Characterization of adaptation mechanisms in sorghum using a multireference back-cross nested association mapping design and envirotyping.

Identifying the genetic factors impacting the adaptation of crops to environmental conditions is of key interest for conservation and selection purposes. It can be achieved using population genomics, and evolutionary or quantitative genetics. Here we present a sorghum multireference back-cross nested association mapping population composed of 3,901 lines produced by crossing 24 diverse parents to 3 elite parents from West and Central Africa-back-cross nested association mapping. The population was phenotyped in environments characterized by differences in photoperiod, rainfall pattern, temperature levels, and soil fertility. To integrate the multiparental and multi-environmental dimension of our data we proposed a new approach for quantitative trait loci (QTL) detection and parental effect estimation. We extended our model to estimate QTL effect sensitivity to environmental covariates, which facilitated the integration of envirotyping data. Our models allowed spatial projections of the QTL effects in agro-ecologies of interest. We utilized this strategy to analyze the genetic architecture of flowering time and plant height, which represents key adaptation mechanisms in environments like West Africa. Our results allowed a better characterization of well-known genomic regions influencing flowering time concerning their response to photoperiod with Ma6 and Ma1 being photoperiod-sensitive and the region of possible candidate gene Elf3 being photoperiod-insensitive. We also accessed a better understanding of plant height genetic determinism with the combined effects of phenology-dependent (Ma6) and independent (qHT7.1 and Dw3) genomic regions. Therefore, we argue that the West and Central Africa-back-cross nested association mapping and the presented analytical approach constitute unique resources to better understand adaptation in sorghum with direct application to develop climate-smart varieties.

Do We Need to Breed for Regional Adaptation in Soybean?-Evaluation of Genotype-by-Location Interaction and Trait Stability of Soybean in Germany.

Soybean is a crop in high demand, in particular as a crucial source of plant protein. As a short-day plant, soybean is sensitive to the latitude of the growing site. Consequently, varieties that are well adapted to higher latitudes are required to expand the cultivation. In this study, we employed 50 soybean genotypes to perform a multi-location trial at seven locations across Germany in 2021. Two environmental target regions were determined following the latitude of the locations. Adaptation and trait stability of seed yield and protein content across all locations were evaluated using Genotype plus Genotype-by-Environment (GGE) biplots and Shukla's stability variance. We found a moderate level of crossing-over type genotype-by-location interaction across all locations. Within the environmental target regions, the genotype-by-location interaction could be minimised. Despite the positive correlation (R = 0.59) of seed yield between the environmental target regions and the same best-performing genotype, the genotype rankings differed in part substantially. In conclusion, we found that soybean can be grown at a wide range of latitudes across Germany. However, the performance of genotypes differed between the northern and southern locations, with an 18.8% higher mean yield in the south. This in combination with the observed rank changes of high-performing genotypes between both environmental target regions suggests that selection targeted towards environments in northern Germany could improve soybean breeding for those higher latitude regions.

The genetic architecture of soybean photothermal adaptation to high latitudes.

Soybean is a major plant protein source for both human food and animal feed, but to meet global demands as well as a trend towards regional production, soybean cultivation needs to be expanded to higher latitudes. In this study, we developed a large diversity panel consisting of 1503 early-maturing soybean lines and used genome-wide association mapping to dissect the genetic architecture underlying two crucial adaptation traits, flowering time and maturity. This revealed several known maturity loci, E1, E2, E3, and E4, and the growth habit locus Dt2 as causal candidate loci, and also a novel putative causal locus, GmFRL1, encoding a homolog of the vernalization pathway gene FRIGIDA-like 1. In addition, the scan for quantitative trait locus (QTL)-by-environment interactions identified GmAPETALA1d as a candidate gene for a QTL with environment-dependent reversed allelic effects. The polymorphisms of these candidate genes were identified using whole-genome resequencing data of 338 soybeans, which also revealed a novel E4 variant, e4-par, carried by 11 lines, with nine of them originating from Central Europe. Collectively, our results illustrate how combinations of QTL and their interactions with the environment facilitate the photothermal adaptation of soybean to regions far beyond its center of origin.

