Fine mapping of major QTL qshgd1 for spontaneous haploid genome doubling in maize (Zea mays L.).

KEY MESSAGE: A large-effect QTL was fine mapped, which revealed 79 gene models, with 10 promising candidate genes, along with a novel inversion. In commercial maize breeding, doubled haploid (DH) technology is arguably the most efficient resource for rapidly developing novel, completely homozygous lines. However, the DH strategy, using in vivo haploid induction, currently requires the use of mutagenic agents which can be not only hazardous, but laborious. This study focuses on an alternative approach to develop DH lines-spontaneous haploid genome duplication (SHGD) via naturally restored haploid male fertility (HMF). Inbred lines A427 and Wf9, the former with high HMF and the latter with low HMF, were selected to fine-map a large-effect QTL associated with SHGD-qshgd1. SHGD alleles were derived from A427, with novel haploid recombinant groups having varying levels of the A427 chromosomal region recovered. The chromosomal region of interest is composed of 45 megabases (Mb) of genetic information on chromosome 5. Significant differences between haploid recombinant groups for HMF were identified, signaling the possibility of mapping the QTL more closely. Due to suppression of recombination from the proximity of the centromere, and a newly discovered inversion region, the associated QTL was only confined to a 25 Mb region, within which only a single recombinant was observed among ca. 9,000 BC1 individuals. Nevertheless, 79 gene models were identified within this 25 Mb region. Additionally, 10 promising candidate genes, based on RNA-seq data, are described for future evaluation, while the narrowed down genome region is accessible for straightforward introgression into elite germplasm by BC methods.

Self-supervised maize kernel classification and segmentation for embryo identification.

Introduction: Computer vision and deep learning (DL) techniques have succeeded in a wide range of diverse fields. Recently, these techniques have been successfully deployed in plant science applications to address food security, productivity, and environmental sustainability problems for a growing global population. However, training these DL models often necessitates the large-scale manual annotation of data which frequently becomes a tedious and time-and-resource- intensive process. Recent advances in self-supervised learning (SSL) methods have proven instrumental in overcoming these obstacles, using purely unlabeled datasets to pre-train DL models. Methods: Here, we implement the popular self-supervised contrastive learning methods of NNCLR Nearest neighbor Contrastive Learning of visual Representations) and SimCLR (Simple framework for Contrastive Learning of visual Representations) for the classification of spatial orientation and segmentation of embryos of maize kernels. Maize kernels are imaged using a commercial high-throughput imaging system. This image data is often used in multiple downstream applications across both production and breeding applications, for instance, sorting for oil content based on segmenting and quantifying the scutellum's size and for classifying haploid and diploid kernels. Results and discussion: We show that in both classification and segmentation problems, SSL techniques outperform their purely supervised transfer learning-based counterparts and are significantly more annotation efficient. Additionally, we show that a single SSL pre-trained model can be efficiently finetuned for both classification and segmentation, indicating good transferability across multiple downstream applications. Segmentation models with SSL-pretrained backbones produce DICE similarity coefficients of 0.81, higher than the 0.78 and 0.73 of those with ImageNet-pretrained and randomly initialized backbones, respectively. We observe that finetuning classification and segmentation models on as little as 1% annotation produces competitive results. These results show SSL provides a meaningful step forward in data efficiency with agricultural deep learning and computer vision.

Development of doubled haploid inducer lines facilitates selection of superior haploid inducers in maize.

Haploid inducers are key components of doubled haploid (DH) technology in maize. Robust agronomic performance and better haploid induction ability of inducers are persistently sought through genetic improvement. We herein developed C1-I inducers enabling large-scale in vivo haploid induction of inducers and discovered superior inducers from the DH progenies. The haploid induction rate (HIR) of C1-I inducers ranged between 5.8% and 12.0%. Overall, the success rate of DH production was 13% on average across the 23 different inducer crosses. The anthesis-silking interval and days to flowering of inducer F1s are significantly correlated with the success rate of DH production (r = -0.48 and 0.47, respectively). Transgressive segregants in DH inducers (DHIs) were found for the traits (days to flowering, HIR, plant height, and total primary branch length). Moreover, the best HIR in DHIs exceeded 23%. Parental genome contributions to DHI progenies ranged between 0.40 and 0.55, respectively, in 25 and 75 percentage quantiles, and the mean and median were 0.48. The allele frequency of the four traits from inducer parents to DHI progenies did not correspond with the phenotypic difference between superior and inferior individuals in the DH populations by genome-wide Fst analysis. This study demonstrated that the recombinant DHIs can be accessed on a large scale and used as materials to facilitate the genetic improvement of maternal haploid inducers by in vivo DH technology.

