Chemical mutagenesis and soybean mutants potential for identification of novel genes conferring resistance to soybean cyst nematode

GE Feng-yong, ZHENG Na, ZHANG Liu-ping, HUANG Wen-kun, PENG De-liang, LIU Shi-ming,

1 Institute of Plant Protection, Chinese Academy of Agricultural Sciences, Beijing 100193, P.R.China

2 College of Plant Protection, Hunan Agricultural University, Changsha 410128, P.R.China

Abstract The resistance of soybean (Glycine max (L.) Merr.) to soybean cyst nematode (SCN, Heterodera glycines Ichinohe), which is a devastating pathogen in soybean production and causes a large quantity of annual yield loss worldwide, can shift during the long-term interaction and domestication. It is vital to identify more new resistance genetic sources for identiflcation of novel genes underlying resistance to SCN for management of this pathogen. In the present study, flrst, two ethane methylsulfonate-mutagenesis soybean M2 populations of PI 437654, which shows a broad resistance to almost all of SCN races, and Zhonghuang 13, which is a soybean cultivar in China conferring strong resistance to lodging, were developed.Many types of morphological phenotypes such as four- and flve-leaflet leaves were observed from these two soybean M2 populations. Second, 13 mutants were identifled and conflrmed to exhibit alteration of resistance to SCN race 4 through the forward genetic screening of 400 mutants of the PI 437654 M2 population, the rate of mutants with alteration of SCN-infection phenotype is 3.25%. Third, these identifled mutants were further verifled not to show any changes in the genomic sequences of the three known SCN-resistant genes, GmSHMT08, GmSNAP18 and GmSANP11, compared to the wildtype soybean; and all of them were still resistant to SCN race 3 similar to the wild-type soybean. Taken together, we can conclude that the 13 mutants identifled in the present study carry the mutations of the new gene(s) which contribute(s) to the resistance to SCN race 4 in PI 437654 and can be potentially used as the genetic soybean sources to further identify the novel SCN-resistant gene(s).

Keywords: soybean, ethane methylsulfonate-mutagenesis populations, mutants, soybean cyst nematode race 4, resistance

1. Introduction

Soybean cyst nematode (SCN),Heterodera glycinesIchinohe, is the most important pathogen on soybean(Glycine max(L.) Merr.) and causes more than US$1 billion in annual yield loss in the United States alone (Koenning and Wrather 2010). SCN has 14 races and often bursts out in the soybean-production areas in China (Penget al.2016). In China, SCN races 1 and 3 are mainly distributed in the Northeast soybean-production areas, while SCN race 4 majorly occurs in the Huanghuaihai soybean-production areas. Planting resistant cultivars is the most effective,economical and environmentally friendly means to control this pathogen. For breeding resistant cultivars, it is of priority to identify the resistant QTL (quantitative trait locus) and genes against SCN and to understand the mechanism of soybean resistance.

Many QTL (more than 160 as of) underlying the resistance of soybean to SCN have been mapped on 19 of 20 chromosomes of soybean (http://www.soybase.org).However, up to now, only two QTLs have been consistently mapped in a variety of soybean germplasm sources, these two QTLs are therhg1on chromosome 18 and theRhg4on chromosome 8, and both have been identifled as the two major resistance QTL in soybean (Concibidoet al. 2004).There are two types of soybeans conferring the resistance to SCN, one is the Peking-type soybeans which require bothrhg1andRhg4to show the resistance to SCN, and the other is PI 88788-type soybeans, in which the solerhg1is sufflcient to exhibit the resistance (Meksemet al. 2001;Concibidoet al. 2004; Cooket al. 2012; Liuet al. 2012,2017). Obviously, the Peking-typerhg1is different from the PI 88788-typerhg1, which was therefore classifled asrhg1-aandrhg1-b, respectively (Brukeret al. 2005; Liuet al. 2017). These three genes,rhg1-a,rhg1-bandRhg4,have been identifled and functionally validated (Cooket al.2012; Liuet al. 2012, 2017). TheGmSNAP18, a soluble NSF (N-ethlmaleimide-sensitive factor) attachment protein gene, is therhg1-agene in the Peking-type soybeans (Liuet al. 2017); therhg1-bconsists of three genes,GmAAT(Glyma18g02580),GmSNAP18(Glyma18g02590), andGmWI12(Glyma18g02610), these three genes locate within a genomic segment of about 31 kb on chromosome 18 and simultaneously contribute to the SCN resistance by high copy number of this segment (2–10 direct repeats)whereas only one copy of this segment presents in susceptible soybeans (Cooket al. 2012, 2014; Leeet al.2015; Yuet al. 2016); and theGmSHMT08, a serine hydroxymethyltransferase gene, is theRhg4gene (Liuet al.2012; Kandothet al. 2017). Recently, Lakhssassiet al.(2017) identifled that theGmSNAP11gene on chromosome 11 also makes minor contributions to the resistance to SCN.Moreover, additional work on the mechanism of soybean resistance to SCN was performed. Baylesset al. (2016)reported that a large quantity of the aberrant GmSNAP18 protein was accumulated at the nematode feeding sites, and then the cytotoxic effects were produced to disrupt SNARE(soluble NSF attachment protein receptor) complexes and vesicle trafflcking, likely causing the resistance of soybeanrhg1to SCN.

