Chengxu Liu; Anthony Yannarell

Restoring cropland to native prairie can improve soil quality. Since 2018, a group of students and faculty members from University of Illinois at Urbana-Champaign have been assisting habitat restoration at Phillips Tract Prairie, Urbana, Illinois. My research project investigates how this restoration has affected the soil properties. I measured soil bulk density for the years 2021 and 2022 (3- and 4-years post-restoration) using oven-dried soil of known volume. I also measured pH with and electronic pH meter for these same years, over three different depth increments (0-10 cm, 10-20 cm, and 20-30 cm). I also created a map of the restoration area from an aerial overview of the prairie using Google Earth. By connecting the soil data to the mapped sampling area, I was able to discover the general patterns of the physical and chemical soil health across the prairie, and to understand how these patterns changed over time, space, and soil depth.

Samantha M. García, Eric R. Larson Ph.D.

Environmental DNA (eDNA) is genetic material present in an environment, shed from organisms inhabiting the area. The presence of a species in an area can be determined by isolating the DNA from an environmental sample and analyzing the sample for the DNA of the target species. This less labor-intensive method of species detection has led to the rise in popularity of eDNA as a tool for invasive species surveillance. Invasive species management is most successful when an invasion is identified and addressed during its early stages when the species abundance is likely to be low. Therefore, it is important for the eDNA detection method to be able to detect low concentrations of DNA. Many factors, including the analytical procedures used, affect the DNA yield recovered from an environmental sample. eDNA storage and extraction methods are essential steps in the procedure that can be adjusted to maximize detectability. In this study, we compared the performance of two eDNA storage and extraction method combinations on the detection of three common aquatic invasive species across five lakes. Environmental samples were split in half creating pairs used to examine the effects of both methods on eDNA detectability. One eDNA storage-extraction method consisted of storing samples in 95% ethanol followed by DNA extraction via Qiagen kit. The other storage-extraction method consisted of storing samples in cetyl trimethylammonium bromide (CTAB) followed by phenol-chloroform-isoamyl (PCI) DNA extraction. The CTAB-PCI method resulted in more positive detections and higher eDNA copy numbers compared to the ethanol-Qiagen method.

Kayla Vittore, Jungwoo Lee, Dr. DoKyoung Lee, Dr. Chunhwa Jang

Fiber crops like Miscanthus x giganteus (Miscanthus) and Panicum virgatum (Switchgrass) are useful for various purposes such as fabrics, fuel, and forage. The quality of harvested fiber is determined by the ratio of lignin, cellulose, and hemicellulose in the fiber, which can be quantified using wet chemistry. There is interest in whether management factors like fertilization could influence the quality of harvested fiber. However, the traditional wet chemistry methods of determining these ratios are costly and time-intensive. Therefore, alternative measurement methods such as Vegetative Indices (VI) are being explored. This study aimed to determine if Nitrogen fertilization rate influenced composition ratios in Miscanthus and Switchgrass, as well as to investigate if mid-season VI’s could be used to estimate the end-season composition ratios. Four fertilization rates were applied to plots in a Miscanthus field established in 2009. Two fertilization rates were applied to plots in a Switchgrass field established in 2020. The results showed that Nitrogen fertilization rate had a positive relationship with Miscanthus fiber component yields but did not strongly influence fiber composition ratios. Switchgrass, on the other hand, had a weak or no relationship between nitrogen fertilization and fiber component yields or composition ratios. It is important to consider the age of the fields as the Miscanthus field was over a decade old, while the Switchgrass field was only two years old. Correlations between the three VI’s and end-season composition ratios were weak, suggesting the selected VI’s were poor predictors of end-season harvest composition or yields. NDRE consistently had the strongest linear correlation with yields of components, possibly due to its better resilience to saturation at higher reflectance values. Based on the results, Nitrogen fertilization may increase Miscanthus fiber component yields but not significantly affect fiber composition ratios. Switchgrass may have a weak or no relationship with nitrogen fertilization and fiber components. The interaction between fertilization and stand-age may contribute to these results, and will be important to study in future projects. Finally, while the selected VI’s were not good predictors of end-season harvest composition or yields, the NDRE results suggest other VI’s may perform better.

