Source: UNIVERSITY OF NEBRASKA submitted to
Sponsoring Institution
National Institute of Food and Agriculture
Project Status
Funding Source
Reporting Frequency
Accession No.
Grant No.
Project No.
Proposal No.
Multistate No.
Program Code
Project Start Date
Dec 15, 2016
Project End Date
Dec 14, 2018
Grant Year
Project Director
Baenziger, P.
Recipient Organization
Performing Department
Agronomy & Horticulture
Non Technical Summary
The long-term goal of this project is to develop the necessary knowledge base, germplasm, and heterotic pools to support the development of hybrid wheat. Wheat (Triticum spp.) yields will need to increase by 1.7% per year (from its current increase of 0.9% per year) to feed a larger global population with increasing dietary needs. Hybrid wheat, which is more climate resilient than pureline wheat, can contribute to achieving this goal. The project objectives are to: (1) screen two large wheat breeding programs for the floral and plant traits needed for efficient hybrid seed production and hybrid performance; (2) create and test hybrids to establish and confirm heterotic pools in wheat; (3) genotype the lines going into the heterotic pools and improve algorithms to separate lines into maximum likelihood pools for future testing and validation; and (4) map restorer genes in T. timopheevi cytoplasm and create a series of cytoplasmic male sterility (CMS) tester lines, their maintainer lines, and a series of elite restorer lines (R-lines) and begin to determine the efficacy of CMS-based hybrid systems. The project team has made great strides in the U.S. and abroad toward developing the tools to foster hybrid wheat development to maximize wheat yield potential. Using an integrated approach involving in-house germplasm, chemical hybridizing agents, breeding, phenotyping, genomic selection, and quantitative trait loci mapping, this project is expected to help create the scientific and germplasm foundations for successfully launching the hybrid wheat industry in the U.S. and is thus relevant to the NIFA-IWYP program.
Animal Health Component
Research Effort Categories

Knowledge Area (KA)Subject of Investigation (SOI)Field of Science (FOS)Percent
Goals / Objectives
The need for increased cereal production is clear as we must feed an ever larger global population with greater dietary needs (Whitford et al., 2013). To meet these needs, it is estimated that wheat (Triticum spp.) yields will need to increase by 1.7% per year; however, current wheat production trends are increasing by only 0.9% per year. In contrast, the much better resourced crop maize (Zea mays L.) has a current yield increase of 1.6% per year. One reason for this trend is that maize is largely a hybrid crop which exhibits substantial heterosis (i.e., hybrid vigor) compared to traditionally self-pollinated crops. In other words, hybrid wheat has the potential to greatly improve the yield per year increase if adequate heterosis can be developed and if practical mechanisms of producing hybrid seed can be implemented.With regard to yield, hybrids like maize have the advantage over traditional crops in that either hybrid parent can be a source of disease or pest resistance, and when the resistance gene is dominant, parents can impart this trait to the hybrid progeny. Seed would be purchased annually, thus providing the resources to support a robust crop improvement program similar to that of maize. Increased grain yields (i.e., through a second Green Revolution) are also needed to preserve land for other uses (Stevenson et al., 2013).The development of commercially successful wheat hybrids first requires cost-effective methods of creating hybrids, sufficient levels of heterosis to make purchasing hybrid seed economically viable, and a suite of methodologies to create and improve heterotic pools using innovative current and forthcoming technologies. With the pressing need to improve wheat (T. aestivum L.) productivity, our long-term goal is to develop the necessary knowledge base, germplasm, and heterotic pools to support the development of hybrid wheat. Toward that end, the objectives of this project are to:Screen two large wheat breeding programs for the floral and plant traits needed for efficient hybrid seed production and performance;Create and test hybrids to establish and confirm initial heterotic pools in wheat;Genotype the lines going into the heterotic pools and improve algorithms to separate lines into maximum likelihood pools for future testing and validation;Map restorer genes in T. timopheevi cytoplasm and create a series of CMS tester lines, their maintainer lines, and a series of elite restorer lines (R-lines) and begin to determine the efficacy of CMS-based hybrid systems.Bringing together an international team of researchers from the University of Nebraska-Lincoln (UNL), the International Maize and Wheat Improvement Center (CIMMYT), Texas A&M University (TAMU), the University of Hohenheim, Kansas State University, Genetics and Crop Plant Research (IPK), and a number of consultants, the proposed project builds on our team's vast experience in highly successful and diverse cultivar breeding efforts and is well poised to develop the necessary publically available knowledge base and germplasm to support the successful development of hybrid wheat (see "Biographical Sketches" and "Key Personnel Roles" for additional details).
