Genomic prediction of grain yield in contrasting environments for white lupin genetic resources Abstract Germplasm collections hold several thousands of white lupin (Lupinus albus L.) accessions. Genome-enabled models with good predictive ability for specific environments could provide a cost-efficient means to identify promising genetic resources for breeding programmes. This study provided an unprecedented assessment of genome-enabled predictions for white lupin grain yield, focusing on (i) a world collection of 109 landraces and 8 varieties phenotyped in three European sites with contrasting climate (Mediterranean, subcontinental or oceanic) and sowing time (data set 1); (ii) 78 geographically diversified landrace genotypes and three variety genotypes phenotyped in moisture-favourable and severely drought-prone managed environments (data set 2). The interest of predictions for individual genotypes was justified by large within-landrace variation for yield responses. Ridge regression BLUP (rrBLUP) and Bayesian Lasso (BL) models exploited allele frequencies (estimated from 3 to 4 genotypes per landrace) of 10,782 polymorphic SNPs for data set 1, and allele values of 9937 polymorphic SNPs for data set 2, following ApeKI-based genotyping-by-sequencing characterization. Compared with BL, rrBLUP displayed similar predictive ability for data set 1 and better predictive ability for data set 2. Best-predictive models displayed intra-environment predictive ability for the five test environments in the range 0.47–0.76. Cross-environment predictions between pairs of environments with positive genetic correlation, i.e., autumn-sown subcontinental vs Mediterranean sites, and moisture-favourable vs drought-prone environments, exhibited a predictive ability range of 0.40–0.51 and a predictive accuracy range of 0.48–0.61. Our results support the exploitation of genomic predictions and provide economic justification for the genotyping of germplasm collections of white lupin.

Functional genomics of the protein kinase superfamily from wheat Abstract Protein kinases (PKs) belonging to a large superfamily play vital roles in plant development and stress tolerance, whereas only a small number of PKs have been functionally studied in wheat. Genome-wide survey and characterization of wheat PK genes (TaPKs) will be valuable in future genetic improvement program. In this study, we performed genome-wide identification of PK in wheat, containing phylogenetic evolution, gene expression pattern, and functional prediction analysis, and to explore structure–function relationship, the secondary and tertiary structures of wheat OST1 were visualized. Totally, 4479 TaPKs composed of 4446 Ser/Thr/Tyr (TaPK1 to TaPK4446) and 33 His PKs (TaHPK1 to TaHPK33) were identified in the TGACv1 assembly. According to phylogenetic analysis, wheat Ser/Thr/Tyr PKs were categorized into eight groups with 57 families and 64 subfamilies, and His PKs were classified into three families. The genes contained on average 5.07 introns, ranging from 0 to 31, and about 54.68% TaPKs exhibited relatively simpler exon–intron organization (0–4 introns). RNA-seq analysis showed 324 TaPKs were significantly differentially expressed after exposure to drought, 372 TaPKs were significantly differentially expressed after inoculation with Puccinia striiformis f. sp. tritici (Pst) and Blumeria graminis f. sp. tritici (Bgt), and a large number of TaPKs exhibited stage- or organ-specific expression patterns. Co-expression analysis showed that 1388 TaPKs were distributed in 19 modules, while 9 modules with 511 TaPKs had the most obvious tissue- or stress-specific expression trends. A suite of sexual reproduction related genes were identified in magenta module where TaPK960, TaPK707–3708, TaPK4243, and TaPK4249 were related to polarized pollen tube growth, TaPK627 and TaPK629 to pollen tube guidance, and TaPK4226–4228 to flowering time regulation. Finally, 12 TaPKs were selected to validate their expression level through qRT-PCR. Our findings obtained will offer a clue to elaborate the roles of PKs in wheat growth and stress resistance for agricultural improvement.