Protocols for In Vivo Doubled Haploid (DH) Technology in Maize Breeding: From Haploid Inducer Development to Haploid Genome Doubling.

Doubled haploid (DH) technology reduces the time required to obtain homozygous genotypes and accelerates plant breeding among other advantages. It is established in major crop species such as wheat, barley, maize, and canola. DH lines can be produced by both in vitro and in vivo methods and the latter is focused here. The major steps involved in in vivo DH technology are haploid induction, haploid selection/identification, and haploid genome doubling. Herein, we elaborate on the various steps of DH technology in maize breeding from haploid induction to haploid genome doubling to produce DH lines. Detailed protocols on the following topics are discussed: in vivo haploid inducer line development, haploid selection using seed and root color markers and automated seed sorting based on embryo oil content using QSorter, artificial genome doubling, and the identification of genotypes with spontaneous haploid genome doubling (SHGD) ability.

Usefulness of temperate-adapted maize lines developed by doubled haploid and single-seed descent methods.

KEY MESSAGE: Spontaneous haploid genome doubling is not associated with undesirable linkage drag effects. The presence of spontaneous doubling genes allows maximum exploitation of variability from the temperate-adapted BS39 population Tropical non-elite maize (Zea mays L.) germplasm, such as BS39, provides a unique opportunity for broadening the genetic base of U.S. Corn Belt germplasm. In vivo doubled haploid (DH) technology has been used to efficiently exploit non-elite germplasm. It can help to purge deleterious recessive alleles. The objectives of this study were to determine the usefulness of BS39-derived inbred lines using both SSD and DH methods, to determine the impact of spontaneous as compared with artificial haploid genome doubling on genetic variance among BS39-derived DH lines, and to identify SNP markers associated with agronomic traits among BS39 inbreds monitored at testcross level. We developed two sets of inbred lines directly from BS39 by DH and SSD methods, named BS39_DH and BS39_SSD. Additionally, two sets were derived from a cross between BS39 and A427 (SHGD donor) by DH and SSD methods, named BS39 × A427_DH and BS39 × A427_SSD, respectively. Grain yield, moisture, plant height, ear height, stalk lodging, and root lodging were measured to estimate genetic parameters. For genome-wide association analysis, inbred lines were genotyped using genotype-by-sequencing and Diversity Array Technology Sequencing (DArTSeq). Some BS39-derived inbred lines performed better than elite germplasm inbreds and all sets showed significant genetic variance. The presence of spontaneous haploid genome doubling genes did not affect performance of inbred lines. Five SNPs were significant and three of them located within genes related to plant development or abiotic stresses. These results demonstrate the potential of BS39 to add novel alleles to temperate elite germplasm.

MYO, a Candidate Gene for Haploid Induction in Maize Causes Male Sterility.

Doubled haploid technology is highly successful in maize breeding programs and is contingent on the ability of maize inducers to efficiently produce haploids. Knowledge of the genes involved in haploid induction is important for not only developing better maize inducers, but also to create inducers in other crops. The main quantitative trait loci involved in maize haploid induction are qhir1 and qhir8. The gene underlying qhir1 has been discovered and validated by independent research groups. Prior to initiation of this study, the gene associated with qhir8 had yet to be recognized. Therefore, this research focused on characterizing positional candidate genes underlying qhir8. Pursuing this goal, a strong candidate for qhir8, GRMZM2G435294 (MYO), was silenced by RNAi. Analysis of crosses with these heterozygous RNAi-transgenic lines for haploid induction rate revealed that the silencing of MYO significantly enhanced haploid induction rate by an average of 0.6% in the presence of qhir1. Recently, GRMZM2G465053 (ZmDMP) was identified by map-based gene isolation and shown to be responsible for qhir8. While our results suggest that MYO may contribute to haploid induction rate, results were inconsistent and only showing minor increases in haploid induction rate compared to ZmDMP. Instead, reciprocal crosses clearly revealed that the silencing of MYO causes male sterility.

Breeding Maize Maternal Haploid Inducers.