Project Methods
Objective 1. We will continue to screen over 350 new lines each year in Nebraska for anther extrusion. We will confirm previous anther extrusion on 150 elite lines each year. A similar set of new TAMU lines will also be screened in Texas. Planted in fall 2015 for evaluation in spring 2016 were the 299 lines in the Hard Winter Wheat Association Mapping Panel (HWWAMP, Guttieri et al., 2015). This panel represents the historic to modern germplasm of the U.S. Great Plains and has been genotyped using the 90K SNP chip. Hence, we can map the QTLs associated with anther extrusion using pre-existing marker data. We will repeat this experiment during the 2016-2017 season. The Nebraska preliminary, advanced and most of the elite yield trials lines are genotyped using genotype-by-sequencing (GBS), and genome-wide association mapping (GWAS) will be used to identify QTLs for anther extrusion (AE) in our breeding programs. We performed preliminary GWAS analysis on 184 lines and found 9 SNP markers (with P<0.001) that were placed above majority of the SNPs in the Manhattan plots. There were no associations that passed the Bonferroni statistical significance threshold (P<0.05). The 9 SNP markers were located on 1BL, 3B, 5BL, 6AS and 7BS chromosomes. Skinnes et al. (2010) identified 4 QTLs for anther extrusion in a hexaploid wheat biparental mapping population on 1AL, 1BL, 4DL and 6AS. We are currently investigating whether the genomic regions on 1BL and 6AS overlapped in the two studies. To identify floral and plant traits needed for efficient hybrid seed production, we will visually score gaping after the use of a CHA and in our CMS lines.Objective 2. We have access to the Saaten-Union CHA Croisor® 100 for this project. Our goal is to produce over 600 wheat hybrids per year. Our parental lines will be from two diverse wheat breeding programs: Texas and Nebraska (Chao et al., 2011). Both sets of parent lines will be chosen based upon flowering time, genetic diversity, floral characteristics (Zhao et al., 2015b; Boeven et al., 2016), and results from our hybrid yield trials in 2016. The female parent lines will also be selected for gaping and the male lines for anther extrusion. All parents will be semi-dwarfs to avoid lodging due to heterosis for height (a conventional height parent line will lead to a hybrid that is too tall and may lodge). The lines will be tested four to seven environments. Our mating design in future will be a balanced missing design based upon Zhao et al. (2015b) where we will have 50 male lines and 100 female lines. We will have 25 crossing blocks in Texas and 25 crossing bocks in Nebraska to avoid the risk of catastrophic losses by inclement weather. Each crossing block will have one male line surrounding randomly assigned 14 female lines, with the "male" line used as a female. Females will be grown in 3 m x 1.2 m plots at twice the normal seeding rate surrounded on both sides by strips of male parents. A sample of heads from every female line from two of the male crossing blocks will be bagged to estimate the efficacy of the CHA. Using the balanced missing crossing block design, each female will be in seven or eight crossing blocks. The experimental hybrids will be evaluated in three locations in Nebraska (Lincoln, North Platte, and Alliance) and in three locations in Texas (Bushland, McGregor, and Farmersville) using 50% to 75% of the normal seeding rate based upon actual seed counts. The experimental design will be an alpha-lattice incomplete block design with repeated check cultivars and parent lines replicated twice to help estimate heterosis and spatial variation (e.g., Guttieri et al., 2015). We estimate the trials will be between 650 and 750 plots per location, depending upon the quantities of hybrid seed produced from each hybrid combination. Finally, we will add 50 lines