The virulent nematode populations have been identifled for most known resistant sources (Mitchumet al. 2007;Colgrove and Niblack 2008; Micthum 2016). During the long-term incompatible interaction between resistant host soybeans and SCN, the soybeans are gradually domesticated under the selection pressure and then partially or even entirely lose their resistance. Recently,Kandothet al. (2017) reported that some SCN-high resistant soybeans such as Peking signiflcantly shifted to moderate resistance, even to moderate susceptibility when they were infected with the SCN population that was continuously selected on a resistant soybean line, while these SCN-resistant soybeans were still high resistant when they were infected with the SCN population that was continuously selected on a susceptible line, for about 60 generations in flve years. Therefore, it is desirable to identify more resistance sources and resistant genes and utilize them to breed new resistant cultivars for the long-term management of this pathogen. Shiet al. (2015) developed three speciflc KASP (kompetitive allele speciflc PCR) SNP(single nucleotide polymorphism) markers to efflciently identify the resistance of soybean sources. This provides a high-throughput and cost-effective method to identify the resistance of soybean.

The reference genome of soybean Williams 82 (Schmutzet al. 2010) and the next generation sequencing are greatly promoting the fast advance in functional genomics of soybean. Many functional genes were identifled recently.However, soybean is a crop that is recalcitrant to genetic transformation, which considerably hinders the validation of gene functions. So far, some soybean genes such as the above-mentionedrhg1a,rhg1-bandRhg4genes (Cooket al. 2012; Liuet al. 2012, 2017),rhg1 LRR-RLK(Melitoet al. 2010) andRhg4 LRR-RLK(Liuet al. 2011) were functioned by overexpression, genetic complementation and RNAi using the hairy roots transformation system or the VIGS (virus-induced gene silencing) gene silence transformation system (Kandothet al. 2013). Recently,some whole soybean transgenic plants were obtained to validate the functions of genes by genetic complementation or CRISPR/Cas9 genome editing (Caiet al. 2017; Kandothet al. 2017). However, all soybean transformation methods are still unstable and low efflcient as of.

Chemical or physical mutagenesis may provide a promising approach. Some chemicals such as ethane methylsulfonate (EMS) can randomly make the nucleotides mutated to form SNPs and small InDels (insertions and deletions) at the whole genome level. After mutagenesis, the large-scale mutants populations can be developed, along with the assistance of PCR (conventional TILLING, targeted induced local lesions in genomes) and/or next generation sequencing (TILLING by sequencing), it is easy to identify the mutants of target genes and then directly validate the functions of genes linking with the corresponding phenotype of mutants. Since McCallumet al. (2000) reported the chemical mutagenesisArabidopsismutants population and TILLING technology, many large-scale chemical mutagenesis populations of crops such as barley, maize,rice, soybean, tomato, watermelon, and wheat (Greeneet al. 2003; Perryet al. 2003; Tillet al. 2004; Cooperet al.2008; Dalmaiset al. 2008; Porcedduet al. 2008; Suzukiet al. 2008; Talamèet al. 2008; Wanget al. 2008; Xinet al.2008; Donget al. 2009) were developed, and many genes were functionally validated using these developed genetic resources. For example, the soybean SCN-resistantRhg4geneGmSHMT08was functionally validated using two mutants (F6266 and F6756) identifled from one EMS-mutagenesis Forrest population by conventional TILLING(Liuet al. 2012) and 13 mutants forward genetically identifled from two other EMS-mutagenesis Forrest populations(Kandothet al. 2017).