Noeleen Brown, Crystal Concepcion, Dean Riechers

S-metolachlor is a very-long-chain fatty acid (VLCFA) elongase-inhibiting herbicide, and widely used for preemergence, residual control of annual grasses and small-seeded broadleaves in crops such as corn, soybean, and cotton. Previous field research investigating multiple herbicide-resistant (MHR) waterhemp (Amaranthus tuberculatus) populations from Illinois demonstrated poor control with S-metolachlor. Several Palmer amaranth (A. palmeri) populations from the U.S. have shown similar responses in the field but detailed metabolism studies with S-metolachlor have not been reported in MHR or sensitive Palmer amaranth. Previous research with MHR waterhemp seedlings demonstrated rapid metabolism of S-metolachlor compared to sensitive populations. The objective of this research was to conduct a comparative analysis of S-metolachlor metabolism in suspected VLCFA inhibitor-resistant Palmer amaranth populations in comparison with sensitive Palmer amaranth and MHR waterhemp populations and tolerant corn seedlings. Metabolism experiments using intact seedlings and radiolabeled S-metolachlor combined with thin layer chromatography (TLC) revealed different rates and patterns of S-metolachlor metabolism between 2 and 12 hours after treatment among Palmer amaranth populations. Current and future research is aimed at quantification and identification of S-metolachlor metabolites in Palmer amaranth seedlings, with and without metabolic inhibitors (NBD-Cl and malathion) pretreatment via TLC analysis. The findings from this research will expand the current knowledge of metabolic detoxification pathways and enzymes specific for weedy dicots and provide metabolic markers for VLCFA-inhibiting herbicide resistance for use in development of new chemistries for selective weed management.

Jie Fu, Brian McKinley, Galya Orr, Kankshita Swaminathan, John Mullet, Amy Marshall-Colon

Bioenergy sorghum is a low-input, drought-resilient, deep-rooting annual crop with high biomass yield potential to enable the sustainable production of biofuels, biopower, and bioproducts. Bioenergy sorghum’s 4-5 m stems account for ~80% of the harvested biomass. Stems accumulate high levels of sucrose that could be used to synthesize bioproducts if information about stem cell-type gene expression and regulation was available to enable engineering. To obtain this information, Laser Capture Microdissection was used to isolate transcriptome profiles from five major cell types present in stems of Sorghum bicolor L. Moench cv. Wray. Transcriptome analysis identified genes with cell-type specific and cell-preferred expression patterns that reflect the distinct characteristics and regulatory functions of each cell type. Analysis of cell-type specific gene regulatory networks (GRNs) revealed that unique TF families contribute to distinct regulatory landscapes, where regulation is organized through various modes and network motifs. Cell-specific transcriptome data was combined with a stem developmental transcriptome dataset to identify the GRN that differentially activates the secondary cell wall formation in stem xylem sclerenchyma and epidermal cells. The cell-type transcriptomic dataset provides a valuable source of information about the function of sorghum stems and GRNs that will enable the engineering of bioenergy sorghum stems.

Zachary Hill, Sarah Lipps, Tiffany Jamann

Fusarium graminearum is the causal agent of Gibberella Ear Rot (GER) of maize and is one of the most common diseases that affect maize in Northern United States, causing large yield losses and posing a significant mycotoxin threat to humans and livestock (Mueller er al., 2020). The most effective way of preventing F. graminearum infection is by breeding resistant varieties (Mesterházy et al., 2012).
Our hypothesis was that as breeders select for higher yield and improved kernel composition quality, they might be inadvertently selecting for GER susceptibility.
Approximately 400 maize lines from the Ames diversity panel were grown out in an augmented alpha-lattice incomplete block design in 2021 (Romay et al., 2013). Kernels from uninoculated replications were used for kernel composition analysis with Near-Infrared Reflectance Spectroscopy (NIR) using a Perten Instruments DA 7200 NIR Analysis System.
The results of our experiment showed that protein and starch were weakly correlated with disease severity. Additionally, all other kernel quality traits were not significantly correlated with disease severity.
We concluded that based on the results breeders are not inadvertently breeding for a more susceptible line to F. graminearum when they breed for different kernel quality traits. However, there is still some literature that came to a separate conclusion about kernel quality traits and susceptibility. Going forward our future research will examine mycotoxin accumulation and disease severity.

