BGRI 2018 Program

Widening the genetic basis of leaf rust resistance is a primary objective of durum wheat breeding programs in North America, especially after the occurrence of new Puccinia triticina races in Mexico and southwestern United States. This study was conducted to characterize and map uncharacterized genes for leaf rust resistance in durum wheat sources and to develop reliable molecular markers for marker-assisted breeding. Seven RIL populations involving crosses of resistant cultivars Geromtel_3 (ICARDA), Tunsyr_2 (ICARDA), Amria (Morocco), Byblos (France), Gaza (Middle East), Saragolla (Italy) and Arnacoris (France) with the susceptible line ATRED #2, were evaluated for reaction to Mexican Pt race BBG/BP. Selective genotyping and bulked segregant analysis using the wheat 90K SNP array identified two genomic regions for leaf rust resistance; 6BS in Geromtel_3 and Tunsyr_2, and 7BL in Amria and Byblos. Composite interval mapping revealed a total of six quantitative trait loci (QTL) for leaf rust resistance; two on chromosomes 6BS and 6BL in Gaza, and major QTL on chromosomes 7BL in Arnacoris and 2BS in Saragolla. Both latter lines had an additional minor QTL on chromosome 1BL. DNA sequences associated with the SNP markers linked to the Lr genes and QTL were anchored to the wild emmer wheat reference sequence of accession Zavitan. KASP markers tightly linked to the Lr genes reported here were developed.

Despite over 70 designated yellow rust resistance genes (Yr) in wheat, few have been cloned. This lack of knowledge hinders efficient marker assisted breeding and exploitation of novel allelic variation. We recently exome sequenced a mutant population of UK cultivar Cadenza which carries Yr7. Screening 1,000 mutagenized individuals with a Yr7-avirulent Pst isolate identified seven susceptible lines presumed to carry mutations in Yr7. To test this, mutational resistance gene enrichment sequencing (MutRenSeq) was carried out on the susceptible lines and a candidate for Yr7 was identified. SNP markers were developed to assess their association with susceptible phenotypes in F2 populations derived from crosses of the susceptible mutant lines to wild type Cadenza. A similar approach was carried out on Yr5 susceptible mutants in the Lemhi-Yr5 near isogenic line background. A candidate gene was also identified for Yr5, which was different from the Yr7 candidate. The parallel Australian work identified susceptible mutants in AvS-Yr7 and AvS-Yr5 and respective candidates using MutRenSeq with similar outcomes. This suggests that Yr5 and Yr7 are very closely linked genes rather than true alleles. We will discuss how distinguishing linkage versus allelic relationships between genes might affect breeding.

The recent release of high-quality sequence information from hexaploid wheat (IWGSC 2014) coupled with the availability of high-density consensus maps for tetraploid wheat (Maccaferri et al. 2014) has accelerated marker and gene discovery in durum wheat, thus facilitating the genetic dissection of agronomic traits. Radiation hybrid (RH) mapping is a promising recombination-independent mapping approach, which involves the use of radiation-induced chromosomal breakage and marker segregation to reconstruct marker order (Tiwari et al. 2016). In this study, for the first time, a RH panel for tetraploid wheat was developed for the reference durum genotype Svevo: the Svevo-Whole-Genome Radiation Hybrid (Sv-WGRH) panel. The Sv-WGRH panel was developed at Kansas State University (USA) according to the protocol reported by Tiwari et al. (2016). Freshly dehiscing pollen of Svevo was irradiated with y-rays (10-Gy) and this was used to pollinate ~150 emasculated spikes of Senatore Cappelli (used as the female parent), producing ~1,300 RH1 seeds, each representing an independent RH event. Greenhouse planting of 1,000 RH1 seeds resulted in ~730 RH1 plants, each representing a RH line of the Sv-WGRH panel. Initial assessment of the panel was performed on the DNA of 613 RH1 plants by means of 23 SSR markers. Results indicated that average marker retention frequency of the panel is ~87%, with 35% of RH1 lines having a retention frequency between 20 and 90%. Approximately175 RH1 plants were obtained from planting 269 RH1 seeds in the greenhouse. DNA of these RH1 lines will be genotyped by GBS in order to construct a physical map. This RH panel is an important resource that will contribute to the assembly of the genome sequence of durum and other tetraploid wheats and will support positional cloning of important genes and QTLs in durum wheat.

New Pst races documented within last two decades have caused severe yield losses worldwide. Genetic bottlenecks associated with wheat polyploidization, domestication and initial selection in agroecosystems decreased wheat genetic diversity and increased its vulnerability to biotic and abiotic stresses. Wild emmer (Triticum turgidum ssp. dicoccoides, DIC), the tetraploid progenitor of common wheat distributed along the Fertile Crescent has valuable residual adaptive diversity in response to diseases, including stripe rust. Yr15, a dominant DIC gene located on chromosome 1BS confers broad-spectrum resistance to stripe rust. Yr15 was incorporated into durum and common wheat cultivars and lines in research and breeding programs in Israel, Europe, Australia, USA, China and India. Comparative genomics, chromosome walking, BAC libraries (DIC and bread wheat), whole genome assemblies, EMS mutagenesis and transgenic approaches enabled us to clone Yr15 and validate its function. The Yr15 protein has a novel structure for R-genes in wheat with two tandem kinase-like domains that are both essential for resistance. Macro- and microscopic observations of fungal development and accumulation of biomass suggest that the hypersensitive response plays a central role in the resistance mechanism. Non-functional alleles of Yr15 in T. dicoccoides, T. durum and T. aestivum differ from the functional allele of DIC G25 by indels, creating truncated proteins. We designed diagnostic markers that differentiate between functional and non-functional Yr15 alleles. Among 85 durum and common wheat accessions only Yr15 introgression lines contained the functional allele, whereas all others contained non-functional alleles. These results suggest that Yr15 has the potential to improve stripe rust resistance in a wide range of tetraploid and hexaploid wheat germplasm. The absence of the functional Yr15 in tested durum and common wheat varieties highlights the value of DIC germplasm as a reservoir of resistance genes for wheat.