Impact of fruit shape selection on genetic structure and diversity uncovered from genome-wide perfect SNPs genotyping in eggplant Abstract The recently constructed eggplant genome provides an opportunity for genome-wide marker exploration to be used for variety identification and genetic analysis. SNPs with high density across the genome and genetic stability are suitable for DNA fingerprint and genetic analysis. In this study, we used the eggplant reference genome of line 63/7 and resequencing data of 45 unique lines to select 219 genome-wide perfect SNPs in the eggplant. The high-throughput SNP genotyping technique target SNP-seq was used to successfully validate the 219 SNPs across 377 eggplant varieties to establish the unique DNA fingerprint for each variety. In addition, we chose 36 SNPs as a core SNP set with the ability to differentiate 95% of 377 eggplant varieties. Model-based structure and principal component analysis suggested three major population groups and indicated a significant impact of long-term selection for fruit shape in eggplant varieties. Further genetic diversity analysis within the three population groups showed a considerably narrow genetic background in round- and oval-fruited eggplants. This could be a result of limited choices of varieties used for breeding of round- and oval-fruited eggplants in Asia, indicated by its low observed heterozygosity (Ho) value (0.152) and high inbreeding coefficient (0.442). Finally, a genome-wide association study based on the 219 SNPs identified five associated SNPs located near the SUN and OVATE homologs, which had conserved function in controlling the fruit shape. This study signals a risk of genetic erosion in the round- and oval-fruited eggplants and provides valuable information for future variety management and breeding programs.

Mining QTLs and candidate genes for seed protein and oil contents across multiple environments and backgrounds in soybean Abstract Soybean is important due to its seed protein and oil, but its seed protein and oil contents need to be improved. In view of this, two genetically related F6:7 recombinant inbred line (RIL) populations, derived from Zheng92116×Qihuang30 (ZQ) and Zheng92116× Liaodou14 (ZL), were genotyped with a SoySNP 6K chip and evaluated for seed protein and oil contents in four different environments. In total, ten stable collocating or overlapping additive QTL clusters, which explained 2.6~23.3% of the phenotypic variation, were identified in the ZQ population in different environments, four of which were further verified in the ZL population. One major and stable QTL controlling seed oil content, qOIL-A1, was detected in all four environments in ZQ and was validated in three environments in the ZL population, with the PVE (phenotypic variation explained) ranging from 4.8 to 22.1% and four of the PVE values exceeding 10% (22.1%, 19.3%, 10.9%, and 12.9%). Another major and stable QTL controlling seed protein content, qPRO-J, was detected in different environments in ZQ and was further verified in the ZL population, with a PVE of 14.1% and 23.2% in ZQ and a PVE of 9.5% in ZL. Moreover, the individuals in the RIL population and another cultivar-based population (320 varieties and landraces) with different genotypes at the common flanking markers for the QTL clusters showed extremely significantly different protein and oil contents, which further validated the QTL mapping results. Based on these findings, eight novel candidate genes with very different expression levels at different soybean seed growth and developmental stages between ZQ and ZL RIL parents were identified in the qOIL-A1 region through transcriptome sequencing and were finally verified by real-time qPCR. Thus, these stable QTL clusters can be applied in marker-assisted selection breeding or map-based candidate gene cloning in soybean for seed protein and oil genetic improvements in the future.

An effective strategy for fertility improvement of indica - japonica hybrid rice by pyramiding S5-n , f5-n , and pf12-j Abstract Though inter-subspecific hybrids from indica and japonica rice have gained great success for yield increases over the recent decade, there are severe sterility problems in such hybrids. To overcome hybrid sterility, we previously obtained the line PL-(S5-n + f5-n) in the genetic background of an elite indica restorer line 9311, via breeding strategy for construction of wide compatibility line, which was proven to be an effective strategy in improving the seed setting rate of indica-japonica hybrid rice. Here, a novel strategy for developing japonica-compatible indica lines was employed by introducing the japonica allele pf12-j into 9311. The obtained NIL-(pf12-j) could significantly improve pollen and embryo-sac fertility by 29.7% and 28.6% in indica-japonica hybrids, thus leading to 20.9% improvement of spikelet fertility. In addition, we achieved the pyramiding line PL-(S5-n + f5-n + pf12-j) by a combinational strategy using both wide compatibility line and japonica-compatible indica line. Compared to PL-(S5-n + f5-n) with 38.9% increase of spikelet fertility, the pyramiding line PL-(S5-n + f5-n + pf12-j) showed 49.7% increase of spikelet fertility, suggesting cumulative effect of wide compatibility alleles S5-n + f5-n and japonica allele pf12-j to shape normal fertility of inter-subspecies hybrid. Interestingly, these lines also showed compatibility to indica. Hence, our results demonstrate that the two strategies could be simultaneously applied for indica-japonica hybrid breeding, and S5-n + f5-n + pf12-j are the optional allelic combination for overcoming hybrid sterility. This finding will greatly enhance our understanding for breeding indica-japonica hybrid rice by molecular-assisted selection strategy.