PI 437654 is a genetic soybean source conferring a broad resistance to almost all of SCN races (Wuet al.2009), including SCN race 4, which is the most virulent race to soybean. To our best knowledge, the soybean genes underlying resistance to SCN race 4 have not yet been achieved any progresses. Soybean cultivar Zhonghuang 13 is widely cultivated in China with many good agronomic traits such as strong resistance to lodging although it is susceptible to SCN. In the present study,PI 437654 and Zhonghuang 13 were utilized to develop two EMS mutagenesis soybean M2 (the second generation mutants) populations, from which many types of signiflcant morphological phenotypes such as multi-leaflet leaves were observed, so the developed M2 mutants were able to be used as the good genetic germplasm sources. Then 21 mutants were identifled, of which 13 mutants were conflrmed with alteration of resistance to SCN race 4 from the developed PI 437654 M2 population of 400 mutants.These identifled mutants are still resistant to SCN race 3,but no changes were detected in the genomic sequences of the three known SCN-resistant genes (Cooket al. 2012;Liuet al. 2012, 2017; Lakhssassiet al. 2017). The identifled mutants can be potentially used for identiflcation of the novel soybean gene(s) conferring resistance to SCN race 4.

2. Materials and methods

2.1. Soybean and SCN

PI 437654 is resistant to SCN whereas soybean cultivar Zhonghuang 13 is susceptible to SCN, which is cultivated in China with many good agronomic traits. The seeds of PI 437654 were raised by ourselves; the seeds of wildtype Zhonghuang 13 were obtained from Institute of Crop Sciences of Chinese Academy of Agricultural Sciences(CAAS). Cysts of SCN races 3 and 4 were isolated from the soils in the soybean-production fleld at Langfang Experimental Base of Institute of Plant Protection of CAAS referring to the method described by Niblacket al. (1993).The cysts were gently crushed to release the eggs using a rubber and the eggs were collected and suspended in tap water or 0.3 mmol L–1ZnCl2solution to be hatched. SCN-susceptible Zhonghuang 13 seedlings were inoculated with the hatched J2 (juvenile stage 2) to propagate SCN using the method described by Brownet al. (2010).

2.2. Development of two EMS-mutagenesis soybean M2 populations

The two soybean M2 populations, PI 437654 and Zhonghuang 13 populations, were developed using EMS mutagenesis employing the methods described by Cooperet al. (2008) and Meksemet al. (2008). The concentration of EMS was optimized at 60 and 36 mmol L–1for the mutagenesis of PI 437654 and Zhonghuang 13 seeds,respectively. The two M2 populations of PI 437654 and Zhonghuang 13 in this study were developed by EMS mutagenesis at the optimized concentrations at Langfang Experimental Base of Institute of Plant Protection of CAAS.

2.3. Morphological phenotypes of the chemically mutagenized soybean populations

During the entire growth and development of the soybean M2 mutants, compared to the wild-type soybeans, all the morphological phenotypes such as leaf color, leaflet numbers,leaf size and shape, branch numbers, pod development,seed size, plant height, and yield and plant numbers of each phenotype were recorded and the pictures were taken.