Unraveling the potential of phenomic selection within and among diverse breeding material of maize (Zea mays L.).

Genomic selection is a well-investigated approach that facilitates and supports selection decisions for complex traits and has meanwhile become a standard tool in modern plant breeding. Phenomic selection has only recently been suggested and uses the same statistical procedures to predict the targeted traits but replaces marker data with near-infrared spectroscopy data. It may represent an attractive low-cost, high-throughput alternative but has not been sufficiently studied until now. Here, we used 400 genotypes of maize (Zea mays L.) comprising elite lines of the Flint and Dent heterotic pools as well as 6 Flint landraces, which were phenotyped in multienvironment trials for anthesis-silking-interval, early vigor, final plant height, grain dry matter content, grain yield, and phosphorus concentration in the maize kernels, to compare the predictive abilities of genomic as well as phenomic prediction under different scenarios. We found that both approaches generally achieved comparable predictive abilities within material groups. However, phenomic prediction was less affected by population structure and performed better than its genomic counterpart for predictions among diverse groups of breeding material. We therefore conclude that phenomic prediction is a promising tool for practical breeding, for instance when working with unknown and rather diverse germplasm. Moreover, it may make the highly monopolized sector of plant breeding more accessible also for low-tech institutions by combining well established, widely available, and cost-efficient spectral phenotyping with the statistical procedures elaborated for genomic prediction - while achieving similar or even better results than with marker data.

High-resolution association mapping with libraries of immortalized lines from ancestral landraces.

KEY MESSAGE: Association mapping with immortalized lines of landraces offers several advantages including a high mapping resolution, as demonstrated here in maize by identifying the causal variants underlying QTL for oil content and the metabolite allantoin. Landraces are traditional varieties of crops that present a valuable yet largely untapped reservoir of genetic variation to meet future challenges of agriculture. Here, we performed association mapping in a panel comprising 358 immortalized maize lines from six European Flint landraces. Linkage disequilibrium decayed much faster in the landraces than in the elite lines included for comparison, permitting a high mapping resolution. We demonstrate this by fine-mapping a quantitative trait locus (QTL) for oil content down to the phenylalanine insertion F469 in DGAT1-2 as the causal variant. For the metabolite allantoin, related to abiotic stress response, we identified promoter polymorphisms and differential expression of an allantoinase as putative cause of variation. Our results demonstrate the power of this approach to dissect QTL potentially down to the causal variants, toward the utilization of natural or engineered alleles in breeding. Moreover, we provide guidelines for studies using ancestral landraces for crop genetic research and breeding.

The performance of phenomic selection depends on the genetic architecture of the target trait.

KEY MESSAGE: The phenomic predictive ability depends on the genetic architecture of the target trait, being high for complex traits and low for traits with major QTL. Genomic selection is a powerful tool to assist breeding of complex traits, but a limitation is the costs required for genotyping. Recently, phenomic selection has been suggested, which uses spectral data instead of molecular markers as predictors. It was shown to be competitive with genomic prediction, as it achieved predictive abilities as high or even higher than its genomic counterpart. The objective of this study was to evaluate the performance of phenomic prediction for triticale and the dependency of the predictive ability on the genetic architecture of the target trait. We found that for traits with a complex genetic architecture, like grain yield, phenomic prediction with NIRS data as predictors achieved high predictive abilities and performed better than genomic prediction. By contrast, for mono- or oligogenic traits, for example, yellow rust, marker-based approaches achieved high predictive abilities, while those of phenomic prediction were very low. Compared with molecular markers, the predictive ability obtained using NIRS data was more robust to varying degrees of genetic relatedness between the training and prediction set. Moreover, for grain yield, smaller training sets were required to achieve a similar predictive ability for phenomic prediction than for genomic prediction. In addition, our results illustrate the potential of using field-based spectral data for phenomic prediction. Overall, our result confirmed phenomic prediction as an efficient approach to improve the selection gain for complex traits in plant breeding.