Genetic dissection of winter barley seedling response to salt and osmotic stress Abstract Seedling establishment is a vulnerable stage in the crop life cycle which can be affected by different abiotic stresses. Drought and salinity are major environmental constraints worldwide, but few studies have genetically compared these two stresses on the same genetic material. In this study, the dynamic response of barley seedling growth to different levels of salt stress and PEG6000-mediated osmotic stress was evaluated in a European winter barley cultivar collection, in parallel with non-stress conditions. Salt and osmotic stress experiments produced different root-response curves. A final set of 56 phenotypic traits was subjected to genome-wide association mapping with 4885 gene-based SNP markers: 28 quantitative trait loci (QTL) were identified; 10 loci were found to be involved in saline conditions, whereas 20 loci were detected under osmotic stress. Four loci on chromosomes 1H, 5H and 6H were detected under more than one growth condition. One co-localized QTL was involved in both root and shoot growth only under salt stress. A set of potential candidate genes with putative pleiotropic effects on seedling growth under different conditions is proposed, based on their physical proximity to the QTL peak markers. To conclude, we found that a QTL controlling seedling growth under one abiotic stress can respond to another stress. QTL promoting faster growth under non-stress conditions were also identified under salt and/or osmotic stresses. These loci might be exploited in breeding programs to overcome environmental stresses at the initial seedling stage.

Genetic dissection of floral traits in anemone-type chrysanthemum by QTL mapping Abstract The anemone-type inflorescence of chrysanthemum features prominent, colored tubular florets. Here, the inheritance of a set of traits associated with floral form was investigated using a quantitative trait locus (QTL) mapping approach applied to a segregating population bred from a cross between the two varieties ‘Nannong Xuefeng’ and ‘Monalisa’. The coefficient of variation of the various traits ranged from 12.8 to 59.2%, with an average of 28.0%, and there was notable transgressive segregation in both directions. A set of 26 QTL was uncovered, explaining a portion of the variation in eight of the eleven traits; individually, the loci accounted for between 6.3 and 11.9% of the phenotypic variation. A further 16 pairs of epistatically acting QTL were identified, associated with a contribution of between 3.7 and 13.6% of the phenotypic variance; the majority of these interactions involved background loci. The conclusion was that both additive and epistatic effects were involved in the genetic determination of floral form in chrysanthemum.

LONG GRAIN 1 : a novel gene that regulates grain length in rice Abstract Grain size is one of the most important traits that influence grain yield in rice. Here we report the isolation and characterization of the long grain1 (lin1) mutant, which exhibited long grains and increased grain weight. A field trial conducted over 2 years indicated that the grain weight increased 3.0–6.6% in the lin1 mutant compared with that of the wild-type Nipponbare. Whole-genome sequencing and genetic linkage analysis indicated that a 1-bp deletion within the coding sequence of Os06g0675200, which showed no homology to previously characterized genes, is responsible for the lin1 phenotype. Quantitative real-time PCR indicated that LIN1 transcripts were more abundant in young panicles than at advanced inflorescence developmental stages, which indicated that LIN1 controls grain length mainly at early stages of grain development. Sequence polymorphism analysis of LIN1 showed that all 15 temperate japonica cultivars tested as well as six out of the nine indica cultivars tested possessed the Nipponbare-type LIN1 allele, whereas the remaining three indica cultivars and one aus cultivar tested harbored an identical missense mutation in LIN1. These results revealed that the mutant allele of LIN1 has not been widely utilized in breeding temperate japonica cultivars currently in cultivation. Our findings indicate that the lin1 mutation may be useful to further improve grain length and presumably grain yield in temperate japonica and indica rice cultivars that harbor the Nipponbare-type LIN1 allele.

