Updated October 11, 2018:
Final Progress Report
Objective 1.1: Develop and evaluate germplasm with new combinations of resistance genes in high yielding backgrounds.
Diers, Scaboo and Nguyen groups have made progress towards diversifying the sources of SCN resistance in soybean germplasm and varieties. The Diers program has entered in the 2018 SCN regional tests two MG III and two MG IV lines that have two SCN resistance genes from G. soja stacked with Rhg1 from PI88788 and five MG III lines and one MG IV line with resistance from both PI 88788 and PI 437654. These tests are in the field and we are awaiting results. In addition, the Diers program has commercialized a variety that has Rhg1 combined with the two resistance genes from G. soja. The Scaboo program is developing MG III and MG IV lines with resistance to HG 1.2.5.7 by using PI 88788, PI 437654, and PI 90763 sources of resistance. We anticipate the first yield-testing of advanced lines during the summer of 2020. The Nguyen program developed ten BC2F2:5 populations carrying up to three resistance genes using sources of resistance from PI 437654, PI 88788 and PI 567516C. To confirm gene stacking and establish contribution of each gene in SCN resistance, all experimental lines are currently being phenotyped for a second time against six SCN races. We anticipate manuscript preparation during the Fall of 2019.
Objective 1.2: Determine resistance gene copy number in the experimental lines for more effective breeding.
During the project, the Diers lab completed Rhg1 copy number evaluations of almost 400 entries in the 2015 and 2016 SCN regional test. Several outliers that have a copy number that do not match expectations based on resistance ratings were identified and will be tested further. In general, the results showed that SCN resistance levels can be predicted by combining the copy number results with the identification of alleles present at Rhg1 and Rhg4. In 2016, Nguyen Lab genotyped 34 experimental lines from the 2015 SCN Regional Tests for the presence of the resistant locus Rhg4. Out of these, eight lines have been selected. In 2017, Nguyen Lab tested 172 experimental lines from the 2016 SCN Regional Tests using available diagnostic markers to detect types of resistant haplotypes of rhg1 and Rhg4. Among them 123 lines were detected as rhg1 PI 88788-type, 15 lines with rhg1 Peking-type, and 13 lines with Rhg4.
Objective 2.1.1: Improve genome assembly
Endosymbiotic bacterial genome
The 1.2 mega-basepair genomic sequence of the soybean cyst nematode’s endosymbiont Candidatus Cardinium hertigii was assembled, annotated, and reported in Genome Announcements (Showmaker et al 2018). This was the first reported genome of an endosymbiont directly sequenced from a plant-parasitic nematode. The genome will not only be helpful for understanding the biology of the endosymbiont, but also for ensuring that the endosymbiont sequence does not contaminate future sequencing projects aimed at the nematode.
Nematode genome sequencing and assembly
Whole-genome sequencing and assembly of multiple strains has been proceeding quickly. As Oxford Nanopore basecalling technology has rapidly improved, the TN10 assembly generated in Spring 2018 with the Oxford technology was revisited. Raw signal files were re-basecalled with the most recent version of the software Albacore 2.3. The reads were assembled using the same methodology as previously reported: error-correction in canu 1.7, assembly in SMARTdenovo, error-polishing with Nanopolish 0.9, and final iterative error-polishing with pilon using Illumina short reads. Parameters in pilon were adjusted to avoid spurious collapsing of tandem repeats in the genome. The new TN10 assembly summed to 132Mb, slightly larger than the assembly produced with the previously available basecaller. The scaffold N50 also slightly increased to 222kb with BUSCO v.3 ortholog completeness scores unchanged at 63% using available Nematoda lineages.
Oxford Nanopore sequencing of TN7 (Hg type 2.5.7), TN22 (Hg type 1.2.5.7), TN8 (Hg type 1.2.3.6.7), OP50 (Hg type 1.2.3.5.6.7), and TN20 (Hg type 1.2.3.4.5.6.7) has been completed. Initial assemblies were produced by the same methodology as described above with metrics similar to that of TN10. With the improvement in TN10 metrics using Albacore 2.3 basecalls, the raw signal files for the above 5 strains were re-basecalled with guppy 1.5.1 using GPU installed on the new Oxford Nanopore GridION instrument. After correcting, assembling, and polishing the guppy 1.5.1 basecalled reads, all assemblies summed to larger sizes even though the increased stringency of the basecaller output led to fewer reads with a smaller nucleotide yield. Assemblies ranged from 112.5Mb to 123Mb, depending on strain. Scaffold N50 improved in some strains while declining in others (134kb - 301kb), possibly reflecting breaks in spurious joins in the previous assemblies. BUSCO v.3 scores remained stable, and all strains have fewer genes in the "missing" category compared to the PacBio/Dovetail assembly. The KAT kmer spectrum analysis tool supports improved assemblies using either Albacore 2.3 or guppy 1.5.1 basecalled reads compared to earlier basecallers.