Can we abandon phosphorus starter fertilizer in maize? Results from a diverse panel of elite and doubled haploid landrace lines of maize (Zea mays L.).

The importance of phosphorus (P) in agriculture contrasts with the negative environmental impact and the limited resources worldwide. Reducing P fertilizer application by utilizing more efficient genotypes is a promising way to address these issues. To approach this, a large panel of maize (Zea mays L.) comprising each 100 Flint and Dent elite lines and 199 doubled haploid lines from six landraces was assessed in multi-environment field trials with and without the application of P starter fertilizer. The treatment comparison showed that omitting the starter fertilizer can significantly affect traits in early plant development but had no effect on grain yield. Young maize plants provided with additional P showed an increased biomass, faster growth and superior vigor, which, however, was only the case under environmental conditions considered stressful for maize cultivation. Importantly, though the genotype-by-treatment interaction variance was comparably small, there is genotypic variation for this response that can be utilized in breeding. The comparison of elite and doubled haploid landrace lines revealed a superior agronomic performance of elite material but also potentially valuable variation for early traits in the landrace doubled haploid lines. In conclusion, our results illustrate that breeding for P efficient maize cultivars is possible towards a reduction of P fertilizer in a more sustainable agriculture.

Genetic Dissection of Phosphorus Use Efficiency in a Maize Association Population under Two P Levels in the Field.

Phosphorus (P) deficiency is an important challenge the world faces while having to increase crop yields. It is therefore necessary to select maize (Zea may L.) genotypes with high phosphorus use efficiency (PUE). Here, we extensively analyzed the biomass, grain yield, and PUE-related traits of 359 maize inbred lines grown under both low-P and normal-P conditions. A significant decrease in grain yield per plant and biomass, an increase in PUE under low-P condition, as well as significant correlations between the two treatments were observed. In a genome-wide association study, 49, 53, and 48 candidate genes were identified for eleven traits under low-P, normal-P conditions, and in low-P tolerance index (phenotype under low-P divided by phenotype under normal-P condition) datasets, respectively. Several gene ontology pathways were enriched for the genes identified under low-P condition. In addition, seven key genes related to phosphate transporter or stress response were molecularly characterized. Further analyses uncovered the favorable haplotype for several core genes, which is less prevalent in modern lines but often enriched in a specific subpopulation. Collectively, our research provides progress in the genetic dissection and molecular characterization of PUE in maize.

Genome-wide association study for deoxynivalenol production and aggressiveness in wheat and rye head blight by resequencing 92 isolates of Fusarium culmorum.

BACKGROUND: Fusarium culmorum is an important pathogen causing head blight of cereals in Europe. This disease is of worldwide importance leading to reduced yield, grain quality, and contamination by mycotoxins. These mycotoxins are harmful for livestock and humans; therefore, many countries have strict regulatory limits for raw materials and processed food. Extensive genetic diversity is described among field populations of F. culmorum isolates for aggressiveness and production of the trichothecene mycotoxin deoxynivalenol (DON). However, the causes for this quantitative variation are not clear, yet. We analyzed 92 isolates sampled from different field populations in Germany, Russia, and Syria together with an international collection for aggressiveness and DON production in replicated field experiments at two locations in two years with two hosts, wheat and rye. The 30x coverage whole-genome resequencing of all isolates resulted in the identification of 130,389 high quality single nucleotide polymorphisms (SNPs) that were used for the first genome-wide association study in this phytopathogenic fungus. RESULTS: In wheat, 20 and 27 SNPs were detected for aggressiveness and DON content, respectively, of which 10 overlapped. Additionally, two different SNPs were significantly associated with aggressiveness in rye that were among those SNPs being associated with DON production in wheat. Most of the SNPs explained only a small proportion of genotypic variance (pG), however, four SNPs were associated with major quantitative trait loci (QTLs) with pG ranging from 12 to 48%. The QTL with the highest pG was involved in DON production and associated with a SNP most probably located within the Tri4 gene. CONCLUSIONS: The diversity of 92 isolates of F. culmorum were captured using a heuristic approach. Key phenotypic traits, SNPs, and candidate genes underlying aggressiveness and DON production were identified. Clearly, many QTLs are responsible for aggressiveness and DON content in wheat, both traits following a quantitative inheritance. Several SNPs involved in DON metabolism, among them the Tri4 gene of the trichothecene pathway, were inferred as important source of variation in fungal aggressiveness. Using this information underlying the phenotypic variation will be of paramount importance in evaluating strategies for successful resistance breeding.