Terence Seldon Kwafo, Justin M. McGrath

Proper concentrations of several nutrients, such as iron (Fe) and zinc (Zn), in the soil are needed for plant growth. Thus, farmers sometimes evaluate soils to determine fertilizer application rates. However, measuring soil and plant nutrient concentrations is relatively time-consuming and expensive. Therefore, usually, only a few samples are collected and analyzed. This limits the scale of research but also poses some limitations on farming practices, as farmers cannot sample fields at high density in order to customize application rates. Cheap, fast, and high spatial resolution methods to measure soil nutrient concentrations would alleviate some of these limitations. This work aimed to determine the potential of hyperspectral imaging (HI) to predict the concentration of some soil nutrients. Soil samples were scanned by visible and near-infrared imaging systems with a total wavelength range of 450 – 1700 nm. Fe and Zn were analyzed.  Partial least-square regression models (PLSR) were used to correlate the relative reflectance of total Fe and Zn in the soil samples. The PLSR models could highly predict Fe concentration (R2 0.9, RMSE% 11.71) and performed moderately well for Zn (R2 0.68, RMSE% 0.57) concentrations. The overall results indicated that the hyperspectral technique coupled with PLSR could be an accurate and reliable method for determining soil nutrient concentrations.

Yudai K. Takenaka, Jed Colquhoun, Scott Lukas, Pavle Pavlovic, Barbara Scott, Mark VanGessel, Martin Williams

To understand the scope of weed problems in lima bean (Phaseolus lunatus), weed communities were surveyed near the time of crop harvest from 2019 to 2022 in two major U.S. production regions. A total of 67 fields were surveyed: 33 fields in the Midwest region and 34 fields in the Atlantic region. Crop management practices used in the fields were obtained from collaborating farmers and vegetable processors. Approximately 52 weed species were observed, with significant differences in weed communities between the two regions. In the Midwest, dominant species included amaranth species (Amaranthus spp.), carpetweed (Mollugo verticillata), common chickweed (Stellaria media), common lambsquarters (Chenopodium album), common purslane (Portulaca oleracea), fall panicum (Panicum dichotomiflorum), and velvetleaf (Abutilon theophrasti). In the Atlantic region, dominant species included annual bluegrass (Poa annua), carpetweed, common chickweed, fall panicum, ivyleaf morningglory (Ipomoea hederacea), nightshade species (Solanum spp.), and smooth pigweed (Amaranthus hybridus). Widely adopted mechanical weed control methods included preplant tillage and interrow cultivation. The most widely used preemergence herbicides were halosulfuron-methyl, pendimethalin, and S-metolachlor. The most widely used postemergence herbicides were bentazon, clethodim, imazamox, and sethoxydim. Despite adoption of several chemical and mechanical control methods, this survey revealed extensive weed problems in many production fields. Greater diversification of integrated weed management systems is needed, particularly targeting Amaranthus species.

Liz Dominguez and Anthony J. Studer

Emerging new technologies can now use fermentable sugars for the production of sustainable aviation fuels (SAF). With this, comes the demand for diverse, high yielding input sugar sources. Sorghum has the potential to supply sugars to the SAF market while also having the added advantages of stress tolerance and the ability to grow on marginal land. However, in order for sorghum to be a viable bioenergy crop, improvements need to be made to increase the quantity of sugar accumulated in the stem. Our work investigates the use of cytoplasmic male sterile (CMS) lines to increase sugar accumulation in the stem by removing the panicle as a sink tissue. To test the extent to which resources could be redirected to the stem, four CMS lines were grown alongside their fertile counterpart in replicated field trials in 2020 and 2022. Bags were placed over the panicles of CMS lines to prevent seed production and fertile lines were open pollinated. By inhibiting seed set we were able to test each line’s ability to increase stem sink strength. Our data shows that preventing seed set can alter stem sink strength; however, the effect was genotype dependent. Only some lines showed a significant difference in sugar production, indicating an underlying difference in the source-sink programming. In addition, increases could be seen in either juice volume, and/or sugar concentration in the stem. These results suggests that multiple mechanisms play a role in the concentration of sugars and accumulation of juice to the stem. Dissecting these mechanisms will be the focus of our future work to improve our understanding of source-sink relationships and allow for the development of higher yielding crops.