that represent 30 to 40 elite European lines and 10 to 20 European hybrids selected by our collaborators Drs. Longin and Reif.The crossing block and hybrid evaluation trials will be repeated in 2018 and 2019. New lines will be added to replace those found to be unsuitable for use in hybrid production. Successful hybrid combinations will be remade for more advanced testing and model validation.Objective 3. Using the approach of Zhao et al. (2015b), we will genotype all our parental lines (50 male lines plus 100 female lines = 150 lines/year) using GBS (Poland et al., 2012). As we will be looking for new parent lines, we will genotype an additional 100 lines based upon floral characteristics for the next crossing block. Furthermore, to look for synergies between and link our hybrid wheat efforts and the hybrid wheat efforts ongoing in Europe, we will genotype 50 lines and hybrids selected by Drs. Longin and Reif. All of the 600-700 hybrids will be tested at four to seven locations to provide the needed phenotypic data on heterosis and other traits (see above in Objective 2). The genetic distance among the lines and the complete hybrid matrix can be predicted (11,175 hybrids for n= 150). This complete matrix will be evaluated using the annealing algorithm of Zhao et al. (2015b) to predict high yielding heterotic groups. Based upon the outcome of the heterotic group predictions, we will determine the optimum size of each heterotic group for further improvement. Historically, approximately 16 individuals per group are sufficient for long-term success (e.g., the creation of high yielding hybrids and the needed genetic variation for long-term, sustained improvement; Zhao et al., 2015b). In addition, the heterotic groups identified with CHA hybrids will be used to guide our CMS lines development in to R-line and B/A-lines.Objective 4. We will map multiple restorer genes to provide full fertility restoration in T. timopheevii-based F1 hybrids from a single source line by using high density Single Nucleotide Polymorphism (SNP) markers. To map the restorer genes, R38568G (an excellent restorer lines from Australia) will be crossed with a CIMMYT A-Line "Navojoa-A," which is a complete male sterile line carrying T. timopheevii cytoplasm. The self-pollination in F1 fertile plants will be confirmed by bagging each heads before flowering. A set of 300 F2 seeds will be space-planted in the field for phenotypic and genotypic analysis. Male fertility in each F2 plants will be assessed during flowering and maturity. To prevent contaminant outcrossing, similar to F1, the heads of individual F2 plants will be covered with glassine pollen-impermeable bags before flowering. At flowering, the anthers in the florets in each head will be visually evaluated as fertile or sterile. To confirm anther fertility, four random samples of pollen from each fertile- and sterile-type anther will be collected and analyzed under a microscope after smearing it with 1% iodine-potassium iodide (I-KI) solution (Sinha et al., 2013). For maker analysis, a leaf sample from each F2 plant will be collected and the DNA from each sample will be used for 90K SNP genotyping. The polymorphic and curated SNP data will be used for linkage map construction and QTL mapping analysis for fertility restoration in the F2 population. Once the linked flanking markers are identified, the SNP sequence will be used to develop breeder-friendly, allele-specific PCR assays (e.g., KASPar).The relevant experimental designs, data analyses and interpretations have been described in each objective. We will use SAS, R, and ASREML for our phenotypic and genomic data analyses. We will collaborate with Drs. Longin and Reif using their annealing and other algorithms related to the development and prediction of heterotic pools (e.g. Zhao et al.; 2013, 2015a,b; Boeven et al., 2016).