Training set design in genomic prediction with multiple biparental families.

Genomic selection is a powerful tool to reduce the cycle length and enhance the genetic gain of complex traits in plant breeding. However, questions remain about the optimum design and composition of the training set. In this study, we used 944 soybean [Glycine max (L.) Merr.] recombinant inbred lines from eight families derived through a partial-diallel mating design among five parental lines. The cross-validated prediction accuracies for the six traits seed yield, 1,000-seed weight, protein yield, plant height, protein content, and oil content were high, ranging from 0.79 to 0.87. We investigated among-family predictions, making use of the special mating design with different degrees of relatedness among families. Generally, the prediction accuracy decreased from full-sibs to half-sib families to unrelated families. However, half-sib and unrelated families also showed substantial variation in their prediction accuracy for a given family, which appeared to be caused at least in part by the shared segregation of quantitative trait loci in both the training and prediction sets. Combining several half-sib families in composite training sets generally led to an increase in the prediction accuracy compared with the best family alone. The prediction accuracy increased with the size of the training set, but for comparable prediction accuracy, substantially more half-sibs were required than full-sibs. Collectively, our results highlight the potential of genomic selection for soybean breeding and, in a broader context, illustrate the importance of the targeted design of the training set.

Genetic architecture underlying the expression of eight α-amylase trypsin inhibitors.

KEY MESSAGE: Wheat cultivars largely differ in the content and composition of ATI proteins, but heritability was quite low for six out of eight ATIs. The genetic architecture of ATI proteins is built up of few major and numerous small effect QTL. Amylase trypsin inhibitors (ATIs) are important allergens in baker's asthma and suspected triggers of non-celiac wheat sensitivity (NCWS) inducing intestinal and extra-intestinal inflammation. As studies on the expression and genetic architecture of ATI proteins in wheat are lacking, we evaluated 149 European old and modern bread wheat cultivars grown at three different field locations for their content of eight ATI proteins. Large differences in the content and composition of ATIs in the different cultivars were identified ranging from 3.76 pmol for ATI CM2 to 80.4 pmol for ATI 0.19, with up to 2.5-fold variation in CM-type and up to sixfold variation in mono/dimeric ATIs. Generally, heritability estimates were low except for ATI 0.28 and ATI CM2. ATI protein content showed a low correlation with quality traits commonly analyzed in wheat breeding. Similarly, no trends were found regarding ATI content in wheat cultivars originating from numerous countries and decades of breeding history. Genome-wide association mapping revealed a complex genetic architecture built of many small, few medium and two major quantitative trait loci (QTL). The major QTL were located on chromosomes 3B for ATI 0.19-like and 6B for ATI 0.28, explaining 70.6 and 68.7% of the genotypic variance, respectively. Within close physical proximity to the medium and major QTL, we identified eight potential candidate genes on the wheat reference genome encoding structurally related lipid transfer proteins. Consequently, selection and breeding of wheat cultivars with low ATI protein amounts appear difficult requiring other strategies to reduce ATI content in wheat products.

Optimizing the P balance: How do modern maize hybrids react to different starter fertilizers?