Li, Catherine H; Arp, Jennifer J; Heller, Nicholas; Moose, Stephen P

The Illinois Long Term Selection Experiment is the longest-running continuous genetics experiment in higher plants. More than 350 cycles of artificial selection performed over 120 years have generated distinct populations representing the phenotypic extremes of kernel protein concentration in maize. To help unravel regions of the maize genome that contributed to the rapid and dramatic responses to phenotypic selection, we have recently derived four populations of near-isogenic lines from combinations of the Illinois Protein Strains. After 6 generations of backcrossing, which included phenotypic selection for the “opposite” kernel protein concentration of the recurrent parent (e.g. higher protein in the Illinois Low Protein background), the populations visibly resemble their recurrent parents.  Kernel protein concentration was measured using both near-infrared reflectance and pink color intensity of the FLOURY2-RFP reporter transgene, an easily visualized and non-destructive marker for alpha-zeins, whose accumulation varies dramatically among the Illinois Protein Strains. One conclusion from these near-isogenic lines is that not all introgressions differed from the recurrent parent, thus not all genomic regions contribute to phenotypic responses to selection. Previous genotype-by-sequencing analyses of the Illinois Protein Strains identified 306 SNPs whose patterns of diversity showed strong correlations with both forward and reverse selection for grain protein. Near-isogenic line pairs for two regions harboring some of these SNPs and candidate genes for asparagine cycling, another pathway strongly affected by selection, displayed phenotypic changes that validate their contribution to grain protein concentration. Updated analyses using the version 5 reference genome and genotyping of these near-isogenic lines will aid in narrowing down the list of candidate genes causing the dramatic variation in kernel protein concentration.

Mae Antonette G. Mercado, Nikita Bhatnagar, Mia Sands, Steven Burgess, Donald Ort, and Anthony J. Studer

Rubisco activase (RCA) is a protein that activates Ribulose-1,5-bisphosphate carboxylase-oxygenase (RUBISCO) by removing sugar phosphates and has been hypothesized to be involved in heat stress response in plants. However, the role of RCA in heat stress has not been experimentally determined in C4 grasses like maize. Some plant species have a single Rca gene that undergoes alternative splicing, but in maize, two Rca genes encode the α and β isoforms. The β isoform is encoded by rca1 (Zm00001eb164390), while the α isoform is encoded by rca3 (Zm00001eb164380) and is up regulated during heat stress. Here we present the characterization of rca3 using both Dissociation (Ds) and Mutator (Mu) transposable element insertion alleles. Sanger Sequencing identified the site of the Ds insertion (rca3-m2::Ds) in the second exon, and the UniformMu insertion (mu1022546) in the third exon of rca3. Characterization of the mutant lines indicate that the insertions disrupted rca3 expression at 42°C. Gas exchange measurements at 42°C were used to identify variability between the response of the wildtype and mutant lines to heat stress. Leaf-level gas exchange measurements taken at three different time points (two hours after dawn, four hours after heat treatment and four hours after recovery) showed that the rca3-m2::Ds mutant plants have significantly lower photosynthetic rates after being exposed to high temperature for four hours. We also observed that the photosynthetic rate after recovery only reached 80% of its initial value. The UniformMu allele had a less severe phenotype than the Ds allele, which we attribute to the position of the insertion, which creates a knockdown rather than a knockout of rca3 expression. Our understanding of the role of RCA in heat stress will guide the improvement of crops that can be more resilient to future temperature fluctuations.

Twohey III, Robert J.; Li, Catherine H.; Hanneman, Marjorie; Wickes-Do, Liam; Gore, Michael A.; Moose, Stephen P.; Studer, Anthony J.