The genetic diversity of wheat has been eroded by human selection during domestication rendering the crop more susceptible to disease. In contrast, the wild progenitors and relatives of wheat represent an important source of genetically diverse disease resistance (R) genes that can be used to improve resistance in cultivated wheat. Most often, R genes encode nucleotide binding and leucine-rich repeat (NLR) immune receptors. We have devised a high-throughput NLR annotation tool and used NLR resistance gene enrichment sequencing (RenSeq), to explore the genetic diversity of NLRs in Aegilops tauschii, the diploid D genome progenitor of hexaploid bread wheat. The sequenced reference accession contains 732 full length NLRs, and we find significant NLR variation between accessions. We will report on our progress to characterize this genetic variation in a panel of 150 geographically and genetically diverse Ae. tauschii accessions, and in associating this to phenotypic variation for rust resistance with the aim of speeding up the discovery and cloning of new functional R genes.

The term "nonhost" resistance (NHR) suggests a resistance apart from host resistance. Evidence now suggests that this is not the case, the same genetic components that confer host resistance have a primary role in NHR, although the regulation of those genetic components is both pathosystem and host-species specific. In this presentation we will explore the potential of NHR for host resistance breeding, looking at the potential of individual "NHR genes", as well as the story told by global transcriptomics responses. Large-scale transcriptomics studies comparing wheat to barley, and three different pathosystems (leaf rust, powdery mildew and cereal blast) indicated that the NHR-related responses were to a great extent pathosystem-specific, the genes differentially expressed between host and nonhost interactions showed only small overlaps between the three cereal pathosystems. This pathosystem-specific reprogramming may reflect different resistance mechanisms operating against non-adapted pathogens with different lifestyles, or equally, different co-option of the hosts by the adapted isolates to create an environment optimal for infection. Comparison of wheat and barley orthologues differentially expressed between host and nonhost interactions also revealed few commonalities, rather suggesting different nonhost responses in the two cereal species. Taken together, our results suggest independent co-evolutionary forces acting on different pathosystems, mirrored by barley- or wheat-specific nonhost responses. At the individual gene level three barley receptor-like kinases (RLKs) have been identified that confer quantitative resistance to wheat powdery mildew when transiently over-expressed in wheat, as reflected by reduced establishment of haustoria in epidermal cells. We will give an update on the phenotypes observed when these HvRLK genes are expressed in wheat transgenics.

Yellow (stripe) rust caused by Puccinia striiformis fsp tritici is one of the most devastating diseases threatening wheat in Ethiopia. The identification of genetic resistance sources help combat the threat. Genome wide association study of yellow rust resistance was conducted on 300 durum wheat accessions comprising 261 landraces and 39 registered varieties. The accessions were evaluated for their field resistance in alpha lattice design (10 X 30) in two replications at Meraro, Kulumsa and Chefe-Donsa in 2015 and 2016 main growing seasons. Scoring was carried out using a modified Cobb scale (0 -100% severity combined with Field responses: Immune, Resistant, Moderately Resistant, Moderately Susceptible and Susceptible) and then converted to Coefficient of Infection (CI). Genotyping was done using 35k SNP chip at Genomic Facility of Bristol University. Analysis of the genotyping data with Axiom Analysis Suit v.2 resulted in a total of 8,682 polymorphic SNPs of which 6,969 satisfied the marker allele frequency (MAF) >= 5% threshold and were used in subsequent analyses. Population structure analysis suggested five sub populations in which the released varieties clearly stood out separately from the landraces. An earlier decay of Linkage disequilibrium (LD) was depicted from LD analysis in TASSEL. Nine SNPs significantly associated with candidate resistant QTLs were identified on chromosome 1A, 1B and 5B via association analysis of Combined CI data of Meraro 2015, 2016 and Kulumsa 2016 using GAPIT in R. Further investigation on approximately 10 million (bp) genomic regions encompassing the identified loci indicated the presence of some disease resistance protein (NBS-LRR class) family and RPM1 far from the loci. This study provides SNPs for tracking the QTL associated with resistance in durum wheat improvement programs. However, the diagnostic value of the identified linked SNPs needs further investigation in independent germplasm.

Breeding for durable resistance to rust diseases in bread wheat (Triticum aestivum L.) relies largely on continually introgressing new resistance genes into adapted varieties, often from wild relatives due to low genetic diversity in the wheat genomes. The process of integrating genes from a different species is not only time consuming, but also not always completely successful in eliminating unfavorable traits brought in by linkage drag. Developing new resistant germplasms in wheat provides a great advantage to overcome the aforementioned challenges. The focus of resistance breeding over the decades has been on using resistance genes or genes that can enhance the host defense response, not enough attention has been given to the host genes targeted by the pathogen effectors to enhance the host susceptibility. Our study was conducted to identify the wheat genes targeted by rust pathogens. In this study, we identified MYC4 transcription factors (TFs) located on the wheat chromosomes 1A, 1B and 1D were upregulated from 1-8 days post- rust inoculation in a susceptible wheat line but not in a resistant wheat line. Down regulation of MYC4 TFs using barley stripe mosaic virus-induced gene silencing (BSMV-VIGS) in the susceptible cultivar Chinese Spring enhanced its resistance to stem rust pathogen. Knock-out of MYC4 TFs from A and B genomes in Cadenza by Ethyl methanesulfonate mutagenesis render new resistance to stem rust pathogen. From this discovery, we have created new germplasms in wheat via modifications of wheat MYC4 transcription factors.