Progress 12/15/16 to 12/14/17

Target Audience:The target audience includes commercial seed producers, purchasers of advanced and improved seed, millers and bakers, biotechnologists, plant breeders, and those that consume wheat. Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?Vikas Belamkar (partially funded from this grant) attended a R-workshop on "Advanced R Programming for Statgen package development" at the Summer Institute in Statistical Genetics 2017, University of Washington, Seattle, WA July 12-14, 2017 and will be transferring this knowledge to our group. How have the results been disseminated to communities of interest?In preparation for the launch of hybrid wheat, talks were given to the Nebraska Crop Improvement Association (80 seed dealers in attendance) in January 2017 and at three seed days in western Nebraska ( approximately 300 growers in attendance) in August, 2017. What do you plan to do during the next reporting period to accomplish the goals?Obj.1: In 2017-2018, we will continue to screen our elite Nebraska germplasm (300 to 400 lines in Nebraska) for anther extrusion and the HWWAMP. In addition, we will begin to study the needed female characteristics (female gape and duration, and stigma receptivity) for increased hybrid seed set. In Texas, the same 80 lines, representing the 2016 UVT and TXE will be grown again in fall 2017 and screened for genotype-by-environment interaction of anther extrusion and female stigma exertion and gape in 2018 at College Station and McGregor. Furthermore, 320 lines representing the advanced pipeline of Texas breeding program, including the 2018 UVT, TXE, STA 1 - 2, and STP 1- 4, will be screened for anther extrusion and female stigma exertion and gape at College Station, TX. Obj.2: We will replant our hybrid crossing block to make additional hybrid seed for testing. We will repeat our experiment to study CHA rates and different adjuvants to reduce phytoxicity and the amount of chemical needed to sterilize wheat plants. We will plant hybrid yield trials at six locations (three in Nebraska and three in Texas). These trials use a balanced missing design hybrid which is critical to use the algorithms developed by our German cooperators to identify heterotic pools. We will continue to analyze our complete 25 x 25 plus one diallel. By having reciprocal hybrids, we will learn if some lines are better as females or males based upon hybrid yield (as opposed to making the decision on floral characteristics). We can also use the diallel hybrid data to selectively drop some of the hybrids and predict their performance from our GBS data and the balanced missing algorithms to increase our understanding of how to better predict hybrids. Obj.3: We were successfully able to genotype all 26 lines in our original hybrid diallel and 143 lines of the 150 in our balanced missing crossing block. The seven lines that were not successfully genotyped will be resubmitted for genotyping, as will all new lines added to the crossing block. Genomic predictions and development of heterotic groups and patterns has to wait until we have the hybrid field trials analyzed and the spatial variation removed or accounted for so that we have the most accurate estimates of the hybrid values. We expect those analyses to be done on the diallel hybrid set by December and we will begin working on the heterotic group identification thereafter. Also, the potential number of hybrids is so much larger using the balanced missing design than with the complete diallel design, the full effort on heterotic groups will have to wait until we can analyze next year's hybrid yield trial. Obj.4: CIMMYT and Texas: To further validate the mapping results and develop KASP markers for Rf genes, a F4:5 recombinant inbred line (RIL) populations is being developed at El Batan. In 2017-18 growing season, each F5 line will be crossed with a CMS line to produce enough F1 seed for field evaluations in 2-3 environments. Once the diagnostic molecular markers are developed, individual Rf genes and combinations will be further evaluated for their ability to restore fertility in order to identify the best set of Rf genes necessary in CMS based hybrid wheat system. Similarly, newly developed CIMMYT restorer lines will also be haplotyped in order to confirm the presence of Rf genes. Nebraska: We will continue to backcross our elite lines into T. timopheevi cytoplasm to create new A-lines and develop sufficient seed for CMS based hybrids. Our convectional breeding program to create new R-lines (R-line x B-line crosses followed by selection for fertility) will continue. As molecular markers are identified, these R-lines will be screened to determine which restorer genes (Rf) they may have. We will also begin making crosses to ensure they have sufficient restorer genes to restore our A-lines in hybrid combinations.