Maize doubled haploid (DH) lines are usually created in vivo, through crosses with maternal haploid inducers. These inducers have the inherent ability of generating seeds with haploid embryos when used to pollinate other genotypes. The resulting haploid plants are treated with a doubling agent and self-pollinated, producing completely homozygous seeds. This rapid method of inbred line production reduces the length of breeding cycles and, consequently, increases genetic gain. Such advantages explain the wide adoption of this technique by large, well-established maize breeding programs. However, a slower rate of adoption was observed in medium to small-scale breeding programs. The high price and/or lack of environmental adaptation of inducers available for licensing, or the poor performance of those free of cost, might explain why smaller operations did not take full advantage of this technique. The lack of adapted inducers is especially felt in tropical countries, where inducer breeding efforts are more recent. Therefore, defining optimal breeding approaches for inducer development could benefit many breeding programs which are in the process of adopting the DH technique. In this manuscript, we review traits important to maize maternal haploid inducers, explain their genetic basis, listing known genes and quantitative trait loci (QTL), and discuss different breeding approaches for inducer development. The performance of haploid inducers has an important impact on the cost of DH line production.

Impact of Spontaneous Haploid Genome Doubling in Maize Breeding.

Doubled haploid (DH) technology has changed the maize-breeding landscape in recent years. Traditionally, DH production requires the use of chemical doubling agents to induce haploid genome doubling and, subsequently, male fertility. These chemicals can be harmful to humans and the plants themselves, and typically result in a doubling rate of 10%-30%. Spontaneous genome doubling and male fertility of maize haploids, without using chemical doubling agents, have been observed to a limited extent, for nearly 70 years. Rates of spontaneous haploid genome doubling (SHGD) have ranged from less than 5% to greater than 50%. Recently, there has been increased interest to forgo chemical treatment and instead utilize this natural method of doubling. Genetic-mapping studies comprising worldwide germplasm have been conducted. Of particular interest has been the detection of large-effect quantitative trait loci (QTL) affecting SHGD. Having a single large-effect QTL with an additive nature provides flexibility for the method of introgression, such as marker-assisted backcrossing, marker-assisted gene pyramiding, and systematic design. Moreover, it allows implementation of new methodologies, such as haploid-inducer mediated genome editing (HI-edit) and promotion of alleles by genome editing. We believe the use of SHGD can further enhance the impact of DH technology in maize.

Mapping of QTL and identification of candidate genes conferring spontaneous haploid genome doubling in maize (Zea mays L.).

In vivo doubled haploid (DH) technology is widely used in commercial maize (Zea mays L.) breeding. Haploid genome doubling is a critical step in DH breeding. In this study, inbred lines GF1 (0.65), GF3(0.29), and GF5 (0) with high, moderate, and poor spontaneous haploid genome doubling (SHGD), respectively, were selected to develop mapping populations for SHGD. Three QTL, qshgd1, qshgd2, and qshgd3, related to SHGD were identified by selective genotyping. With the exception of qshgd3, the source of haploid genome doubling alleles were derived from GF1. Furthermore, RNA-Seq was conducted to identify putative candidate genes between GF1 and GF5 within the qshgd1 region. A differentially expressed formin-like protein 5 transcript was identified within the qshgd1 region.

Accelerating plant breeding.

The growing demand for food with limited arable land available necessitates that the yield of major food crops continues to increase over time. Advances in marker technology, predictive statistics, and breeding methodology have allowed for continued increases in crop performance through genetic improvement. However, one major bottleneck is the generation time of plants, which is biologically limited and has not been improved since the introduction of doubled haploid technology. In this opinion article, we propose to implement in vitro nurseries, which could substantially shorten generation time through rapid cycles of meiosis and mitosis. This could prove a useful tool for speeding up future breeding programs with the aim of sustainable food production.

PollenCALC: software for estimation of pollen compatibility of self-incompatible allo- and autotetraploid species.

BACKGROUND: Self-incompatibility (SI) is a biological mechanism to avoid inbreeding in allogamous plants. In grasses, this mechanism is controlled by a two-locus system (S-Z). Calculation of male and female gamete frequencies is complex for tetraploid species. We are not aware of any software available for predicting pollen haplotype frequencies and pollen compatibility in tetraploid species. RESULTS: PollenCALC is a software tool written in C++ programming language that can predict pollen compatibility percentages for polyploid species with a two-locus (S, Z) self-incompatibility system. The program predicts pollen genotypes and frequencies based on defined meiotic parameters for allo- or autotetraploid species with a gametophytic S-Z SI system. These predictions can be used to obtain expected values for for diploid and for (allo- or autotetraploidy SI grasses. CONCLUSION: The information provided by this calculator can be used to predict compatibility of pair-crosses in plant breeding applications, to analyze segregation distortion for S and Z genes, as well as linked markers in mapping populations, hypothesis testing of the number of S and Z alleles in a pair cross, and the underlying genetic model.

