32.
- 47. Iqbal MJ, Yaegashi S, Ahsan R, Shopinski KL, Lightfoot DA. Root response to Fusarium solani f. sp. glycines: temporal accumulation of transcripts in partially resistant and susceptible soybean. Theor Appl Genet. 2005;110(8):1429–38. pmid:15815926
- 48. Srivastava SK, Huang X, Brar HK, Fakhoury AM, Bluhm BH, Bhattacharyya MK. The genome sequence of the fungal pathogen Fusarium virguliforme that causes sudden death syndrome in soybean. PLos One. 2014;9(1):e81832. pmid:24454689
- 49. Luckew AS, Cianzio SR, Leandro LF. Screening method for distinguishing soybean resistance to Fusarium virguliforme in resistant x resistant crosses. Crop Sci. 2012;52(5):2215–23.
- 50. Hayden RW. A Review of: “An R companion to applied regression, Second Edition, by J. Fox and S. Weisberg”. J Biopharm Stat. 2012;22(2):418–9.
- 51.
Mendiburu FD. Agricolae: statistical procedures for agricultural research. R Package Version 1.2–3. https://cran.r-project.org/web/packages/agricolae/index.html. 2015.
- 52. Song QJ, Hyten DL, Jia GF, Quigley CV, Fickus EW, Nelson RL, et al. Development and evaluation of SoySNP50K, a high-density genotyping array for soybean. PLos One. 2013;8(1). pmid:23372807
- 53.
Bates D, Maechler M, Bolker B, Walker S. lme4: Linear Mixed- Effects models using Eigen and S4. R. package version 1. http://CRAN.R-project.org/package=lme4. 2014.
- 54. Zhang Z, Ersoz E, Lai C-Q, Todhunter RJ, Tiwari HK, Gore MA, et al. Mixed linear model approach adapted for genome-wide association studies. Nature Genet. 2010;42(4):355–U118. pmid:20208535
- 55. Lipka AE, Tian F, Wang Q, Peiffer J, Li M, Bradbury PJ, et al. GAPIT: genome association and prediction integrated tool. Bioinformatics. 2012;28(18):2397–9. pmid:22796960
- 56. Ng A, Xavier RJ. Leucine-rich repeat (LRR) proteins Integrators of pattern recognition and signaling in immunity. Autophagy. 2011;7(9):1082–4. pmid:21606681
- 57. Huang LF, Lin KH, He SL, Chen JL, Jiang JZ, Chen BH, et al. Multiple patterns of regulation and overexpression of a ribonuclease-like pathogenesis-related protein gene, OsPR10a, conferring disease resistance in rice and Arabidopsis. PLos One. 2016;11(6). pmid:27258121
- 58. Park CJ, Seo YS. Heat shock proteins: A review of the molecular chaperones for plant immunity. Plant Pathol J. 2015;31(4):323–33. Epub 12/30. pmid:26676169.
- 59. Singh NK, Kumar KRR, Kumar D, Shukla P, Kirti PB. Characterization of a pathogen induced thaumatin-like protein gene AdTLP from Arachis diogoi, a wild peanut. PLos One. 2013;8(12). pmid:24367621
- 60. Seifi HS, Van Bockhaven J, Angenon G, Höfte M. Glutamate metabolism in plant disease and defense: friend or foe? Mol Plant-Microbe Interact. 2013;26(5):475–85. pmid:23342972
- 61. Sharma M, Pandey GK. Expansion and function of repeat domain proteins during stress and development in plants. Front Plant Sci. 2015;6:1218. pmid:26793205
- 62. Kaundal A, Ramu VS, Oh S, Lee S, Pant B, Lee H-K, et al. GENERAL CONTROL NONREPRESSIBLE4 degrades 14-3-3 and the RIN4 complex to regulate stomatal aperture with implications on nonhost disease resistance and drought tolerance. Plant Cell. 2017;29(9):2233–48. pmid:28855332
- 63. Li B, Feng H, Zhang Q, Zhang X, Hong Y. A cellular gene as a double surveillance agent for plant to combat pathogen. Plant Signal Behav. 2013;8(10):e26817.