2.4. Infection of SCN and survey of phenotypes

The inoculation and infection of SCN to the soybean seedlings were performed according to the method described by Brownet al. (2010). Wild-type Zhonghuang 13 was used as the SCN-susceptible control. The cysts in the roots of mutants were counted under an Olympus SZX7 stereomicroscope (Olympus, Japan). The average numbers of cysts in the roots of each wild-type Zhonghuang 13 plant were calculated. All the wild-type soybeans and mutants were set up to 10 replicates in each experiment. Female index (FI, %)=(The cyst numbers in a plant/The average cyst numbers in a Zhonghuang 13 plant)×100. As a standard,the plant will be high resistant to SCN while the FI＜10%,whereas the plant will become susceptible to SCN while the FI＞30%. Because FI of each mutant and wild-type PI 437654 was calculated separately, all the FI values were not statistically analyzed.

2.5. Extraction of genomic DNA

The young leaves on the top of seedlings were harvested to extract genomic DNA. The extraction of DNA was carried out using the Qiagen DNeasy Plant Mini Kit (Qiagen Sciences,USA) per the manufacture instructions. The concentration and quality of extracted DNA was measured using a nanodrop microvolume spectrophotometer and fluorometer(Thermo Scientiflc, USA).

2.6. Extraction of RNA and synthesis of cDNA

The young leaves were harvested to extract RNA using general Trizol method. The cDNA was then synthesized employing the TransScript One-Step cDNA Synthesis SuperMix Kit (TransGen Biotech, China) per the instructions from the manufacture.

2.7. PCR amplification and sequencing of known SCN-resistant genes

The genomic sequences of three known SCN-resistant genes,GmSHMT08(Glyma08g11490, GenBank acc. no.JQ714083),GmSNAP18(Glyma18g02590, GenBank acc.no. KX147329) andGmSNAP11(Glyma11g35820), of the mutants and wild-type PI 437654, and the cDNA sequence ofGmSNAP11of wild-type PI 437654 were sequenced. We designed the primers forGmSNAP11, and the other primers forGmSHMT08andGmSNAP18are from the references(Liuet al. 2012, 2017). All the primers used in this study are listed in Table 1. The PCR was conducted using the extracted genomic DNA or the synthesized cDNA by the following program: denatured at 94°C for 4 min, 35 cycles of 94°C for 30 s, 50–60°C for 30 s and 72°C for 1–1.5 min,with extension at 72°C for 7 min. The PCR products were separated on the 1.5% agarose gel and then the band at the corrected size was cut and recovered, and purifled using AxyPrep DNA Gel Extraction Kit (Axygen, USA).The purifled band was then cloned into pMD19-T vector(TaKaRa, Japan) to form the plasmid. The plasmid was transformed into competent bacteriaEscherichia coliDH5α(TaKaRa, Japan) using heat shock treatment method and the transformed bacteria were overnight cultured on the LB plate with induction and selection of IPTG and X-gal at 37°C. Subsequently, the white colonies were selected and overnight cultured to extract the plasmid using EasyPure Plasmid MiniPrep Kit (TransGen Biotech, China), in the end,the extracted plasmid was sequenced at Sangon Biotech Co.(Beijing, China) to obtain the genomic or cDNA sequences of the genes.

3. Results

3.1. Development and major morphological phenotypes of two chemical mutagenesis soybean M2 populations

The optimal concentration of EMS was tested, and then EMS was used to mutagenize soybean PI 437654 which is resistant to SCN and Zhonghuang 13 which is susceptible to SCN employing the methods described by Cooperet al. (2008) and Meksemet al. (2008). The optimized concentration of EMS were 60 and 36 mmol L–1for the mutagenesis of PI 437654 and Zhonghuang 13 seeds,respectively. Then, a soybean PI 437654 M2 population containing 682 plants (mutants) and a Zhonghuang 13 M2 population containing 723 plants were developed by mutagenesis using the optimal concentrations of EMS mentioned above. Subsequently, two M2 seeds were planted from each mutant to raise to M3 seeds. From these M2 plants, many types of signiflcant morphological phenotypes were observed. Pictures of partial morphological phenotypes of mutants are shown in Fig. 1. Each phenotype shown in Fig. 1 was observed in at least eight mutants. Besides these phenotypes listed, there are still many other morphological phenotypes such as no pods, small pods,small seeds, early maturing, and narrow but long leaves(data not shown). Interestingly, the multiple leaflet leaves(four- or flve-leaflet leaves, Fig. 1-E and F, respectively)mutants were only observed in the Zhonghuang 13 M2 population, but not observed in the PI 437654 M2 population(both wild-type PI 437654 and Zhonghuang 13 soybeans are only three-leaflet leaves plants). Furthermore, some mutants contain both four- and flve-leaflet leaves in a single plant. Totally, 26 multiple leaflet leaves mutants were obtained from the developed Zhonghuang 13 population.Also, a tall and much more pods in Zhonghuang 13 mutant,which has 338 pods and is 107 cm in height. In contrast, the wild-type Zhonghuang 13 has only 132 pods and is 75 cm in height on the average, increasing by 156.1% for pod numbers and 42.7% for height (Fig. 1-N), and eight tall, more pods and late maturing mutants like the mutant shown in Fig. 1-M from the PI 437654 M2 population were obtained.