Stomata control CO2 and water vapor movement across the leaf surface by altering their aperture via guard cells. SLAC1 is a guard cell-specific anion channel controlling stomatal movement. Our lab characterized a null mutant slac1-2 in Zea mays that is unable to close its stomata in response to multiple environmental stimuli. Here we present the use of slac1-2 to investigate the level of stomatal control as Z. mays responds to increasing vapor pressure deficit (VPD), how stomata affect nitrogen uptake, and determine if stomata limit photosynthetic efficiency in the C4 crop Z. mays.
As atmospheric VPD increases, there is an increased demand for water from the surrounding air. Past studies have identified VPD breakpoints where the linear transpirational response to increasing VPD changes in slope. It has been hypothesized that breakpoints are controlled by hydraulic conductance. We performed VPD curves on slac1-2 to determine if a plant without stomatal control shows a VPD breakpoint due to alternate intracellular regulators of transpiration. Breakpoints were not observed in slac1-2 hybrids, indicating that stomata are the main regulators during VPD response. 
Since the slac1-2 mutant had previously only been evaluated in an inbred genetic background, single-cross hybrids homozygous for the slac1-2 mutation were produced. The slac1-2 hybrids and wild type comparisons were planted in three locations during the 2022 field season. Midday gas exchange measurements and yield data were collected from the plots to determine if a constant open stomata phenotype removes a possible CO2 photosynthetic limitation in a hybrid background. Stomatal conductance was significantly higher in the slac1-2 hybrid however, this did not result in significantly higher yields at any of the field sites. Plant nitrogen content was also measured in the hybrids, and results support slac1-2 having an impact on nitrogen accumulation and partitioning.

Mirai Inaoka, Dr. Alexander E. Lipka

Leaf angle (LA) and tassel branch number (TBN) are important traits for improving grain yield in maize (Zea mays L.). To assist plant biologists with understanding which genes control LA and TBN and then incorporate this knowledge into breeding decisions, it is crucial to pinpoint which regions of the maize genome are most likely to contain loci associated with these two traits. We are currently performing a genome-wide association study (GWAS) to identify a subset of markers across the maize genome that exhibit statistically significant associations with LA and TBN. Thus far, the observed results indicate that there are statistically significant marker-trait associations for LA and TBN, suggesting the presences of several nearby large-effect genes. These results will aid in plant breeders’ efforts in the future when they are breeding for maize plants that contain the appropriate ratio of LA to TBN to ensure high grain yield.

Isabel S. Werle, Lucas K. Bobadilla, Aaron G. Hager, and Patrick J. Tranel

S-metolachlor is an herbicide typically applied to corn and soybean fields for waterhemp (Amaranthus tuberculatus Moq.) control in the Midwestern United States. In Illinois, a waterhemp population denominated CHR has evolved metabolic resistance to S-metolachlor. In this study, we investigated the inheritance of S-metolachlor resistance in CHR. Whole plant dose-response was conducted with four waterhemp populations including resistant (R) and sensitive (S) parental lines, and two F1 lines, designated F1-4 and F1-9. S-metolachlor rates ranged from 8.5 to 8,456 g ai ha-1. Survival counts were taken 21 days after treatment. Data from the reciprocal F1 lines did not differ and therefore were pooled. The effective dose needed to reduce waterhemp emergence by 50% (ED50) was determined using a three-parameter Weibull II model with the drc package in R. The dominance degree (D) of the resistant trait was calculated based on the ED50 values of each population. Dose-response analysis revealed an ED50 of 105 g ai ha-1 of S-metolachlor for the S population. The ED50 of the R population and the pooled-F1 line were 476 and 261 g ai ha-1, respectively. The R population was 5-fold more resistant than the S population, but a lower resistance level (2-fold) was recorded for the pooled-F1 line. The degree of dominance resulted in D=0.20, indicating that S-metolachlor resistance in CHR is incompletely dominant. The similar inheritance patterns observed in the two F1 lines suggest that the resistance is behaving as a nuclear-encoded trait. S-metolachlor segregation in F2 and backcross lines is currently being investigated.

Rashmi Dangol, Nicholas Heller

In the U.S. Midwest, crop production systems rely heavily on corn-soybean rotation which includes low-diversity cropping systems with high fertilizer inputs that leave the land fallow for around eight months of the year. Integrating an annual winter crop into the summer-annual-dominant cropping system could be an approach to increase diversity and sustainability in farming practices providing both economic benefits and ecosystem services. Pennycress (Thlaspi arvense) has been developed as a winter annual cash cover crop oilseed that can be integrated into a traditional corn-soybean rotation. Despite its potential as feedstock and biodiesel production, limited information is known about the best agronomic management practices for its high production. The objective of the study is to optimize when and how we can grow pennycress between different crop rotation systems to maximize crop productivity and oilseed yields. This study is conducted at Illinois State University Research Farm, Lexington using a randomized block experimental design to compare four treatments systems: Cereal Rye-Pennycress-Soybean, Silage Corn-Pennycress-Soybean, Grain Corn-Pennycress-Soybean, and Soybean-Pennycress-Corn. The annual summer crops were grown in spring 2022, and pennycress was drilled in fall 2022. Following the pennycress, soybean and maize will be planted as summer cash crops. Cash crops biomass and yield and pennycress spring canopy cover, biomass, yield, and the total net income of each plot will be evaluated to compare these systems.