Frequent variations in tandem repeats pSc200 and pSc119.2 cause rapid chromosome evolution of open-pollinated rye Abstract Variation in heterochromatin is thought to play an important role in chromosome evolution in cereal rye (Secale cereale). This theory has been based on only limited numbers of sampled individuals, as it is difficult to handle large numbers in detailed cytogenetic analyses. In this study, we analyzed a relatively large number of individuals comprising 300 plants from six open-pollinated rye cultivars using oligonucleotides TAMRA-oligo-6 and FAM-pSc119.2-1 that represent tandem repeats of pSc200 and pSc119.2, respectively. Seventy-three types of heterochromatin blocks were identified on all seven chromosomes: 43 from pSc119.2 and 30 from pSc200. Eight block types were fully conserved, and variants in 65 (89.04%) had frequencies ranging from 0.17 to 49.67% and an average of 14.47%. Block 6R-g3 (49.67%) had the highest frequency of variation. Higher average frequencies of variation in block types were revealed by pSc200 (19.33%) than by pSc119.2 (11.08%). A total of 322 polymorphic chromosomes (combinations of different block types) were identified, including 79 major types with frequencies higher than expected, including 14 that were fivefold higher. Six preferentially transmitted chromosomes were confirmed by karyotype analysis in 40 wheat-rye F1 plants. Among the 300 plants investigated, there were 63.00 to 86.66% heterozygosity and 43.66 to 90.32% heterogeneity for each chromosome. Results from this research confirmed rapid chromosome change and high levels of chromosome diversity in heterochromatin blocks in rye.

Multi-environment QTL analysis using an updated genetic map of a widely distributed Seri × Babax spring wheat population Abstract Seri/Babax spring wheat RIL population was developed to minimize the confounding effect of phenology in the genetic dissection of abiotic stress traits. An existing linkage map (< 500 markers) was updated with 6470 polymorphic Illumina iSelect 90K array and DArTseq SNPs to a genetic map of 5576.5 cM with 1748 non-redundant markers (1165 90K SNPs, 207 DArTseq SNPs, 183 AFLP, 111 DArT array, and 82 SSR) assigned to 31 linkage groups. We conducted QTL mapping for yield and related traits phenotyped in several major wheat growing areas in Egypt, Sudan, Iran, India, and Mexico (nine environments: heat, drought, heat plus drought, and yield potential). QTL analysis identified 39 (LOD 2.5–23.6; PVE 4.8–21.3%), 36 (LOD 2.5–15.4; PVE 2.9–21.4%), 30 (LOD 2.5–13.1; PVE 3.6–26.8%), 39 (LOD 2.7–14.4; PVE 2.6–15.9%), and 22 (LOD 2.8–4.8; PVE 6.8–12.9%) QTLs for grain yield, thousand-grain weight, grain number, days to heading, and plant height, respectively. The present study confirmed QTLs from previous studies and identified novel QTLs. QTL analysis based on high-yielding and low-yielding environmental clusters identified 11 QTLs (LOD 2.6–14.9; PVE 2.7–19.7%). The updated map thereby provides a better genome coverage (3.5-fold) especially on the D genome (4-fold), higher density (1.1-fold), and a good collinearity with the IWGSC RefSeq v1.0 genome, and increased the number of detected QTLs (5-fold) compared with the earlier map. This map serves as a useful genomic resource for genetic analyses of important traits on this wheat population that was widely distributed around the world.