Table 1 Information of the primers used in this work

3.2. Forward genetic screening for identification of soybean mutants with alternation of resistance against SCN race 4

Two seeds from each mutant of the PI 437654 M2 population of the flrst 400 mutants were randomly selected and planted to perform the phenotyping with the infection of SCN race 4 in the greenhouse according to the method described by Brownet al. (2010). Twenty-one mutants with alteration of SCN-infection phenotype (FI＞10%) were primarily identifled, compared to the resistant wild-type PI 437654,although some mutants were still in segregation, using SCN-susceptible Zhonghuang 13 as the control (data not shown).Afterwards, 10 seeds of each identifled-above mutant were planted, but some seeds did not germinate. Totally, 92 plants were obtained. Subsequently, these plants with the infection of SCN race 4 were phenotyped. The results indicate that,flrst, 13 of the 21 primarily identifled mutants were conflrmed the alteration of phenotype (Table 2); second, because most of mutants were still under segregation (Table 2), the FI of only 24 of 92 plants was over 10% (Fig. 2). The rate of mutants with alteration of SCN-infection phenotype was 3.25%. In contrast to the control Zhonghuang 13 with the average cyst numbers of 142 each plant, the average cyst numbers on each wild-type PI 437654 plant were 8.2, while the cyst numbers of the mutants maximally reach to 24 (#16-215-6), almost the triple. The FI of wild-type PI 437654 was 5.8% while the FI of #16-215-6 was 16.9%.

3.3. Phenotyping of the identified mutants with the infection of SCN race 3

Fig. 1 Major morphological phenotypes of the mutants observed from both PI 437654 and Zhonghuang 13 M2 (the second generation mutants) populations. A, wild-type PI 437654. B, wild-type Zhonghuang 13. C, albino leaves. D, partial or entire chlorisis leaves.E, four-leaflet leaves. F, flve-leaflet leaves. G and H, wrinkled leaves. I, dwarf and compacted leaves. J, three branches and short internode. K, abortion of pod development. L, short internode. M, tall, more pods and late maturing (PI 437654). N, tall,more pods and high yield (Zhonghuang 13). O, seed coat color change (bottom, mutant; top, wild-type PI 437654). P, seed size increase (bottom, mutant; top, wild-type Zhonghuang 13).

Table 2 Survey of the cysts of soybean cyst nematode (SCN) race 4 in the roots of primarily identifled PI 437654 mutants

Fig. 2 Soybean cyst nematode (SCN) race 4 infection phenotype of partial single plants of the identifled mutants. Only the plants with FI (female index)＞10% are included, totally, 24 plants in 13 mutants. WT-PI 437654, wild-type PI 437654.

All the progresses on the identiflcation of the soybean genes conferring resistance to SCN were made in the studies of the resistance of soybean to SCN race 3 (Cooket al.2012; Liuet al. 2012, 2017; Lakhssassiet al. 2017), and PI 437654 was identifled as a genetic source with a broad resistance to almost all of SCN races (Wuet al. 2009). To further validate the resistance of these identifled mutants to SCN, seven representative mutants were selected to conduct the phenotyping with the infection of SCN race 3.The results (Table 3) showed that each plant of all these seven mutants was still resistant to SCN race 3 like the wild-type PI 437654.