Phosphorus (P) is an essential macronutrient for plants, but also a limited resource worldwide. Strict regulations for fertilizer applications in the European Union are a consequence of the negative environmental effects in case of improper use. Maize is typically grown with the application of P starter fertilizer, which, however, might be reduced or even omitted if suitable varieties were available. This study was performed with the 20 commercially most important maize hybrids in Germany evaluated in multi-location field trials with the aim to investigate the potential to breed for high-performing maize hybrids under reduced P starter fertilizer. At the core location, three starter fertilizers with either phosphate (triple superphosphate, TSP), ammonium nitrate (calcium ammonium nitrate, CAN), or a combination of ammonium and phosphate (diammonium phosphate, DAP) were evaluated relative to a control and traits from youth development to grain yield were assessed. Significant differences were mainly observed for the DAP starter fertilizer, which was also reflected in a yield increase of on average +0.67 t/ha (+5.34%) compared to the control. Correlations among the investigated traits varied with starter fertilizer, but the general trends remained. As expected, grain yield was negatively correlated with grain P concentration, likely due to a dilution effect. Importantly, the genotype-by-starter fertilizer interaction was always non-significant in the multi-location analysis. This indicates that best performing genotypes can be identified irrespective of the starter fertilizer. Taken together, our results provide valuable insights regarding the potential to reduce starter fertilizers in maize cultivation as well as for breeding maize for P efficiency under well-supplied conditions.

Genetic Architecture of Cereal Leaf Beetle Resistance in Wheat.

Wheat production can be severely damaged by endemic and invasive insect pests. Here, we investigated resistance to cereal leaf beetle in a panel of 876 winter wheat cultivars, and dissected the genetic architecture underlying this insect resistance by association mapping. We observed an effect of heading date on cereal leaf beetle infestation, with earlier heading cultivars being more heavily infested. Flag leaf glaucousness was also found to be correlated with resistance. In line with the strong effect of heading time, we identified Ppd-D1 as a major quantitative trait locus (QTL), explaining 35% of the genotypic variance of cereal leaf beetle resistance. The other identified putative QTL explained much less of the genotypic variance, suggesting a genetic architecture with many small-effect QTL, which was corroborated by a genomic prediction approach. Collectively, our results add to our understanding of the genetic control underlying insect resistances in small-grain cereals.

Speed breeding short-day crops by LED-controlled light schemes.

KEY MESSAGE: A simple and rapid speed breeding system was developed for short-day crops that enables up to five generations per year using LED lighting systems that allow very specific adjustments regarding light intensity and quality. Plant breeding is a key element for future agricultural production that needs to cope with a growing human population and climate change. However, the process of developing suitable cultivars is time-consuming, not least because of the long generation times of crops. Recently, speed breeding has been introduced for long-day crops, but a similar protocol for short-day crops is lacking to date. In this study, we present a speed breeding protocol based on light-emitting diodes (LEDs) that allow to modify light quality, and exemplarily demonstrate its effectiveness for the short-day crops soybean (Glycine max), rice (Oryza sativa) and amaranth (Amaranthus spp.). Adjusting the photoperiod to 10 h and using a blue-light enriched, far-red-deprived light spectrum facilitated the growth of short and sturdy soybean plants that flowered ~ 23 days after sowing and matured within 77 days, thus allowing up to five generations per year. In rice and amaranth, flowering was achieved ~ 60 and ~ 35 days after sowing, respectively. Interestingly, the use of far-red light advanced flowering by 10 and 20 days in some amaranth and rice genotypes, respectively, but had no impact on flowering in soybeans, highlighting the importance of light quality for speed breeding protocols. Taken together, our short-day crops' speed breeding protocol enables several generations per year using crop-specific LED-based lighting regimes, without the need of tissue culture tools such as embryo rescue. Moreover, this approach can be readily applied to a multi-storey 96-cell tray-based system to integrate speed breeding with genomics, toward a higher improvement rate in breeding.

Misexpression of a transcriptional repressor candidate provides a molecular mechanism for the suppression of awns by Tipped 1 in wheat.

Awns are bristle-like structures formed at the tip of the lemma on the florets of some cereal grasses. Wild-type wheat is awned, but awnletted and awnless variants have been selected and nowadays all forms are cultivated. In this study, we dissected the genetic control underlying variation of this characteristic feature by association mapping in a large panel of 1110 winter wheat cultivars of worldwide origin. We identified the B1 (Tipped 1) locus on chromosome 5A as the major determinant of awnlessness globally. Using a combination of fine-mapping and expression analysis, we identified a putative C2H2 zinc finger protein with an EAR domain, characteristic of transcriptional repressors, as a likely candidate for Tipped 1. This gene was found to be up-regulated in awnless B1 compared with awned b1 plants, indicating that misexpression of this transcriptional regulator may contribute to the reduction of awn length in B1 plants. Taken together, our study provides an entry point towards a better molecular understanding of the evolution of morphological features in cereals through selection and breeding.