The wheat yellow rust pathogen Puccinia striiformis has been hypothesized to have its centre of origin in the Himalayan region, where the pathogen is highly diverse. However, the detection of pathogen diversity on same host varieties at a given location and across locations still needs to be assessed. In present study we attempted to assess the genetic diversity in P. striiformis sampled on the same host variety at multiple locations across Pakistan. The recently released high yielding but highly susceptible variety Galaxy was cultivated throughout the country during 2016. The variety was scored and sampled in 46 locations from 24 districts across three provinces of Pakistan. Out of 46 locations surveyed the variety was highly infected (>60S score) at 30 locations across the country, with 16 locations at Khyber Pakhtunkhwa province, 12 at Punjab province and 2 at Sindh province. Microsatellite data of a total of 90 samples collected from these 46 locations on Galaxy-2013, resulted in the detection of 84 multilocus genotypes (MLGs) from at least five genetic groups of P. striiformis, as identified through population genetics analyses. At least two genetic lineages were detected at a given district on the same variety, when assessed across all 24 districts. Similarly, multiple MLGs were also detected on Galaxy-2013 at a given district. This could be expected from the high diversity and recombinant structure of P. striiformis in Pakistan, where isolates from distinct genetic groups carrying virulence to a given variety could infect the host. The pattern, however, needs to be confirmed on other susceptible varieties present at multiple locations and over temporal scale.

Each year new races of P. triticina are found in the field and arise from selection pressure due to the monoculture of just a few wheat cultivars. It is assumed that changes in effectors in the rust, caused by deletions, nucleotide changes, amino acid changes, and expression differences are the causal agents of the virulence shifts. The P. triticina genome has been published and almost 1400 secreted effector candidates were identified. Crosses are difficult to make in the rust, so a mutagenesis approach was employed to identify specific effectors recognized by Lr resistance genes. Purified spores of BBBD Race 1 were exposed to fast neutrons and inoculated onto the Thatcher Lr2A and Lr2C isolines. Typically, BBBD induces a 0; response on these lines, however, pustules that formed were assumed to be mutants. Two rounds of single pustule culture, under selection on the specific isoline were used to increase the spores. Seven mutants each were isolated for virulence to Lr2A and Lr2C. DNA has been isolated and sequenced. Effector gene candidates are being identified and will be reported.

Durum wheat (Triticum turgidum L. ssp. durum) producers can experience significant yield and grain losses due to crown rot (CR) disease, caused primarily by the fungal pathogen, Fusarium pseudograminearum. Losses due to CR are exacerbated when disease infection coincides with terminal drought. Durum wheat is very susceptible to CR and resistant germplasm is not currently available in elite breeding pools. Deploying physiological traits for drought adaption (e.g. deeper roots), to reduce stress due to water deficit may, therefore, potentially minimise losses due to CR infection. The rapid generation advance technology, "speed breeding", was used to rapidly develop recombinant inbred lines (RIL) populations (F6) derived from crosses between Australian cultivars and ICARDA elite breeding lines pre-selected for drought adaptation in Syria and Morocco. Populations were evaluated in the field and under controlled conditions for several physiological traits, including seminal root angle and number and CR severity. This provided the genetic predisposition of lines for rooting behaviour and CR susceptibility in the absence of water stress. Field experiments were established in Queensland, Australia, which allowed an examination of the value of root development traits to improve adaptation to each of the stresses. NDVI measurements were recorded weekly, which enabled modelling of the senescence pattern and calculation of stay-green traits for each genotype. Genome-wide association studies using DArT markers identified key genomic regions underpinning the traits. Our genetic analyses highlighted the genetic relationships between yield as well as above- and below-ground physiological traits. Through this study, we have provided new insights into the genetic controls and value of these traits, which we anticipate will assist breeders to design improved durum varieties that may mitigate production losses due to water deficit and CR.

As the primary interface for resource acquisition, plant roots play a key role in growth regulation. Evidence from rice, maize and sorghum demonstrates that the below-ground plant architecture significantly impacts plant performance under abiotic constraints. Roots assume critical functions in water uptake, nutrient acquisition and anchorage, an essential characteristic to maintain plant stability under increased grain load. Despite their fundamental importance, knowledge about genetic control of root growth in major grain crops is limited and very little is known about interactions between below-ground and above-ground plant development. Here we demonstrate that VERNALIZATION1 (VRN1), a key regulator of flowering behavior in cereals, also modulates root architecture in wheat and barley. Associations of VRN1 haplotypes to root growth habit were discovered in wheat by genome-wide association studies, and confirmed by allelic analyses in wheat and barley populations. Functional characterization in transgenic barley confirmed that VRN1 influences root growth angle directly, via gravitropism. These discoveries provide unexpected insight into underground functions of a major player in the well-characterized flowering pathway, revealing the intersection of above-ground gene regulation with the largely unexplored genetic architecture of plant root development. Understanding the pleiotropic involvement of this key developmental gene in overall plant architecture will help to breed cereal cultivars adapted to specific environmental scenarios.

Genome-wide association studies (GWAS) are useful tool for accelerating crop improvement and rate of genetic gain via enhanced knowledge of marker-trait associations. Synthetic hexaploid wheat (Triticum aestivum) are used as a means of introducing novel genetic variation into bread wheat, however, reports on genome wide association studies (marker-trait associations) remain limited in synthetic hexaploid wheat under drought stressed conditions. GWAS for grain yield and yield-related traits were conducted using a diverse panel of 126 primary synthetic wheats and two bread-wheat (checks) genotypes under drought stressed conditions across two growing seasons (2016 and 2017). The population was genotyped using genotype-by-sequencing (GBS) approach and 37,592 high quality SNPs were used in GWAS analysis. Analysis of variance showed significant genotype effects for days to heading and maturity, grain filling period, plant height, grain and biomass yield, grain protein content, thousand grain weight, grain and spike harvest index. A number of genotypes were superior compared to checks for yield (seven entries in 2016 and four entries in 2017) and yield related traits. Markers associated with multiple traits observed on chromosomes 1A, 1B, 1D, 2D, 3A, 3B, 3D, 4A, 4D, 5A, 5B, 6A, 6B,7A, 7B, and 7D from a GWAS model that accounted for kinship and population structure. These markers associated with yield and yield related traits will be validated in a different population and will be converted to KASP markers that can be used in wheat breeding programs to improve grain yield under drought stressed conditions. Additionally, these lines showed resistance several biotic stresses such as stem rust (21 entries), leaf rust (80), stripe rust (13), common bunt (46), cereal cyst nematode (11), crown root rot (3), and barley yellow dwarf virus (15). Therefore, these synthetic lines are potential source for integration of resistance genes into high yielding modern wheat germplasm.