Table 3 Survey of the cysts of soybean cyst nematode (SCN) race 3 in the roots of identifled PI 437654 mutants

3.4. Genomic analyses of the known SCN-resistant genes of those identified mutants

TheGmSHMT08(Glyma08g11490),GmSNAP18(Glyma18g02590) andGmSNAP11(Glyma11g35820)are the three reported genes that confer the resistance to SCN(Cooket al. 2012; Liuet al. 2012, 2017; Lakhssassiet al.2017). Regarding those PI 437654 mutants identifled with alteration of resistance to SCN race 4, to verify them as the novel genetic mutant sources that can be used for the identiflcation of new SCN resistant gene(s), the cloning and genomic analyses by sequencing of these three known SCN-resistant genes of the identifled mutants and the wildtype PI 437654 were conducted. The genomic sequence ofGmSHMT08of PI 437654 can refer to Liuet al. (2012) in addition to the sequence of ForrestGmSHMT08(GenBank acc. no. JQ714083). The genomic sequence ofGmSNAP18of PI 437654 is identical to that of ForrestGmSNAP18(GenBank acc. no. KX147329). The genomic sequence ofGmSNAP11of PI 437654 was deposited in NCBI GenBank with acc. no. MH899132. The cDNA sequence of PI 437654GmSNAP11(Fig. 3-A) shows that PI 437654GmSNAP11is a truncatedα-SNAPgene, only encoding an α-SNAP protein of 239 amino acids, while Williams 82GmSNAP11encodes an α-SNAP protein of 289 amino acids (Fig. 3-B).The sequence alignment results indicate that all the genomic sequences of these three genes of these mutants are identical to the corresponding sequences of wild-type PI 437654 (data not shown), meaning that none of these 13 identifled mutants carry any mutations in these three genes.

4. Discussion

SCN causes a large quantity of annual yield loss worldwide(Koenning and Wrather 2010; Penget al. 2016). Identifying the resistant genes and utilizing them to breed and cultivate the resistant cultivars is one key for the control of this pathogen. PI 437654 is a soybean source that confers a broad resistance to almost all of SCN races(Wuet al. 2009), so the chemical mutagenesis PI 437654 M2 population is a good genetic source for the functional identiflcation of the genes conferring resistance to SCN in soybean. Zhonghuang 13 is a soybean cultivar in China with many good agronomic traits such as resistance to lodging although it is susceptible to SCN, the Zhonghuang 13 mutant population is therefore a good genetic source for studying the other agronomic traits of soybean. Thus, these two soybeans were selected to develop two EMS-mutagenesis mutants populations for the new soybean TILLING resources, from which many mutants with morphological phenotypes were obtained (Fig. 1). Especially, a tall and much more pods mutant was obtained from the Zhonghuang 13 M2 population developed in the present study (Fig. 1-N).The pod numbers and height of this mutant were increased by 156.1 and 42.7%, respectively, compared to the wild-type Zhonghuang 13. Potentially, this Zhonghuang 13 mutant is likely bred to a high-production germplasm which is being tested in the fleld. The mutation rate of the various mutants populations developed by TILLING are quite different in the same species. In the present work, 13 mutants were identifled and conflrmed with alteration of SCN resistance from the PI 437654 M2 population of 400 mutants (Table 2).The rate of mutants with alteration of SCN infection phenotype is 3.25%. Liuet al. (2012) reported that two mutants (F6266 and F6756) ofGmSHMT08, which is a major SCN-resistant gene (Rhg4) in Peking-type soybeans,with alteration of SCN infection phenotype were reverse genetically identifled by conventional TILLING screening of an EMS-mutagenesis Forrest M2 population of 1 920 mutants, Forrest is a Peking-type soybean that is resistant to SCN race 3. Kandothet al. (2017) reported that thirteenGmSHMT08mutants with alteration of SCN infection phenotype were identifled by forward genetic screening of two other EMS-mutagenesis Forrest M2 populations(one has 1 225 mutants, and the other has 1 109 mutants).Obviously, the rate of mutants with SCN phenotype alteration in the mutagenized PI 437654 population developed in the present study is much higher than those of the three Forrest mutant populations mentioned above. The mutation rate of the chemically mutagenized soybean populations may depend on the varieties and targeted genes. These two M2 populations developed by EMS-mutagenesis in the present study are good mutants populations as the soybean genetic resources.