What was accomplished under these goals? Obj. 1: We have continued to screen our advanced breeding materials for anther extrusion. Previously, in Nebraska we screened the preliminary, advanced, and elite yield trials, as well as the regional performance nursery for a total of 288 lines in the field for the second year. A select group of lines (29 lines) was also screened for anther extrusion in the greenhouse. The anther extrusion ranged from 2 to 7 (on a scale where one indicated that few anthers, or only the tip of most anthers are visible outside of the glumes at anthesis, and nine indicating that many anthers were fully exposed outside of the florets at anthesis) in the field, but there was considerable genotype by environment (GxE) interactions. Over fifty percent of the lines had scores of 5 or above (considered to be adequate to be a male line in hybrid production). While some lines (e.g. Freeman, scored as 7) consistently had high values for anther extrusion and other lines consistently had poor anther extrusion (e.g. Camelot scored as 2), many lines exhibited higher environmental variability. The broad sense heritability for anther extrusion was relatively high (ranging from 0.62 to 0.85 depending upon the nursery). In 2017 we screened an additional 120 lines and 299 lines the Hard Winter Wheat Association Mapping Panel (HWWAMP) developed as a resource for the Great Plains by the previous TCAP grant. The HWWAMP will be used in genome wide association studies to map QTLs for anther extrusion. In Texas, we have screened TAMU wheat breeding nurseries (SOBS, AOBS, STP1-STP4, and AP1-AP10) and lines from UNL (NIN, TRP, IRDR, RPN; approximately 600 lines) for floral characteristics in spring 2014, and 100 promising lines were selected and planted at College Station during fall 2014. A set of promising 180 TAM germplasm lines, selected for floral characteristics, was planted in the field at College Station and McGregor in fall 2016 (representing STP 1 - 4, TXE, UVT) to look at genotype-by-environment interaction for anther extrusion and female stigma exertion and gape. Obj. 2: We used Croisor® 100 make hybrids from a 25 female x 25 male complete diallel. For efficient field work, an additional female line was added making the crossing block 25 males by 26 females (hereafter referred to as a 25 x 25 diallel plus one). Visual ratings for CHA induced floral gaping and phytotoxicity due to CHA application was recorded. Few heads in every female plot were bagged to estimate the efficacy of the CHA. Ideally, the bagged heads were completely sterile indicating that the CHA worked, and the seed set on the unbagged heads was good indicating that there is good cross-pollination and few phytotoxic effects. The female plots were harvested to measure grain yield. In 2015, 80% of bagged heads in the females had 10 seeds or fewer, whereas 93% had 5 seeds or fewer in 2016, so adequate sterility was achieved for our hybrid seed production. We completed our first year of hybrid evaluation trials in 2016 (three locations in NE and one in TX) and our second year (three locations in NE and two in TX) in 2017. An augmented incomplete block design was used as hybrid seed is precious and the trials are very large (complete replications would be impossible). Due to limited seed, not every hybrid was tested at every location. However, at least 550 hybrids were tested at the five locations in 2017. Complete sets of the diallel crosses (650 hybrids) were harvested at two locations in Nebraska, in Lincoln and at North Platte. We are now working on analyzing the data, removing the spatial variation, comparing reciprocal crosses, and estimating heritability and heterosis. We successfully used the balanced missing design pioneered by our German cooperators for the 2017 crossing block. In this crossing block, we had 50 males and 100 females (total of 150 lines with 25 males and 50 females selected by both UNL and TAMU) where each male was crossed to 14 females. To ensure that environmental hazards did not destroy the crossing block, we grew three crossing blocks (1 in NE and 2 in TX). The NE (25 males and 100 females) and one of the TX (the remaining 25 males and the same 100 females) crossing blocks worked very well with a range of seed produced from 65g to 2070g with an average of 657g in Lincoln and 24g to 1182g with an average 326g in Texas. The seed from both crossing blocks has been sent to Lincoln for designing the hybrid yield trials for six locations (three in Nebraska and three in Texas). Obj. 3: DNA isolation, genotyping-by-sequencing, SNP calls: The 150 lines in the crossing block were genotyping-by-sequencing (GBS) at the Wheat Genetics and Germplasm Improvement Laboratory (WGGIL) at Kansas State University. Subsetting the SNP calls for the 150 parental lines and filtering SNPs and lines with appropriate quality control (excluded SNPs and lines with >20% missing information and