Durum wheat is one of the most important crops in Mediterranean countries, where drought conditions often occur. Therefore, optimisation of root system architecture (RSA) is an important objective for enhancing yield and the sustainability of durum wheat grown under different water regimes. In the present study, linkage and association mapping (AM; panel of 183 elite accessions; Maccaferri et al. 2005) for RSA evaluated at the seedling stage evidenced 20 clusters of quantitative trait loci (QTLs) for root length and number as well as 30 QTLs for root growth angle (RGA). The most divergent RGA phenotypes observed by seminal root screening were validated by root phenotyping of field-grown adult plants. QTLs were mapped based on the Illumina 90K SNPs profiles, thus allowing cross-referencing of RSA QTLs between durum and bread wheat. QTL analysis of RSA and grain yield data indicates root growth angle (RGA) as a valuable target to enhance grain yield and yield stability across different soil moisture regimes (Maccaferri et al. 2016). Based on their relative additive effects, allelic distribution in the AM panel and co-location with QTLs for yield, eight RGA QTLs have been prioritised in terms of breeding interest and value. These QTLs were investigated for gene content based on the chromosomal pseudomolecules of Chinese Spring T. aestivum, Zavitan T. dicoccoides genome assembly (Distelfeld et al. 2017) and Svevo durum genome. The chromosome regions contained 25 to 242 predicted genes (123 on average). In six RGA QTLs, from one to four gene annotations were involved in auxin pathways. The comparison between the three assemblies indicates their high quality and usefulness to identify candidates to explore the polymorphism and the structural variation present in the A and B wheat genomes.

Phenotyping is the cornerstone of plant breeding but until the advent of remote sensing (RS) technologies, selection was largely restricted to relatively heritable traits like phenology and a few essential complex traits including end-use quality and yield per se. However, RS permits integrative traits -that by definition are genetically complex- to be evaluated on a scale that overcomes some of the disadvantages of their relatively low heritability. For example, canopy temperature is a good predictor of yield and root function under abiotic stress, and a number of spectral indices –including NDVI and water index- can be used to measure other physiological and performance traits like biomass and stay-green. Our data show that aerial phenotyping platforms permit physiological traits to be measured with increased precision resulting in better prediction of yield compared with ground-based sensing approaches. Additional traits that lend themselves to aerial high throughput screening include photosynthetic pigments and photo-inhibition using indices in the visible spectrum, while water-sensitive indices in infra-red (IR) and near IR bands are associated with hydration status of tissue. Many of the RS indices mentioned, by virtue of being high throughput, lend themselves well to large-scale genetic resource screening as well as genetic analysis. However, integrative traits have less power to predict performance under high yield environments. Here, physiological breeding approaches focus more on the characterization of potential parents in order to design complementary crosses. We have developed conceptual models for different mega-environments that consider a range of traits for which there is evidence of useful unexplored genetic diversity. One novel example of such a trait is spike photosynthesis (SPS) which is difficult to measure and as a result has been largely ignored in breeding, despite the fact that spikes intercept up to half of the radiation during grain-filling. The trait is also interesting because spikes have extremely high water use efficiency and are often the only organs that remain green under post-anthesis stress. Three innovations have been applied to better characterize SPS including a 360 degree LED illumination chamber for gas exchange measurement, long duration spike shading treatments, and use of stable carbon isotopes analysis of grain, all of which help estimate the integrated contribution of SPS to yield, leading to identification of useful genetic sources for crossing as well as novel QTL for SPS. An innovative crossing strategy -involving complex traits- has recently demonstrated genetic gains compared to local checks, and in several cases CIMMYT elite checks, in international yield trials. The crosses are designed to combine complementary traits, specifically sources of high biomass (source) with good sources of spike fertility (sink). The rationale is to translate expression of high radiation use efficiency -often coming from exotics like primary synthetics and landraces- into yield by crossing with elite lines that optimize partitioning of biomass to grain. This pre-breeding exercise has already provided novel genetic sources of yield potential and heat tolerance to national programs, broadening the genetic basis of their crossing blocks, and in some cases delivering candidates for release, for example in Pakistan.

Wheat is the main food crop in Central and West Asia and North Africa (CWANA) regions with average consumption of 200 kg/capita/year. However, its productivity is very low (2.5 t/ha) mainly due to drought, heat and stripe rust. The wheat breeding program at ICARDA applies both conventional and molecular breeding tools following inter-countries shuttle breeding and key locations yield testing approaches to combine yield potential and wide adaptation with resistance to biotic and abiotic stresses. Following this approach, climate resilient wheat varieties combining high yield potential, rust resistance with heat and drought tolerance have been identified. Using marker assisted selection, major genes have been pyramided into adapted varieties with minor gene background for rust resistance. Molecular markers linked to heat tolerance and yellow rust resistances have been identified. Yield levels of some of these genotypes range up to 6t/ha at Wadmedani station of Sudan under extreme heat stress, 7t/ha at Merchouch station of Morocco under terminal moisture stress (260 -300 mm) and 11 t/ha at Sids station in Egypt under optimum conditions. Pedigree analysis showed that resistance sources for heat and drought tolerance in such elite germplasm were introgressed from synthetic wheats and wild relatives mainly T. dicoccoides. On an annual basis, ICARDA’s spring bread wheat program composes and distributes about 400 elite set of these genotypes through international nurseries and yield trials to countries in the CWANA and SSA regions and beyond upon request for potential direct release/or parentage purposes. In the last 5 years, more than 32 spring bread wheat varieties of ICARDA origin have been released by the national programs in the CWANA and SSA regions. It is important to deploy these varieties rapidly with improved crop management technologies to increase wheat production and enhance food and nutritional security.