Refining the genetic architecture of flag leaf glaucousness in wheat.

KEY MESSAGE: The cuticle is the plant's barrier against abiotic and biotic stresses, and the deposition of epicuticular wax crystals results in the scattering of light, an effect termed glaucousness. Here, we dissect the genetic architecture of flag leaf glaucousness in wheat toward a future targeted design of the cuticle. The cuticle serves as a barrier that protects plants against abiotic and biotic stresses. Differences in cuticle composition can be detected by the scattering of light on epicuticular wax crystals, which causes a phenotype termed glaucousness. In this study, we dissected the genetic architecture of flag leaf glaucousness in a panel of 1106 wheat cultivars of global origin. We observed a large genotypic variation, but the geographic pattern suggests that other wax layer characteristics besides glaucousness may be important in conferring tolerance to abiotic stresses such as heat and drought. Genome-wide association mapping identified two major quantitative trait loci (QTL) on chromosomes 3A and 2B. The latter corresponds to the W1 locus, but further characterization revealed that it is likely to contain additional QTL. The same holds true for the major QTL on 3A, which was also found to show an epistatic interaction with another locus located a few centiMorgan distal to it. Genome-wide prediction and the identification of a few additional putative QTL revealed that small-effect QTL also contribute to the trait. Collectively, our results illustrate the complexity of the genetic control of flag leaf glaucousness, with additive effects and epistasis, and lay the foundation for the cloning of the underlying genes toward a more targeted design of the cuticle by plant breeding.

Copy number variation of Ppd-B1 is the major determinant of heading time in durum wheat.

BACKGROUND: Heading time is an important adaptive trait in durum wheat. In hexaploid wheat, Photoperiod-1 (Ppd) loci are essential regulators of heading time, with Ppd-B1 conferring photoperiod insensitivity through copy number variations (CNV). In tetraploid wheat, the D-genome Ppd-D1 locus is absent and generally, our knowledge on the genetic architecture underlying heading time lacks behind that of bread wheat. RESULTS: In this study, we employed a panel of 328 diverse European durum genotypes that were evaluated for heading time at five environments. Genome-wide association mapping identified six putative QTL, with a major QTL on chromosome 2B explaining 26.2% of the genotypic variance. This QTL was shown to correspond to copy number variation at Ppd-B1, for which two copy number variants appear to be present. The higher copy number confers earlier heading and was more frequent in the heat and drought prone countries of lower latitude. In addition, two other QTL, corresponding to Vrn-B3 (TaFT) and Ppd-A1, were found to explain 9.5 and 5.3% of the genotypic variance, respectively. CONCLUSIONS: Our results revealed the yet unknown role of copy number variation of Ppd-B1 as the major source underlying the variation in heading time in European durum wheat. The observed geographic patterns underline the adaptive value of this polymorphism and suggest that it is already used in durum breeding to tailor cultivars to specific target environments. In a broader context our findings provide further support for a more widespread role of copy number variation in mediating abiotic and biotic stress tolerance in plants.

Evaluation of the genetic architecture and the potential of genomics-assisted breeding of quality traits in two large panels of durum wheat.