Dilkaran Singh, Kavya Kannan, Amy Marshall-Colon

Rising atmospheric CO2 levels directly affect the C and N metabolism and their interactions. Elevated CO2 grown C3 plants has altered allocation of N causing reduced investment in Rubisco and has reduced carboxylation rate (Vc,max). These changes contribute to the negative acclimation of photosynthetic capacity. A decline in plant N concentration is also observed under elevated CO2 leading to decreased nutritional quality of C3 plants. Altering genetic factors controlling CxN metabolic crosstalks could provide genotypes better suited for high CO2 conditions, i.e., has little to no acclimation of photosynthetic capacity and maintain N concentration similar to ambient CO2-grown C3 plants.
We constructed a gene regulatory network of Arabidopsis plants grown under CO2 and N treatments.  We found bZIP1 as one of the regulators of CxN responsive genes. Next, we analyzed the overexpression of bZIP1 TF experimentally. bZIP1-overexpression (bZIP1-OX) line accumulated more plant biomass under elevated CO2 conditions than wildtype plants, while N and protein concentration remained similar between the two genotypes. RNA seq analysis of bZIP1-OX plants revealed CxN responsive genes differentially regulated by bZIP1-OX are enriched in Jasmonic Acid (JA) signaling and synthesis pathways. Further, co-expression network analysis was utilized to identify the gene modules correlated with above-ground biomass. Modules having a significant positive correlation with biomass constitute genes related to JA-related processes.
In conclusion, our study shows bZIP1 as a potential target for biomass increase under elevated CO2 & biomass increase, and JA-related genes’ expression are correlated in bZIP1-OX plants.

Ximin Piao, Edward B Lochocki, Megan L. Matthews

Most plant models allocate photo-assimilated carbon (C), typically in the form of sugars, among organs based on empirical partitioning tables, ignoring the structural mechanisms resulting in the C allocation. This undermines the predictive power of the model under changing conditions, such as water stress and rising CO2 level. Thornley (1972) proposed a phenomenological model where the C partitioning is determined by the substrate utilization and translocation processes. We extended Thornley’s model to include the reproduction and senescence stages triggered by the development indices, and this new partitioning model is referred to as the Thornley model.
BioCro is a crop growth simulation platform that integrates models of key physiological and biophysical processes (Humphries and Long, 1995). We integrated the Thornley model into the BioCro framework. The Thornley model parameters were trained with the biomass data from soybean grown at the ambient CO2 level at the SoyFACE facility in 2002 and 2004. It was then tested on data from soybean grown at the ambient level in 2004 and 2006, and four years (2002, 2004, 2005, and 2006) at the elevated CO2 level at the SoyFACE facility. The Thornley model was also tested on data from a different soybean cultivar grown at the Energy Farm in 2021 and 2022, using updated photosynthetic parameters but the same partitioning parameters. Preliminary results have shown that the Thornley model performs as well as the partitioning model for the soybean grown at SoyFACE and outperforms the partitioning model for the cultivar grown at the Energy Farm.

Montana Hernandez, Dr. Andrew Leakey, Dr. Daniel Tejeda-Lunn

The ratio of carbon assimilation and water loss through transpiration is termed intrinsic water use efficiency (iWUE) and is a major target for crop improvement. Stomata are tiny pores on the leaf surface that open and close to control the trade-off between CO2 uptake and water loss. In C4 grasses, atmospheric CO2 concentrations are saturating for photosynthesis meaning we can lower iWUE by reducing the leaf area available for transpiration. Reducing stomatal density (StoDen) by manipulating developmental regulators is one way to lower the leaf area available for transpiration. We expressed a novel gene identified by transcriptome analysis of Sorghum Bicolor (Sb002g35600) in our target crop Saccharum officinarum (sugarcane) to determine the impact on epidermal patterning, stomatal density, and gas exchange. Expressing Sb002g35600 in sugarcane reduced stomatal density by 16% on the adaxial surface and 18% on the abaxial leaf surface, due to there being fewer stomata per cell file. However, there was no change in gas exchange, leaf physiology, or the length of stomatal complexes. These data suggest that gas exchange remains constant via a compensatory mechanism of stomatal opening and closing at lower stomatal density.