- 64. Sharfman M, Bar M, Schuster S, Leibman M, Avni A. Sterol-dependent induction of plant defense responses by a microbe-associated molecular pattern from Trichoderma viride. Plant Physiol. 2014;164(2):819–27. pmid:24351686
- 65. Li YB, Han LB, Wang HY, Zhang J, Sun ST, Feng DQ, et al. The thioredoxin GbNRX1 plays a crucial role in homeostasis of apoplastic reactive oxygen species in response to Verticillium dahliae infection in cotton. Plant Physiol. 2016;170(4):2392–406. pmid:26869704
- 66. Camejo D, Guzman-Cedeno A, Moreno A. Reactive oxygen species, essential molecules, during plant-pathogen interactions. Plant Physiol Biochem. 2016;103:10–23. pmid:26950921
- 67. Singh S, Cornilescu CC, Tyler RC, Cornilescu G, Tonelli M, Lee MS, et al. Solution structure of a late embryogenesis abundant protein (LEA14) from Arabidopsis thaliana, a cellular stress-related protein. Protein Sci. 2005;14(10):2601–9. pmid:16155204
- 68. Powell JJ, Carere J, Sablok G, Fitzgerald TL, Stiller J, Colgrave ML, et al. Transcriptome analysis of Brachypodium during fungal pathogen infection reveals both shared and distinct defense responses with wheat. Scientific Reports. 2017;7(1):17212. pmid:29222453
- 69. Ruan L, Zhang JB, Xin XL, Zhang CZ, Ma DH, Chen L, et al. Comparative analysis of potassium deficiency-responsive transcriptomes in low potassium susceptible and tolerant wheat (Triticum aestivum L.). Sci Rep. 2015;5. pmid:25985414
- 70. Wang H, Wijeratne A, Wijeratne S, Lee S, Taylor CG, St Martin SK, et al. Dissection of two soybean QTL conferring partial resistance to Phytophthora sojae through sequence and gene expression analysis. BMC Genomics. 2012;13(1):428. pmid:22925529
- 71. Narasimhan ML, Damsz B, Coca MA, Ibeas JI, Yun DJ, Pardo JM, et al. A plant defense response effector induces microbial apoptosis. Mol Cell. 2001;8(4):921–30. pmid:11684026
- 72. Malvick DK, Bussey KE. Comparative analysis and characterization of the soybean sudden death syndrome pathogen Fusarium virguliforme in the northern United States. Can J Plant Pathol. 2008;30(3):467–76.
- 73. Mbofung GYC, Harrington TC, Steimel JT, Navi SS, Yang XB, Leandro LF. Genetic structure and variation in aggressiveness in Fusarium virguliforme in the Midwest United States. Can J Plant Pathol. 2012;34(1):83–97.
- 74. Anderson J, Akond M, Kassem MA, Meksem K, Kantartzi SK. Quantitative trait loci underlying resistance to sudden death syndrome (SDS) in MD96-5722 by ‘Spencer’ recombinant inbred line population of soybean. 3 Biotech. 2015;5(2):203–10. pmid:28324582
- 75. Swaminathan S, Abeysekara NS, Liu M, Cianzio SR, Bhattacharyya MK. Quantitative trait loci underlying host responses of soybean to Fusarium virguliforme toxins that cause foliar sudden death syndrome. Theor Appl Genet. 2016;129(3):495–506. pmid:26678962
- 76. Swaminathan S, Abeysekara NS, Knight JM, Liu M, Dong J, Hudson ME, et al. Mapping of new quantitative trait loci for sudden death syndrome and soybean cyst nematode resistance in two soybean populations. Theor Appl Genet. 2018;131(5):1047–62. pmid:29582113
- 77. Arumuganathan K, Earle ED. Nuclear DNA content of some important plant species. Plant Mol Biol Rep. 1991;9(3):208–18.
- 78. Schmutz J, Cannon SB, Schlueter J, Ma J, Mitros T, Nelson W, et al. Genome sequence of the palaeopolyploid soybean. Nature. 2010;463:178. pmid:20075913
- 79. Hyten DL, Choi IY, Song Q, Shoemaker RC, Nelson RL, Costa JM, et al. Highly variable patterns of linkage disequilibrium in multiple soybean populations. Genetics. 2007;175(4):1937–44. pmid:17287533