Over 70% varieties released in Africa, Middle East, Africa and Latin America during the last decade trace to CGIAR germplasm either released directly or used as a parent. A large-scale breeding, utilizing conventional and modern strategies and tools are enabling the genetic yield gains together with resistance to rusts and other diseases, tolerance to drought and heat, end-use and nutritional (high Zn) qualities. Diverse field sites in Mexico and Kenya permit two generations per year shuttle breeding for selection, and managed phenotyping for grain yield and other relevant traits. High throughput phenotyping and genomic selection are under testing and implementation to enhance genetic gains. Utilization of slow rusting, minor but additive genes based adult plant resistance (APR) to leaf rust has been main-streamed in the breeding program, which has resulted in a rapid decline of leaf rust where CIMMYT-derived varieties are grown. About 10-15% germplasm distributed internationally in recent years possesses near-immune APR to races belonging to the Ug99 lineage and being promoted for release in East Africa where high stem rust pressures are common and resistance durability is crucial to reduce crop losses. In addition, 40-50% germplasm carries moderate but adequate levels of APR and 20-30% possesses diverse race-specific resistance genes (R-genes) to Ug99 races. The worldwide spread and fast evolution of aggressive and temperature tolerant lineages of yellow rust fungus has posed a new challenge to breeding APR especially for areas where yellow rust infection initiates in seedling growth stages under prolonged cool and humid conditions. To counter this challenge we have implemented a strategy that allows field based selection of APR genes with small to intermediate effects R-genes for achieving all-stage resistance while maintaining resistance complexity. The superior wheat germplasm should result in releases of new popular varieties; enhance productivity and profitability in targeted regions.

While plateauing yield gains, climate change and biotic stresses are becoming serious constraints to wheat production, integration of innovative cutting-edge technologies in breeding programs is critical for the development of high-yielding wheat varieties with increased climate resiliency and durable disease resistance. One promising technology that could increase genetic gains for quantitative traits and enable simultaneous selection on several desired superior alleles is genomic selection (GS), which uses genome-wide marker information to predict the breeding value of individuals prior to phenotyping, thereby eliminating poor-performing progenies before the next generation of costly field-testing. To evaluate the potential integration of GS as a breeding tool, we tested genomic prediction for grain yield and stem rust in CIMMYT's yield trial (YT) and elite yield trial (EYT) nurseries from 2014-2017. Each of the YTs comprise 9,000 lines, genotyped using genotyping-by-sequencing and phenotyped for response to Ug99 stem rust race in Njoro, Kenya and grain yield in Obregon, Mexico. We observed that the mean within-year prediction accuracies for stem rust and grain yield were 0.6 and 0.5, respectively and the mean across-year/across-nursery prediction accuracies were 0.52 and 0.35, respectively. A subset of the YTs comprising 1,000 lines constitute the EYTs, that were phenotyped for grain yield in different stressed environments in Obregon. The mean within-year prediction accuracies for grain yield in fully-irrigated (IR), severe drought (SD), and late-heat (LH) environments ranged from 0.35 to 0.52, 0.35 to 0.55, and 0.35 to 0.51, respectively and the mean across-year/across-nursery prediction accuracies ranged from -0.07 to 0.32 for IR, -0.12 to 0.25 for SD, and 0.08 to 0.29 for LH. Although some of these results are promising, further efforts on understanding genotype x environment interactions, yield stability and improving prediction accuracies in unrelated populations are crucial for the successful implementation of GS in applied wheat breeding programs.

Wheat is one of the major cereal crops in Ethiopia predominantly grown by small-scale farmers under rainfed condition. Currently, about 4.9 million farmers produce close to 4.5 million tons of wheat across 1.7 million hectares of land with average productivity of 2.6 t/ha. For the last two decades, wheat area production and productivity increased by 38,762.98 ha/year, 0.16 million tons/year and 0.07 tons/ha/year, respectively. It shows increasing scenario of area (92.29%), production (321.61%) and productivity (119.26%). However, the productivity of wheat per unit area is low as compare to world average (3.41 t/ha). And, still the country importing around 30% of the demand from other countries. Several factors contribute for low productivity but wheat rust diseases especially stem and yellow rust has become frontline constraints. In response, the research and development sector have been done a number of efforts. Development and deployment of resistant varieties is one of the key tools to combat frequent outbreak of rust. These could only be achieved by having effective contingency planning for organized fast-track development and release of new varieties and accelerated seed multiplication under pinned by flexible policies, and commitments by national and/or international stakeholders. Currently, this strategy significantly shortens the time lag between variety release and availability of quality seeds directly in to the hands of farmers to combat the rapidly evolving rust problem in Ethiopia. In the late 1990s the average age of varieties being used in farmers’ fields was greater than 14 years. However, after the strategies developed for fast-track variety development and seed multiplication, susceptible wheat varieties were replaced in shortest possible time. Generally, fast-track variety development and pre-release seed multiplication of rust-resistant varieties in Ethiopia is tackling Ethiopia’s wheat rust effect. As a result, it protecting Ethiopian farmers from crop losses and helping them to raise their production, achieve high incomes, and strengthen national food security.

Improving the rate of genetic gains in wheat for grain yield is a challenge that requires integration of traditional breeding methods with emerging technologies in genomics and phenomics. Advances in phenomics has enabled the development of aerial high throughput phenotyping (HTP) systems that can be used for large scale field based characterizing and screening of breeding populations. International Maize and Wheat Improvement Center (CIMMYT) is evaluating the potential of integrating HTP of advanced wheat lines in prediction of grain yield and other traits. Advanced lines (~1000) in the 2nd year of yield testing were evaluated for grain yield and other traits in multiple environments at the Norman E. Borlaug Research Station at Ciudad Obregon, Sonora, Mexico from 2014-2017. Hyperspectral and thermal cameras were used in aerial HTP across the crop cycle. Spectral indices and canopy temperatures were estimated from the images collected during the crop cycle. Advanced lines were also genotyped with genotyping-by-sequencing (GBS) markers. Pedigree, genomic and phenomic data was used for predicting grain yield. Initial results indicate that the spectral indices improve prediction accuracy for grain yield within a year (~0.4 – 0.5), though across years it is considerably lower. Environment or year had significant effects on the phenomic traits. The recent development of genomic selection models that include genotype by environment interaction effect could improve the predictive ability of the spectral traits. Plant height and lodging estimated using aerial imagery showed significant correlation (r = 0.79 for height, r = 0.88 for lodging) with the manual height measurements and lodging scores, respectively.