Raysa Gevartosky, Jessica Rutkoski

To meet the growing demand for food and to improve the profitability and sustainability of agriculture, yield gains in wheat must accelerate. Genomic selection (GS) and high-throughput phenotyping (HTP) can potentially improve rates of genetic gain for yield in breeding programs. However, new breeding strategies that can leverage the potential of these technologies and increase overall breeding efficiency are needed. Obtaining high GS accuracies for yield remains challenging because testing large numbers of lines needed for accurate model training is costly. To address this problem, we developed a Phenomic Assisted Genomic Selection approach that relies on HTP-based yield imputation, using k-Nearest Neighbor, Random Forest, and RuleFit machine learning algorithms, to help increase the size of the GS training set without increasing costs. To begin testing this approach, we used yield trials data from conventional phenotyping and HTP across three locations in the state of Illinois, and 2,420 unique entries to evaluate phenotype imputation accuracies for yield. We found moderate to high phenotype imputation accuracies, indicating that imputed phenotypes could be useful for GS model training. If successful, Phenomic Assisted Genomic Selection could nearly double the selection accuracy among un-tested breeding candidates without increasing costs, thereby increasing rates of genetic gain. This study will help breeding programs utilize HTP and GS synergistically, as a result enhancing crop improvement both regionally and globally.

Jeremy Logrono, Eric Olson, Jessica Rutkoski

The University of Illinois at Urbana-Champaign Small Grains Breeding Lab develops cutting-edge methods to rapidly develop new soft red varieties and increase genetic gains per year. Currently, we are developing a speed breeding protocol that incorporates advancing F2 populations using ‘Modified Minibulk’ system followed by advancement of space planted F3 single plants that are vernalized indoors and transplanted into the field in the spring. The Modified Minibulk system uses 22-hr photoperiod length and high density seeding (0.7 seeds per cm²) in the greenhouse. The F3 single plant nursery is seeded into a 264-cell Paper Chain Pots (one seed per cell) to allow for vernalization. In spring, a paper pot transplanting system is used to transplant F3 plants, achieving 30.48 cm spacing between single plants. Whole plants are selected and harvested in late summer and seeds are planted in a 5m²  plot for seed increase. The entire speed breeding protocol can potentially replace the conventional bulk breeding and headrows system for deriving F3:F4 lines, saving two years in the line development pipeline and producing sufficient seeds for multi-location testing one year earlier.  Early results from this system demonstrate that it takes 132 days from seed to seed (including the 10-week vernalization) to complete the F2 generation and harvest the F3 bulk seeds in the greenhouse.  The 2022 F3 single plant yields from the field nursery ranged from 5.8 – 24.5 g with mean of 9.51 g (n = 651).  Our goal is to continue improving our methods to maximize the amount of seed that can be generated from a single plant for the seed increase generation.

Mara Krone & Santiago Mideros

Fusarium graminearum causes Fusarium Head Blight (FHB) in wheat, producing significant yield losses and mycotoxin contamination. Quantitative host resistance is the best method to control FHB. However, there is a lack of understanding on how disease resistance effects the evolution of plant pathogens. This study aimed to determine if the wheat’s level of resistance influenced F. graminearum aggressiveness and to identify genes under selection due to host resistance. Thirty-one isolates from highly susceptible and 26 isolates from moderately resistant wheat lines were used. The isolate’s aggressiveness was measured by AUDPC, visually damaged kernels, and deoxynivalenol contamination. DNA was extracted and sequenced to create a SNP data set that was used to perform a whole-genome scan for selection and a phylogenetic analysis. The results of this research found that the population of F. graminearum from susceptible wheat had more aggressive isolates. This led to the conclusion that resistant wheat varieties have not selected for a population with primarily aggressive isolates but mixed levels of aggressiveness. The population of isolates from highly susceptible wheat were primarily aggressive suggesting that on a susceptible host aggressive isolates might be outcompeting weaker isolates. The level of wheat resistance added selection pressure to the F. graminearum isolates as both populations had unique regions of the genome and genes under selection. The fixed level of isolate aggressiveness and the increased number of genes under selection suggests that there is stronger selection pressure occurring in the population from highly susceptible wheat.

