Durable resistance to blast, the most significant fungal disease of rice, represents an agronomically relevant character. Gigante Vercelli (GV) and Vialone Nano (VN) are two old temperate japonica Italian rice cultivars with contrasting response to blast infection: GV displays durable and broad resistance while VN is highly susceptible. RNA-seq was used to dissect the early molecular processes deployed during the resistance response of GV at 24 h after blast inoculation. Differential gene expression analysis identified 1,070 and 1,484 modulated genes, of which 726 and 699 were up regulated in response to infection in GV and VN, respectively. Gene ontology (GO) enrichment analyses revealed a set of GO terms enriched in both varieties but, despite this commonality, the gene sets contributing to common GO enriched terms were dissimilar. The expression patterns of genes grouped in GV-specific enriched GO terms were examined in detail including at the transcript isoform level. GV exhibited a dramatic up-regulation of genes encoding diterpene phytoalexin biosynthetic enzymes, flavin-containing monooxygenase, class I chitinase and glycosyl hydrolase 17. The sensitivity and high dynamic range of RNA-seq allowed the identification of genes critically involved in conferring GV resistance during the early steps of defence perception-signalling. These included chitin oligosaccharides sensing factors, wall associated kinases, MAPK cascades and WRKY transcription factors. Candidate genes with expression patterns consistent with a potential role as GV-specific functional resistance (R) gene(s) were also identified. This first application of RNA-seq to dissect durable blast resistance supports a crucial role of the prompt induction of a battery of responses including defence-related genes as well as members of gene families involved in signalling and pathogen-related gene expression regulation.

BackgroundApparent Amylose Content (AAC), regulated by the Waxy gene, represents the key determinant of rice cooking properties. In occidental countries high AAC rice represents the most requested market class but the availability of molecular markers allowing specific selection of high AAC varieties is limited.ResultsIn this study, the effectiveness of available molecular markers in predicting AAC was evaluated in a collection of 127 rice accessions (125 japonica ssp. and 2 indica ssp.) characterized by AAC values from glutinous to 26%. The analyses highlighted the presence of several different allelic patterns identifiable by a few molecular markers, and two of them, i.e., the SNPs at intron1 and exon 6, were able to explain a maximum of 79.5% of AAC variation. However, the available molecular markers haplotypes did not provide tools for predicting accessions with AAC higher than 24.5%. To identify additional polymorphisms, the re-sequencing of the Waxy gene and 1kbp of the putative upstream regulatory region was performed in 21 genotypes representing all the AAC classes identified. Several previously un-characterized SNPs were identified and four of them were used to develop dCAPS markers.ConclusionsThe addition of the SNPs newly identified slightly increased the AAC explained variation and allowed the identification of a haplotype almost unequivocally associated to AAC higher than 24.5%. Haplotypes at the waxy locus were also associated to grain length and length/width (L/W) ratio. In particular, the SNP at the first intron, which identifies the Wx a and Wx b alleles, was associated with differences in the width of the grain, the L/W ratio and the length of the kernel, most likely as a result of human selection.

BackgroundFusarium fujikuroi is the causal agent of bakanae, the most significant seed-borne disease of rice. Molecular mechanisms regulating defence responses of rice towards this fungus are not yet fully known. To identify transcriptional mechanisms underpinning rice resistance, a RNA-seq comparative transcriptome profiling was conducted on infected seedlings of selected rice genotypes at one and three weeks post germination (wpg).ResultsTwelve rice genotypes were screened against bakanae disease leading to the identification of Selenio and Dorella as the most resistant and susceptible cultivars, respectively. Transcriptional changes were more appreciable at 3 wpg, suggesting that this infection stage is essential to study the resistance mechanisms: 3,119 DEGs were found in Selenio and 5,095 in Dorella. PR1, germin-like proteins, glycoside hydrolases, MAP kinases, and WRKY transcriptional factors were up-regulated in the resistant genotype upon infection with F. fujikuroi. Up-regulation of chitinases and down-regulation of MAP kinases and WRKY transcriptional factors were observed in the susceptible genotype. Gene ontology (GO) enrichment analyses detected in Selenio GO terms specific to response to F. fujikuroi: ‘response to chitin’, ‘jasmonic acid biosynthetic process’, and ‘plant-type hypersensitive response’, while Dorella activated different mechanisms, such as ‘response to salicylic acid stimulus’ and ‘gibberellin metabolic process’, which was in agreement with the production of gibberellin A3 in Dorella plants.ConclusionsRNA-seq profiling was performed for the first time to analyse response of rice to F. fujikuroi infection. Our findings allowed the identification of genes activated in one- and three- week-old rice seedlings of two genotypes infected with F. fujikuroi. Furthermore, we found the pathways involved in bakanae resistance, such as response to chitin, JA-dependent signalling and hypersensitive response. Collectively, this provides important information to elucidate the molecular and cellular processes occurring in rice during F. fujikuroi infection and to develop bakanae resistant rice germplasm.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-016-2925-6) contains supplementary material, which is available to authorized users.