Genetic gains were assessed in Turkey in 2009-2013 in three groups of wheat germplasm: 1) Fourteen varieties of winter wheat for irrigated conditions (released 1963-2004) tested in 16 environments; 2) Twenty-one1 varieties of winter wheat for dry rainfed conditions (released 1966-2006) tested in 21 environments; and 3) 35 Thirty-five spring wheat varieties (released 1964-2010) tested in 15 environments. Grain yields of group one varieties increased by 58 kg/ha (1.37%) per year, compared to gains of 12.5 kg/ha (0.50%) per year and 30.9 kg/ha (0.62%) per year for groups two and three, respectively. For irrigated winter wheat, gains were mainly achieved by shortening the plant height and increasing harvest index. Yield gains were accompanied by improved stripe- and leaf-rust resistance, primarily utilizing adult plant resistance genes. For dry rainfed varieties, modern varieties had improved yield potential and tolerance to moisture stress. The number of spikes per unit area decreased by 10% from 1966 to 2006, while the number of kernels per spike and thousand kernel weight increased by 10%. Rusts are less important for dryland wheat, but some improvements in resistance were recorded. For spring wheat, genetic yield gains were associated increases in harvest index, number of spikes per unit area, and number of grains/spike. Leaf rust resistance made substantial contributions to yield genetic gains in group three. Productivity gains were reduced by 50% in a trial with fungicide protection, demonstrating the important effect of rust resistance in genetic gain. In all three groups, the newest varieties combined higher yields with yield stability. Grain quality of the new varieties did not deteriorate over time, though most were inferior to superior bread-making quality checks. Breeding strategies should utilize the diverse modern gene pool and genetic resources in integrated breeding programs that capitalize on advances in phenotyping and genetic tools, combined with "speed breeding".

Introgressions of rye chromosome arm 1RS have contributed to abiotic and biotic stress resistance in wheat, but 1RS possesses the secalin locus that is deleterious for bread-making quality. 1RS sources derived from Petkus and Insave rye carry stem rust resistance genes Sr31 and Sr1RSAmigo, respectively. We derived a Chinese Spring//Pavon 1RSPetkus.1AL/Nekota 1RSInsave.1AL topcross population to investigate the allelic relationship of Sr31 and Sr1RSAmigo. Assessment of reaction in TC1F2 families to Pgt races TKKTP, TTKSK (Ug99), TRTTF and TTTTF facilitated the mapping of Sr31 and Sr1RSAmigo, independently. In addition to stem rust testing, the 10K rye SNP genotyping platform and other 1RS molecular markers combined with a secalin protein assay facilitated identification of recombinant 1RS chromosome arm lines harboring both Sr31 and Sr1RSAmigo and lines not possessing either resistance gene. Our data indicate that the secalin locus maps between Sr31 and Sr1RSAmigo. In order for Sr31-Sr1RSAmigo combination lines to be most useful for breeding programs, the deleterious secalin locus will require deletion. We have engineered and are currently testing several CRISPR/Cas9 reagents that target multiple sites throughout the secalin locus to gauge their mutation efficiency and efficacy in wheat. These include Arabidopsis and wheat optimized Cas9 cassettes, Csy4 and tRNA splicing systems that allow for equimolar release of multiple gRNAs into the cell and the TREX2 exonuclease that has been shown to increase the size of deletions. The reagents will be tested in wheat protoplasts prior to Agrobacterium transformation. The development of an improved 1RS chromosome arm with two broadly effective stem rust resistance genes, but without the secalin locus, could also serve as a landing point for the CRISPR/Cas9 ‘knock-in’ of additional beneficial genes that would co-segregate with 1RS.

The population of Puccinia striiformis f.sp. tritici (PST) has become more variable in the UK, since multiple new genetically dissimilar strains appeared in 2012.These were discovered with the aid of the "Field Pathogenomics" system which sequences infected leaves and genotypes the fungus within the leaf tissue. There is potential in using this system to monitor the population of PST in more detail than previous methods due to its ability to genotype the population within a short timeframe. As the different genetics groups of PST largely correspond with different pathotypes, this should allow for more directed pathotyping, focusing on characterizing new groups or unusual results. With the aim of developing a more effective monitoring program in the UK, this project has used the technology to study in detail the spatial and temporal spread of yellow rust across key field sites in the UK in 2016 and 2017. Our sampling was performed on plots sown with susceptible varieties, to give an impression of the overall population of PST in the region without any bias introduced by host selection pressure The fields were monitored from November through June, to study how the disease develops over the year, the impact of weather on disease spread and to investigate if PST races present early on could be used to predict future epidemics. Multiple samples were taken from each field at each time-point whenever possible, to try and identify novel incursions of rust mid season and to allow for estimates of within field genetic variability. This study tests the potential of using "field pathogenomics" as a monitoring tool for characterising yellow rust populations and looks into the dynamics of rust spread and development to ensure that sampling methods can be optimised to provide a more accurate picture of what is occurring in practical situations.

Wheat blast (or brusone), caused by Magnaporthe oryzae (syn Pyricularia oryzae) pathotype Triticum, is one of the most fearsome and intractable wheat diseases in recent decades. Beyond the Americas, the first outbreak of this disease was detected in Bangladesh in late February 2016. The disease spread to about 15% of Bangladesh's total wheat area with significant yield losses. This underlines the urgent need for collaborative research, policy support and awareness to mitigate the threat of wheat blast. In collaboration with CIMMYT, major research on wheat blast was conducted by Wheat Research Centre, BARI in the 2016-17 wheat growing season. A total of 412 farmers wheat fields were surveyed, of which 77 fields had varying levels of blast disease. Several grass weeds were identified as suspected alternative hosts for the disease. Regarding seed to plant transmission, the fungus was mostly found on the coleoptiles, sometimes on the primary roots and a few on the first leaf. Among 100 wheat genotypes screened for resistance, some were free from blast infection or had low (< 10%) levels of infection, while the susceptible varieties showed 70-90% disease severity. Wheat planted at optimum times (Nov. 15-30) either escaped or had low disease levels compared to late sowing dates. Seed treatment with Provax (Carboxin + Thiram) was found effective in reducing seed infection with improved germination. Preventive foliar spray with Triazoles, particularly in combination with Strobin was found effective in controlling wheat blast infection with significant increase in grain yield.