Matthew D. Murphy, Samuel B. Fernandes, Gota Morota, Alexander E. Lipka

Identifying specific genomic regions contributing to yield stability could help us select cultivars that thrive in harsher and more unpredictable environmental conditions. Identifying these variance-controlling loci has been elusive due to the conservativeness arising from testing for genotype-by-environment (GxE) interactions that could underlie these genomic regions. While variance genome-wide association studies (vGWAS) have successfully identified GxE loci in human and livestock species, their use in plants has been limited. To effectively use vGWAS to search for GxE loci in plants, it is vital to assess their performance with both real and simulated plant data. In this study, we used publicly available genome-wide marker data from Arabidopsis and maize to simulate traits controlled by GxE interactions. We evaluated the true and false-detection rates of two commonly used vGWAS approaches: the Brown-Forsythe Test (BFT) and the double generalized linear model (DGLM). Both methods had high true positive detection rates for simulated GxE loci, but the sample size was crucial in identifying these interactions. We also examined how well these vGWAS approaches detected GxE interactions in two flowering-time traits in maize across two different environments. Although we detected statistically significant associations after a genome-wide Bonferroni correction at α=0.05, we also observed many false-positive associations. We recommend that future studies use larger genotypic datasets (i.e., n > 2,815) and more sophisticated vGWAS tests that adequately control for false positives; otherwise, the full potential of vGWAS to identify GxE loci will not be realized.

Sarah Lipps, Carolyn Butts-Wilmsmeyer, Martin Bohn, Tiffany Jamann

Host resistance is one of the most effective forms of disease management. However, in some pathosystems, such as the Fusarium graminearum – maize system, understanding the genetic basis of resistance is challenging due to the complex relationship between quantitative disease resistance, environmental factors, and kernel composition. F. graminearum is the causal agent of Gibberella ear rot in maize, a disease of high economic importance. The fungus colonizes the ear and produces the carcinogenic mycotoxins deoxynivalenol, zealaranone, and nivealnol, which are harmful to both humans and livestock when consumed. Our main objectives are to use an association mapping approach to identify putative regions of the maize genome associated with resistance to F. graminearum in diverse germplasm, and kernel quality information to understand the impact of physical kernel structure on F. graminearum and the fungus’ ability to colonize the ear. In 2021, we inoculated and screened 320 inbred lines from the Ames diversity panel for resistance to F. graminearum. While disease levels varied between environment we estimated a heritability of 0.74. Using a genome wide association study approach, we identified five significant single nucleotide polymorphisms associated with resistance. Kernel quality data on healthy seed were collected using near-infrared spectroscopy (NIR). NIR provides insight into the composition of kernels including starch, protein, ash, oil, and fiber. In the future we plan to examine toxin accumulation and phenylpropanoid profiles on all lines screened for disease. The results from our study will improve the understanding of resistance to F. graminearum in maize and can be used to develop resistant germplasm.

Lucas Berger Munaro, Fred Kolb, Jessica Rutkoski

In Illinois, wheat is mostly cultivated in a wheat-soybean double crop system and The University of Illinois wheat breeding program aims to improve the profitability of wheat-soybean double cropping by developing high-yielding, early maturing wheat varieties with acceptable grain quality. Our objective was to estimate the genetic trends for grain yield, test weight, heading date, and plant height in the past 21 years of breeding at the University of Illinois to determine if these trends align with our current breeding goals. We retrieved the phenotypic data available from advanced yield trials, where 1690 genotypes were evaluated from 2001 to 2021 in 5 locations, resulting in a total of 67,349 observations. The presence of two check genotypes allowed us to use the control population method to estimate the genetic trends, in which the check genotypes are used to determine whether phenotypic changes over time are due to genetic or non-genetic causes. Using mixed models in ASReml-R, we identified a positive genetic trend for grain yield (33-33.6 kg ha-1) and test weight (0.82-0.91 g L-1). For heading date and plant height genetic trends were not significantly different from zero. Our study revealed that The University of Illinois wheat breeding program has been contributing to improve the profitability of wheat-soybean double cropping. Progress have been made by developing germplasm with higher grain yield and test weight without extending the growth cycle.