KEY MESSAGE: New QTL for important quality traits in durum were identified, but for most QTL their effect varies depending on the investigated germplasm. Most of the global durum wheat (Triticum turgidum ssp. durum) production is used for human consumption via pasta and to a lower extent couscous and bulgur. Therefore, durum wheat varieties have to fulfill high demands regarding quality traits. In this study, we evaluated the quality traits protein content, sedimentation volume, falling number, vitreousity and thousand kernel weight in a Central European (CP) and a Southern and Western European panel (SP) with 183 and 159 durum lines, respectively, and investigated their genetic architecture by genome-wide association mapping. Except for protein content, we identified QTL explaining a large proportion of the genotypic variance for different traits. However, most of them were identified only in one panel. Nevertheless, for sedimentation volume a genomic region on chromosome 1B appeared important in both durum panels and a BLAST search against the emmer and bread wheat reference genomes points toward the candidate gene Glu-B3. This was further supported by the protein subunit banding pattern via SDS-PAGE gel electrophoresis. For vitreousity, genomic regions on chromosome 7A explained a larger proportion of the genotypic variance in both panels, whereas one QTL, possibly related to the Pinb-2 locus, also slightly influenced the protein content. Within each panel, high prediction abilities for genomic selection were obtained, which, however, dropped considerably when predicting across both panels. Nevertheless, the across-panel prediction ability was still larger than 0.4 for protein content and sedimentation volume, underlining the potential for genomics-aided durum breeding, if laboratory and logistical facilities are available.

The soybean experiment '1000 Gardens': a case study of citizen science for research, education, and beyond.

KEY MESSAGE: Citizen science, an approach that includes normal citizens in scientific research, holds great potential also for plant sciences and breeding and can be a powerful research tool to complement traditional approaches. Citizen science is an approach that includes normal citizens in scientific research, but has so far not been exploited by the various disciplines in plant sciences. Moreover, global threats challenge human well-being and science can provide solutions, but needs to leave the ivory tower in the mind of the broader public. In 2016, we performed the '1000 Gardens-the soybean experiment' citizen science project, that aimed at finding citizens in Germany who would grow soybean lines in their own gardens and evaluate them for a range of traits related to adaptation and agronomic performance. Here, we describe details of this project, i.e. the recruitment, performance, and compliance of the citizen scientists. A total of 2492 citizen scientists volunteered for the project, but through the high media coverage a much broader audience than just the participants was reached. Our 1000 Gardens project was successful in collecting a scientifically unique data set with heritabilities ranging up to 0.60 for maturity date or 0.69 for plant height. Our results suggest that the citizen science approach holds great potential also for plant sciences and can be a powerful research tool to complement traditional approaches. Our project was also successful in raising public awareness about the importance of plant breeding and in communicating key messages on the manifold benefits of legumes for a sustainable agriculture to a broader public. Thus, citizen science appears as a promising avenue to demonstrate the value of breeding and science to the general public by including normal citizens in scientific research.

Assessing the variation and genetic architecture of asparagine content in wheat: What can plant breeding contribute to a reduction in the acrylamide precursor?

KEY MESSAGE: A large genetic variation, moderately high heritability, and promising prediction ability for genomic selection show that wheat breeding can substantially reduce the acrylamide forming potential in bread wheat by a reduction in its precursor asparagine. Acrylamide is a potentially carcinogenic substance that is formed in baked products of wheat via the Maillard reaction from carbonyl sources and asparagine. In bread, the acrylamide content increases almost linearly with the asparagine content of the wheat grains. Our objective was, therefore, to investigate the potential of wheat breeding to contribute to a reduction in acrylamide by decreasing the asparagine content in wheat grains. To this end, we evaluated 149 wheat varieties from Central Europe at three locations for asparagine content, as well as for sulfur content, and five important quality traits regularly assessed in bread wheat breeding. The mean asparagine content ranged from 143.25 to 392.75 mg/kg for the different wheat varieties, thus underlining the possibility to reduce the acrylamide content of baked wheat products considerably by selecting appropriate varieties. Furthermore, a moderately high heritability of 0.65 and no negative correlations with quality traits like protein content, sedimentation volume and falling number show that breeding of quality wheat with low asparagine content is feasible. Genome-wide association mapping identified few QTL for asparagine content, the largest explaining 18% of the genotypic variance. Combining these QTL with a genome-wide prediction approach yielded a mean cross-validated prediction ability of 0.62. As we observed a high genotype-by-environment interaction for asparagine content, we recommend the costly and slow laboratory analysis only for late breeding generations, while selection in early generations could be based on marker-assisted or genomic selection.