Puccinia graminis f. sp. tritici (Pgt), the causal agent of stem rust of wheat and barley, continues to be a "shifty enemy" for farmers across the globe. Recent advances in genomic technologies have led to the genotyping of hundreds of Pgt isolates, and have begun to lay the foundation of the genetic landscape of this important plant pathogen. By mining this genomic data, a core set of single-nucleotide polymorphisms (SNPs) markers were developed to distinguish between current critical Pgt lineages. Preliminary testing of 30 SNP markers identified 17 robust markers that were able to distinguish between Pgt isolates representing the four major clades: I, Ug99 race group; II, race JRCQC; III, TRTTF/RRTTF race group; IV, TKTTF race group. Further validation of this core SNP assay, using an expanded set of 200 Pgt isolates, confirmed the specificity of the assay and demonstrated that all of the previously defined sub-clades (III-A, III-B, IV-A.1, IV-A.2, IV-B, IV-C, IV-D, IV-E.1 and IV-E.2) were clearly separated. Further testing of the assay using survey 187 d-samples (ethanol-killed samples) collected from Azerbaijan, Egypt, Eritrea, Ethiopia, Iraq and Kenya identified 43 multiple-locus genotypes (MLGs). The majority of the samples (72.3%) belonged to clades I, III and IV, with III-B, IV-B and IV-E.2 being the most frequently, 14.8%, 20.7% and 14.8%, respectively. The remaining samples (27.7%) consisted of 17 MLGs, belonging to five new MLG classes. The majority of these samples (38 samples, 20.2%) belonged to a new MLG.14 class and was found in collections from Azerbaijan, Eritrea and Iraq. A MLG nomenclature for this core SNP assay will be presented. The development a core SNP assay for Pgt expands the molecular toolbox for genotyping and surveillance of this "shifty enemy".

A data-driven stochastic simulation framework has been developed and tested that makes use of finely resolved meteorological data and high-performance computing resources to trace and predict atmospheric dispersal of Puccinia graminis f. sp. tritici (Pgt) urediniospores. The simulation framework is used as a tool to derive best estimates of disease risks when facing on-going threats. We present two aspects of our work: First, a series of case-studies is summarized, emphasizing how surveillance and dispersal modelling can be combined to enhance our understanding of spatiotemporal disease dynamics, but also touching upon the difficulties of validating dispersal models in the face of data scarcity on regional and continental scales. In particular, we discuss comparisons between simulation results and available field disease data in two case-studies: (i) identification of key airborne Pgt-urediniospore migration routes in Southern/East Africa, the Middle East, and Central/South Asia; and (ii) the analysis of likely incursion routes of Pgt race TKTTF into Ethiopia. In the second part of the presentation, a novel short-term forecasting system for wheat rusts in Ethiopia is introduced, which has been recently launched and is currently being developed further. The Ethiopia Early Warning System links field disease surveillance results directly to atmospheric dispersal simulations, environmental suitability assessments, and spatial epidemiological models of disease progress. A brief overview of the concept is provided, followed by a report of lessons learnt and successes from the ongoing wheat season in Ethiopia (2017-2018).

Although relatively little wheat is grown in Ecuador, all three rust pathogens are commonly found. In February 2016, a severe epidemic of P. graminis f.sp. tritici (Pgt) was reported on an experimental UC-Davis line containing Sr38. The line, UC11075, was originally sent to the southern INIAP station in Ecuador in 2012. The Pgt pathotype was determined at the Ag-Canada Morden Laboratory as RRTTF. Isolates of RRTTF race were also found later that year at the INIAP station further north near Quito on CIMMYT lines. Independent genotyping of the isolates at the Ag-Canada Ottawa Laboratory using SSR markers and at the USDA Cereal Disease Laboratory in Minnesota using SNP markers showed the Ecuador samples were identical to isolates previous reported in Pakistan, Yemen, and Ethiopia. Other Pgt isolates collected at the northern INIAP station that same year showed varying virulence to Sr13 and Sr27, not in the standard differential set. Due to the low amount of wheat grown in Ecuador, it is difficult to determine how prevalent RRTTF is or how long it has been in the country. However, based on the severity of the epidemic in the research plot in southern Ecuador, and its appearance in the north, the RRTTF race was likely not a recent introduction. The long-term presence of the RRTTF in Ecuador is uncertain. This may be influenced by the continuous growing seasons, the few, small and changing landscape of wheat fields, infection of voluntary wheat and other grass hosts, or sexual recombination on one or more of the 30 Berberis species found in the Ecuadorian highlands. Race RRTTF represents a new threat to wheat production in the western hemisphere as this is virulent on most Sr resistance genes, and supports the need for global surveillance of movement and detection of the rust pathogens of wheat.

Wheat rusts have been associated with crop failures and famine throughout history. Recent outbreaks of stem and yellow (stripe) rust in Europe have been linked to expansions in pathogen geographic distribution, exotic incursions and increased virulence. For example, in the past two decades, wheat yellow rust, caused by Puccinia striiformis f. sp. tritici (PST), has seen the emergence of new races that are adapted to warmer temperatures, have expanded virulence profiles, and are more aggressive than previous races, leading to wide-scale epidemics. Our rapid "field pathogenomics" strategy, that uses gene sequencing of infected wheat leaves taken directly from the field, has enabled us to gain insight into the population structure of PST over successive seasons. For instance, our analysis uncovered a dramatic shift in the PST population in the UK and supports the hypothesis that a recent introduction of a diverse set of exotic PST lineages may have displaced the previous populations. In addition, we uncovered potential seasonal and varietal specificity for certain genotypes of PST and the first evidence of mutations in the demethylation inhibitor (DMI) fungicide target gene, cyp51. Furthermore, by expanding our technique to wheat stem rust, we have also gained new insight into the recent emergent outbreaks of this disease in Europe. Developing and applying a genomics-driven approach to pathogen surveillance, we have generated valuable new knowledge that could be extremely useful for various aspects of disease management.