SNPs with minor allele frequency (MAF) <0.05) resulted in 143 lines and 39,960 high-quality SNPs. This dataset filtered with high-stringency was then used to investigation population structure of the parental lines. Seven parental lines with more than 20% missing information will be re-genotyped using GBS in 2017-2018. The 280,285 SNPs obtained in the SNP calling were filtered to retain high-quality SNPs and the missing sites were imputed using Beagle v4.1. The parental lines were subdivided and SNPs with MAF>0.05 were retained. This resulted in 114,436 SNPs across 150 lines. Subsequently, we built 11,175 hybrids in silico using CreateHybridGenotypesPlugin in TASSEL. The hybrid dataset was filtered for retaining SNPs with MAF>0.05 and maximum missing sites less than 20%, which resulted in 56,719 high-quality SNPs across the hybrids. This indicates, we will very likely have more than 50K SNPs across the 11,175 hybrids to run the prediction in 2018-2019. For investigating the population, we added an additional filter and excluded hybrids with >20% missing sites. This resulted in 56,470 high-quality SNPs across 10,731 hybrids for the population structure analysis in the hybrids. The population structure was investigated using principal component analysis (PCA) and a phylogeny based on Neighbor joining (NJ) algorithm. We found that the parental lines are quite diverse based on PCA and there were two outgroups, one comprising five lines from the Nebraska (NE) breeding program and the other containing five lines from the Texas (TX) program. The remaining lines were weakly structured into:1) mainly lines from NE, 2) mainly lines from TX, and 3) lines from both from NE and TX. Structure among the hybrids is similar to the parents but with a significantly higher overlap/increased kinship. Obj. 4: A BC1F1 mapping population of spring wheat genotypes (n = 280) was developed by crossing a restorer source (T. timopheevii cytoplasm) with an advanced line (T. aestivum cytoplasm) at El Batan, CIMMYT. In 2016-17 growing cycle, the BC1F1 population was phenotyped for fertility restoration by counting the number of fertile lateral florets and total number of seeds per spike present in a set of three spikes per genotype. The spikes were covered with glassine bags before flowering to prevent cross fertilization and contamination. The individual BC1F1 plants have also been genotyped using a 20K SNP chip. The genotypic data will be filtered for monomorphic markers and missing data and will be used to map regions corresponding to fertility restoration. Software package ICIMapping (Integrated Breeding Platform, will be initially used to map genomic regions corresponding to fertility restoration using simple interval mapping and inclusive composite interval mapping (ICIM) algorithm. We expect to have preliminary mapping results by the end of 2017.


  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Hybrid Wheat for the Great Plains: Improving Yields and Water Efficiency? Poster presented by Amanda Easterly at the Daugherty Water for Food Global Institute at the University of Nebraska 2017 Annual Conference in April 2017. Authors: Amanda Easterly, Vikas Belamkar, Nick Garst, Jackie Rudd, Amir Ibrahim, Anil Adhikari, P. Stephen Baenziger.
  • Type: Journal Articles Status: Under Review Year Published: 2017 Citation: Evaluation and implementation of genomic selection in preliminary yield trials in the University of Nebraska winter wheat breeding program. Vikas Belamkar, Mary J. Guttieri, Waseem Hussain, Diego Jarqu�n, Ibrahim El-basyoni, Jesse Poland, Aaron J. Lorenz, P. Stephen Baenziger. Manuscript submitted on August 01, 2017 to Theoretical and Applied Genetics. Status: Pending major revisions.
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Screening Great Plains Winter Wheat (Triticum aestivum L.) Germplasm for Anther Extrusion and Evaluating the Effect of Anther Extrusion on Hybrid Seed Production Yields Nicholas Garst Masters Defense seminar and thesis
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Diversity and Performance of Wheat Hybrids in the Great Plains. Poster and 5 minute oral presentation by Amanda Easterly at the 2016 ASA/CSSA/SSSA International Meeting in Phoenix, AZ. Authors: Amanda Easterly, University of Nebraska - Lincoln; Nicholas Garst, University of Nebraska - Lincoln; Vikas Belamkar, University of Nebraska - Lincoln; Jackie C. Rudd, Texas A&M AgriLife Research; Amir M.H. Ibrahim, Texas A&M University; P. Stephen Baenziger, University of Nebraska - Lincoln
  • Type: Conference Papers and Presentations Status: Published Year Published: 2017 Citation: Modeling Male Sterility Induced by Chemical Hybridization of Wheat (Triticum aestivum L.) Poster presented by Amanda Easterly at the Conference in Applied Statistics, Manhattan, KS, April 2017. Authors: Amanda Easterly, Nicholas Garst, Vikas Belamkar and P. Stephen Baenziger  University Nebraska-Lincoln