Since the resistance of soybean to SCN can shift during the interaction with SCN (Mitchumet al. 2007; Colgrove and Niblack 2008; Mitchum 2016; Kandothet al. 2017),identiflcation of new resistant genes is required for the control of this pathogen. In the present study, we forward genetically identifled and conflrmed 13 PI 437654 mutants.Compared to the wild-type PI 437654, the numbers of cysts developed in the roots of some plants of these mutants were signiflcantly increased (Table 2; Fig. 2), so the SCN infection phenotype of these mutants altered.

The genomic sequences of three known resistance genes:GmSHMT08,GmSNAP18andGmSNAP11(Cooket al. 2012; Liuet al. 2012, 2017; Lakhssassiet al. 2017),which, to our best knowledge, are all the genes identifled and functionally validated to confer resistance to SCN up to now, were analyzed in the present study. The results clearly showed that the genomic sequences of all the three genes in the identifled mutants are identical to those of the wild type PI 437654 (data not shown). So, there is (are)novel gene(s) contributing to the resistance of PI 437654 to SCN race 4.

Because all the three known SCN-resistant genes were identifled by the infection of SCN race 3 (Cooket al.2012; Liuet al. 2012, 2017; Lakhssassiet al. 2017), we still phenotyped them with the infection of SCN race 3 to further carry out the veriflcation. The results clearly indicated that the identifled mutants did not signiflcantly change the resistance to SCN race 3 (Table 3). Taken together with the analytical results of genomic sequences, we can conclude that these identifled and conflrmed mutants listed in Table 2 are all the new genetic sources potential for the identiflcation of novel genes conferring resistance to SCN.

In the conflrmation of the phenotype of the primarily identifled mutants, 24 plants having phenotype alteration from 92 plants were obtained (Table 2; Fig. 2), the percentage of mutant plants is 26%. Although these plants were the progenies of different mutants, we assume that the mutations might only take place within one gene. In addition, their phenotype range widely (FI from 10.6 to 16.9%), the mutations might be occurred at various positions in one gene.

Abeet al. (2012) and Takagiet al. (2013, 2015) developed the MutMap and QTL-seq methods to rapidly identify plant QTL and trait genes through whole-genome resequencing;the two rice populations they used were derived from the backcross of the EMS-mutagenesis mutants with the wildtype rice and showed extremely opposite trait values for a given phenotype in a segregating progeny; the rice QTL,such as partial resistance to the fungal rice blast disease and seedling vigor, were successfully identifled. These 13 identifled PI 437654 mutants may be used to identify the new SCN-resistant gene(s) in PI 437654 employing the MutMap and QTL-seq technologies in the near future.

5. Conclusion

In the present study, two chemical mutagenesis M2 populations of soybeans, PI 437654 and Zhonghuang 13,were developed, showing many types of signiflcant morphological phenotypes. Thirteen mutants were forward genetically identifled and conflrmed from 400 mutants of PI 437654 M2 population that their SCN race 4-infection phenotype was altered compared to the wild-type PI 437654. Furthermore, all those 13 identifled mutants were still resistant to SCN race 3 like the wild-type PI 437654; and their genomic sequences of three known SCN-resistant genes,GmSHMT08,GmSNAP18andGmSNAP11, were not changed compared to the wild-type PI 437654. Put together, those 13 identifled mutants carry the mutations of gene(s) contributing to the resistance to SCN race 4 in PI 437654, and they can be potentially used as the new genetic soybean sources to map and identify the gene(s) conferring resistance to SCN race 4.

Acknowledgements

This work was flnancially supported by the Innovation Program and Youth Elite Program of Chinese Academy of Agricultural Sciences and the Special Fund for Agro-scientiflc Research in the Public Interest of China (201503114). We thank the colleagues and students in our lab and the workers in the Langfang Experimental Base of Institute of Plant Protection of Chinese Academy of Agricultural Sciences for their kind help and assistance in development of the PI 437654 and Zhonghuang 13 mutants populations in the fleld.