The Institut National de la Recherche Agronomique is the only public organization in charge of varietal development of major field crops especially cereals. The cereal breeding program which started with "massale" selection using local germplasm allowed the release of ten national varieties. Since the 1980's, the program gradually shifted to hybridization between national and international material to improve grain yield, resistance to the main diseases, and quality. To date, 103 cereal varieties are released and registered. During the lastthree years, an increase was noticed in terms of release and/or registration. This year, 16 cereal varieties, including three durum wheat varieties from INRA/ICARDA partnership are under registration to the national catalog. INRA has always had the responsibility of variety maintenance of breeder seeds G0 of their own varieties. Since 2005, it has disengaged from the G2/G3 seed production and keeps supplying G1 seeds and pre-basic seeds G2 upon the request of some seed companies. Before 2002, INRA had licensed its own varieties to seed companies, through calls for tenders. The use right is granted to the company that has the highest offer. During this period, 78 cereal varieties were conceded to seed companies especially to SONACOS. However a small number of these varieties are still under commercial seed production. The use rate is still low. Given the fact that the calls of tenders were not fruitful until 2014, INRA has offered its new varieties via a new agreement, AMI based on new contracts where the royalties and some articles have been revised. Eighteen varieties including 14 cereals were conceded. Despite the efforts made, all the seed partners including INRA should reinforce their commitment to deal with the challenges facing the cereal seed sector.

The study was conducted in Arsi and West Arsi zones of Ethiopia. It is designed to assess the factors that affect the adoption of rust resistant wheat varieties and the impact of these technologies on wheat productivity and commercialization. The study employed a double hurdle model to determine the factors that affect adoption of rust resistant wheat varieties and propensity score matching to carry out the impact study. The sample consists of cross-sectional data of 158 small-scale wheat farmers selected from the study area. The study found that among the total sampled respondents, about 58% were adopters of rust resistant wheat varieties and on average the adopters allocated about 73% of their total wheat area to rust resistant improved wheat varieties. According to the result of the double hurdle model, it was found that oxen ownership and participation to extension services positively affects the likelihood of adoption of wheat technologies. Furthermore, female headed households are more likely to adopt wheat technologies compared to their male headed household counterparts. With regard to the intensity of adoption of rust resistant wheat varieties, experiences on improved crop management practices positively affects the proportion of area allocated to improved wheat technologies while oxen ownership affects negatively the proportion of area allocated to improved wheat technologies. According to the result of the propensity score matching method, rust resistant wheat variety adoption on average increased wheat productivity of adopters by about 1.2 ton per hectare than the non-adopters. Similarly, the result of the propensity score matching estimates also showed that on average adopters sell 13% more of their wheat produce than the non-adopters. The results provide empirical evidence that agricultural technology adoption can contribute to improving productivity and transforming the Ethiopian traditional subsistence smallholding farming system.

Stem rust, caused by Puccinia graminis (Pg), remains one of the most destructive diseases of wheat. The emergence and spread of the Ug99 family of stem rust races out of East Africa rekindled concern over the impact of Pg on global wheat security and has underscored the need for durable sources of genetic resistance. Intrinsically durable, non-host resistance (NHR) presents a compelling strategy; but NHR to Pg remains largely unexplored as a protection strategy in wheat, in part due to the challenge of developing a genetically tractable system in which Pg-NHR segregates. In this study, we approached the question of Pg-NHR through the pathogen's sexual host, barberry (Berberis spp.). Specifically, we developed a bi-parental, interspecific mapping population (F1) via a controlled cross between Pg-susceptible European barberry (B. vulgaris; Bv) and Pg-resistant Japanese barberry (B. thunbergii; Bt), to dissect the genetic mechanism of the apparent Pg-NHR exhibited by Bt. To facilitate this dissection, we built a reference genome for Bt and constructed genetic linkage maps for both parental species. Specifically, using long-read PacBio data from 116 SMRT cells, we assembled a 1.3 Gb draft genome for Bt using FALCON-Unzip; and from a total of 37,283 variants (SNPs and indels) identified between the parents via genotyping-by-sequencing, we constructed genetic linkage maps with 704 and 304 markers for Bt and Bv, respectively. Within each species, the markers segregated into 14 linkage groups, in agreement with the 14 chromosomes present in these diploid Berberis spp. QTL analysis for Pg-NHR led to the identification of a single QTL, spanning 7.21 cM and containing 4 contigs (1.74 Mbp). With this study, we demonstrate for the first time the feasibility of dissecting the genetics of a source of Pg-NHR, which may contribute insight into possible novel mechanisms of durable rust resistance in wheat.

Berberis holstii is native to the highlands of East Africa and frequently grown in proximity to fields of small grain cereals. The species is susceptible to Puccinia graminis and P. striiformis in artificial inoculations. In an earlier study, we obtained stem rust infections on susceptible lines of wheat, barley, rye and oats using aeciospores collected from naturally infected B. holstii in North Shewa (Ethiopia) and concluded that P. graminis completes its sexual cycle in Ethiopia. Subsequently, single-pustule isolates were derived from six aecial samples that infected wheat 'Line E' and 'Morocco', and barley 'Hiproly'. These isolates were evaluated on wheat stem rust differential lines and other genetic stocks carrying various stem rust resistance genes. Several races of P. graminis f. sp. tritici were identified with virulence combinations ranging from broadly avirulent to broadly virulent. Results from this study established for the first time that B. holstii is functional as an alternate host for P. graminis f. sp. tritici in Ethiopia. The presence of the sexual cycle has likely played a role in generating the virulence diversity observed in the Puccinia graminis f. sp. tritici population in Eastern Africa.