Selection of haploid maize kernels from hybrid kernels for plant breeding using near-infrared spectroscopy and SIMCA analysis.

Samples of haploid and hybrid seed from three different maize donor genotypes after maternal haploid induction were used to test the capability of automated near-infrared transmission spectroscopy to individually differentiate haploid from hybrid seeds. Using a two-step chemometric analysis in which the seeds were first classified according to genotype and then the haploid or hybrid status was determined proved to be the most successful approach. This approach allowed 11 of 13 haploid and 25 of 25 hybrid kernels to be correctly identified from a mixture that included seeds of all the genotypes.

"PolyMin": software for identification of the minimum number of polymorphisms required for haplotype and genotype differentiation.

BACKGROUND: Analysis of allelic variation for relevant genes and monitoring chromosome segment transmission during selection are important approaches in plant breeding and ecology. To minimize the number of required molecular markers for this purpose is crucial due to cost and time constraints. To date, software for identification of the minimum number of required markers has been optimized for human genetics and is only partly matching the needs of plant scientists and breeders. In addition, different software packages with insufficient interoperability need to be combined to extract this information from available allele sequence data, resulting in an error-prone multi-step process of data handling. RESULTS: PolyMin, a computer program combining the detection of a minimum set of single nucleotide polymorphisms (SNPs) and/or insertions/deletions (INDELs) necessary for allele differentiation with the subsequent genotype differentiation in plant populations has been developed. Its efficiency in finding minimum sets of polymorphisms is comparable to other available program packages. CONCLUSION: A computer program detecting the minimum number of SNPs for haplotype discrimination and subsequent genotype differentiation has been developed, and its performance compared to other relevant software. The main advantages of PolyMin, especially for plant scientists, is the integration of procedures from sequence analysis to polymorphism selection within a single program, including both haplotype and genotype differentiation.

Frequency, type, and distribution of EST-SSRs from three genotypes of Lolium perenne, and their conservation across orthologous sequences of Festuca arundinacea, Brachypodium distachyon, and Oryza sativa.

BACKGROUND: Simple sequence repeat (SSR) markers are highly informative and widely used for genetic and breeding studies in several plant species. They are used for cultivar identification, variety protection, as anchor markers in genetic mapping, and in marker-assisted breeding. Currently, a limited number of SSR markers are publicly available for perennial ryegrass (Lolium perenne). We report on the exploitation of a comprehensive EST collection in L. perenne for SSR identification. The objectives of this study were 1) to analyse the frequency, type, and distribution of SSR motifs in ESTs derived from three genotypes of L. perenne, 2) to perform a comparative analysis of SSR motif polymorphisms between allelic sequences, 3) to conduct a comparative analysis of SSR motif polymorphisms between orthologous sequences of L. perenne, Festuca arundinacea, Brachypodium distachyon, and O. sativa, 4) to identify functionally associated EST-SSR markers for application in comparative genomics and breeding. RESULTS: From 25,744 ESTs, representing 8.53 megabases of nucleotide information from three genotypes of L. perenne, 1,458 ESTs (5.7%) contained one or more SSRs. Of these SSRs, 955 (3.7%) were non-redundant. Tri-nucleotide repeats were the most abundant type of repeats followed by di- and tetra-nucleotide repeats. The EST-SSRs from the three genotypes were analysed for allelic- and/or genotypic SSR motif polymorphisms. Most of the SSR motifs (97.7%) showed no polymorphisms, whereas 22 EST-SSRs showed allelic- and/or genotypic polymorphisms. All polymorphisms identified were changes in the number of repeat units. Comparative analysis of the L. perenne EST-SSRs with sequences of Festuca arundinacea, Brachypodium distachyon, and Oryza sativa identified 19 clusters of orthologous sequences between these four species. Analysis of the clusters showed that the SSR motif generally is conserved in the closely related species F. arundinacea, but often differs in length of the SSR motif. In contrast, SSR motifs are often lost in the more distant related species B. distachyon and O. sativa. CONCLUSION: The results indicate that the L. perenne EST-SSR markers are a valuable resource for genetic mapping, as well as evaluation of co-location between QTLs and functionally associated markers.