Nematode genome annotation
We have started annotating these assemblies, starting with repeat analysis. Preliminary, repeat modeling (RepeatModeler) and repeat quantification (RepeatMasker) of the UIUC sequenced SCN strains identified the genomic repeat content ranging from 28.76 to 31.77 %; comparable to the 34% repeat reported for the SCN reference genome (Masonbrink 2018). Alignments of previously reported SCN RNAseq reads (Gardner et al 2018) resulted in an alignment rate range of 89.61% to 92.30%, this is higher than the alignment rate to the current reference TN10 Dovetail assembly (87.13%). The RNAseq alignment rate shows that more genetic information was likely captured in the Nanopore sequencing than the PacBio method used by the reference genome. The repeat and RNAseq read information are currently being incorporated into the gene prediction models for the UIUC sequenced strains using the Breaker2 gene prediction pipeline. Future work will include functional annotation, effector mining, and comparative genomics.
Objective 2.1.2: Genome curation and annotation.
The SCN genome was used to explore effector duplicity and variation within the sequenced population. A total of 431 putative effectors were identified in the genome using four distinct methods: direct alignment of 80 previously identified H. glycines effectors, MEME motif finding among these 80 with subsequent queries of the motifs to all genomic genes with FIMO, and genes with DOG box enrichment in promoters. Of these predicted effectors, 216 have a predicted signal peptide and lack a transmembrane domain. This incongruency may illustrate both the birth and death of effectors. We also identified 161 genes that we can say with high confidence were acquired by SCN through horizontal gene transfer from other species. Seven of these genes coincide with the putative effectors identified above. Finally, we believe that transposable elements and tandem duplications may contribute to effector expansion in H. glycines and other nematodes.
A genome web portal has been created at SCNBase.org using a GMOD instance of Tripal. This site houses the H. glycines genome, a plethora of annotation information, as well as tabs that address website curators/collaborators, current collaborative research, tutorials on SCNBase use, downloadable files, and tools. The primary focus has been to functionally characterize the H. glycines genome and populate SCNBase’s tools with this annotation information. The BLAST tool allows all four types of blast to both the genome and predicted proteome. The JBrowse tool is used to explore the genome at the nucleotide level and has been populated with many tracks that include gene predictions (genomic gene predictions, mitochondrial gene preditions, SignalP secretion predictions, multiple effector predictions, horizontal gene transfers), multiple expression tracks (predicted transcripts, PA3 transcriptome, isoseq transcriptome, NCBI ESTs, all raw RNAseq, all raw isoseq), multiple repeat predictions (RepeatModeler repeats, Helitrons, LTR retrotransposons, DNA transposons, tandemly duplicated loci), genomic data alignments (raw and assembled PacBio reads and NCBI’s nucleotide sequences for H. glycines), population-level structural data with SNPs and indels from 15 H. glycines populations, syntenic relationships with multiple species (Globodera ellingtonae, Globodera rostochiensis, Globodera pallida, Meloidogyne hapla, Meloidogyne incognita, and an early H. glycines assembly (PRJNA28939)). A third tool, feature search, allows for the retrieval of functional annotation and sequence information for each gene and predicted transcript in the genome. All data is downloadable with a direct link on the SCNBase.org homepage, all of which is easily accessible through a search tool that is heavily integrated throughout the entire site.
A number of studies are ongoing that will be published and included in SCNBase. Some of the current work involves dissecting genes associated with Hg-type using variation found in fifteen populations of H. glycines. To better understand the expansion and contraction of gene families among plant-parasitic nematodes, a comparative study among all sequenced plant-parasitic nematodes may identify gene families important to specific nematode pest clades.
Objective 2.2: Conduct comparative population studies to identify genes associated with SCN virulence and evaluate utility as novel resistance targets.
A series of virulent SCN populations differing in HG type were provided for genome sequencing (above) and a de novo assembly of the SCN virulent transcriptome was completed. Comparative analysis between virulent populations on resistant soybean identified differences in gene expression and polymorphisms between populations. These expression changes and polymorphisms are being pursued for validation and use as potential candidates for virulence markers. In addition, the functional impact of effector variants unique to virulent populations is being assessed to determine importance to virulence. SCN populations (below) experimentally selected on a series of single resistance sources provides a new set of genetic material for genomic selection studies to identify the genes associated with SCN virulence.
Objective 2.3: Determine unique resistance gene stacks that would be beneficial in rotations to enhance durability of SCN resistance.
The Universities of Illinois and Missouri tested a population of 95 experimental lines that segregated for rhg1 from PI 88788, two SCN resistance alleles from G. soja, and one from PI 567516C. Across six environments of testing, the only resistance gene that had a significant, positive effect on yield was one of the G. soja resistance genes. More work is needed to understand why a positive impact on yield was not observed for the other genes.
During 2018, yield and agronomic testing was initiated for a population of 134 experimental lines that segregates for the SCN resistance genes rhg1 and rhg4 from PI 437654, two SCN resistance genes from G. soja, and one from PI 567516C. This population is being grown in two environments in both Illinois and Missouri during 2018. These tests will provide more information about the impact of SCN resistance genes on yield and agronomic traits. To determine the impact of resistance genes on SCN populations, the Nguyen Lab developed experimental lines that segregate for rhg1 PI 88788-type (elite line SA12-1455) and Hartwig-type (PI 437654), Rhg4 and qSCN11 locus (PI 437654), and qSCN10 and qSCN18 loci (PI 567516C). The preliminary results suggest that Hartwig-type rhg1 with qSCN10 and qSCN18 significantly increases resistance to SCN race 1. Also, PI 88788-type rhg1 with Rhg4 and qSCN11 increased resistance to race 1. We observed that qSCN11 might be responsible for resistance to race 2. PI 88788-type rhg1 with Rhg4 and qSCN11 increased resistance to race 3. PI 88788-type rhg1 with qSCN10 and qSCN18 increased resistance to race 1, 2, 2, 5, 14, and LY1; and therefore, these lines seem to have the highest potential. Seeds will be increased in winter nursery. The lines will be tested for yield drag in 2019.
Limited options for rotating resistance sources are available, however, genes from new sources such as PI 468916 (cqSCN-006 and cqSCN-007) and PI 567516C (chromosome 10 QTL) have been identified. One of our objectives was to evaluate unique resistance gene stacks that have these new resistance sources to determine what would be beneficial in rotation to combat the increase in virulent SCN and limit nematode adaptation to resistant cultivars. In a greenhouse experiment, eight SCN populations were developed by selecting an SCN field population (HG type 1.2.5.7) on a single resistance source or on a rotation of resistance gene stacks. Egg density data were collected after each generation and HG type tests were conducted after eight generations. Continuous use of rhg1-b (887) or 006/007 (468) had limited effectiveness for reducing SCN density. Continuous use of rhg1-a/Rhg4 resistance reduced SCN density but retained virulence on PI 88788-type resistance and selected for increased SCN virulence (HG type 1.2.3.5.6.7) on Peking-type resistance. Rotation of rhg1-a/Rhg4 and a stack of rhg1-b (887) + 006/007 (468) + CHR 10 (567) was the most effective at reducing population density and minimized selection pressure. Our results provide important information for the implementation of a strategic rotation plan to effectively manage the widespread virulence on PI 88788 and enhance durability of SCN resistance.
Objective 3: Coordinate the testing of publicly developed SCN resistant experimental lines.
During 2018, yield tests of SCN resistant soybean experimental lines and varieties were organized and are being grown in 39 locations in 10 north central states and Canada. The tests include 9 MG 00, 16 MG 0, 24 MG I, 61 MG II, 46 MG III, and 26 MG IV experimental lines and check varieties. The tests were planted in spring 2018 and were maintained through the summer and data are being taken on the date of maturity, plant height, and lodging and the plots are being harvested to estimate seed yield. In addition, soil samples were taken from field locations and the number of nematode eggs/100 cc of soil and HG type of the nematodes is being determined for each location. The SCN resistance level of each experimental lines is also being determined with two SCN isolates. The data from these environments will be analyzed, summarized, and the distributed as a printed test report and also posted online.
Over the period of last three years, our group has presented our results in the form of oral presentation, poster presentation and peer reviewed research articles. The details of those are attached in a separate file labeled as "NCSRP_result distribution".
View uploaded report 
For this project, we followed a two-pronged strategy. In our first approach, we worked on diversifying resistance resources against the soybean cyst nematode (SCN). The long-term aim of this approach is to generate a broader spectrum of soybean cultivars with robust SCN resistance. As a part of this project, we coordinated multi-site testing of SCN-resistant soybean lines. SCN resistance testing conducted in different field locations was very impactful as it gauged how these soybean lines resist different nematode populations, how they performed in different soil conditions and under different environments. The development and commercialization of a soybean variety with the Rhg1 resistance locus combined with two additional resistance genes from G. soja was one highlight of this particular approach. In our second approach, we are studying the molecular basis of nematode virulence. Nematodes overcome plant resistance by delivering protein molecules into the plant that molecularly interfere with the plant’s ability to mount a defense reaction. Since some nematode populations are better at overcoming soybean resistance than others, it is likely that these populations deliver different, more powerful proteins. The long-term goal of this project is to identify the differences within the proteins delivered by the virulent and non-virulent SCN populations. These differences are encoded within the genetic code of the nematodes. Thus, analyzing the genetic code of SCN populations differing in virulence will help us identify the proteins that SCN delivers to overcome soybean resistance. We generated and compared such information from virulent and non-virulent nematode populations. From these analyses, we generated large amount of data, which we are currently analyzing. We also developed an online portal on which we have uploaded all data. This portal will ensure free and global access of this information. Overall, this project has brought together a powerful group of university scientist that work collaboratively with the goal to empower farmers in managing this pest.