Oryza sativa, a model plant for Arbuscular Mycorrhizal (AM) symbiosis, has both host and non-host roots. Large lateral (LLR) and fine lateral (FLR) roots display opposite responses: LLR support AM colonization, but FLR do not. Our research aimed to study the molecular, morphological and physiological aspects related to the non-host behavior of FLR. RNA-seq analysis revealed that LLR and FLR displayed divergent expression profiles, including changes in many metabolic pathways. Compared with LLR, FLR showed down-regulation of genes instrumental for AM establishment and gibberellin signaling, and a higher expression of nutrient transporters. Consistent with the transcriptomic data, FLR had higher phosphorus content. Light and electron microscopy demonstrated that, surprisingly, in the Selenio cultivar, FLR have a two-layered cortex, which is theoretically compatible with AM colonization. According to RNA-seq, a gibberellin inhibitor treatment increased anticlinal divisions leading to a higher number of cortex cells in FLR. We propose that some of the differentially regulated genes that lead to the anatomical and physiological properties of the two root types also function as genetic factors regulating fungal colonization. The rice root apparatus offers a unique tool to study AM symbiosis, allowing direct comparisons of host and non-host roots in the same individual plant.

BackgroundIn this study we carried out a genome-wide association analysis for plant and grain morphology and root architecture in a unique panel of temperate rice accessions adapted to European pedo-climatic conditions. This is the first study to assess the association of selected phenotypic traits to specific genomic regions in the narrow genetic pool of temperate japonica. A set of 391 rice accessions were GBS-genotyped yielding—after data editing—57000 polymorphic and informative SNPS, among which 54% were in genic regions.ResultsIn total, 42 significant genotype-phenotype associations were detected: 21 for plant morphology traits, 11 for grain quality traits, 10 for root architecture traits. The FDR of detected associations ranged from 3 · 10−7 to 0.92 (median: 0.25). In most cases, the significant detected associations co-localised with QTLs and candidate genes controlling the phenotypic variation of single or multiple traits. The most significant associations were those for flag leaf width on chromosome 4 (FDR = 3 · 10−7) and for plant height on chromosome 6 (FDR = 0.011).ConclusionsWe demonstrate the effectiveness and resolution of the developed platform for high-throughput phenotyping, genotyping and GWAS in detecting major QTLs for relevant traits in rice. We identified strong associations that may be used for selection in temperate irrigated rice breeding: e.g. associations for flag leaf width, plant height, root volume and length, grain length, grain width and their ratio. Our findings pave the way to successfully exploit the narrow genetic pool of European temperate rice and to pinpoint the most relevant genetic components contributing to the adaptability and high yield of this germplasm. The generated data could be of direct use in genomic-assisted breeding strategies.

Fusarium head blight (FHB), caused by the fungus Fusarium graminearum, represents one of the major wheat diseases worldwide, determining severe yield losses and reduction of grain quality due to the accumulation of mycotoxins. The molecular response associated with the wheat 2DL FHB resistance QTL was mined through a comprehensive transcriptomic analysis of the early response to F. graminearum infection, at 3 days post-inoculation, in spikelets and rachis. The analyses were conducted on two near isogenic lines (NILs) differing for the presence of the 2DL QTL (2-2618, resistant 2DL+ and 2-2890, susceptible null). The general response to fungal infection in terms of mRNAs accumulation trend was similar in both NILs, even though involving an higher number of DEGs in the susceptible NIL, and included down-regulation of the primary and energy metabolism, up-regulation of enzymes implicated in lignin and phenylpropanoid biosynthesis, activation of hormons biosynthesis and signal transduction pathways and genes involved in redox homeostasis and transcriptional regulation. The search for candidate genes with expression profiles associated with the 2DL QTL for FHB resistance led to the discovery of processes differentially modulated in the R and S NILs related to cell wall metabolism, sugar and JA signaling, signal reception and transduction, regulation of the redox status and transcription factors. Wheat FHB response-related miRNAs differentially regulated were also identified as putatively implicated in the superoxide dismutase activities and affecting genes regulating responses to biotic/abiotic stresses and auxin signaling. Altered gene expression was also observed for fungal non-codingRNAs. The putative targets of two of these were represented by the wheat gene WIR1A, involved in resistance response, and a gene encoding a jacalin-related lectin protein, which participate in biotic and abiotic stress response, supporting the presence of a cross-talk between the plant and the fungus.

A rice GWAS panel of 281 accessions of japonica rice was phenotypically characterized for 26 traits related to phenology, plant and seed morphology, physiology and yield for 2 years in field conditions under permanent flooding (PF) and limited water (LW). A genome-wide analysis uncovered a total of 160 significant marker-trait associations (MTAs), of which 32 were LW-specific, 59 were PF-specific, and 69 were in common between the two water management systems. LW-specific associations were identified for several agronomic traits including days to maturation, days from flowering to maturation, leaf traits, plant height, panicle and seed traits, hundred grain weight, yield and tillering. Significant MTAs were detected across all the 12 rice chromosomes, while clusters of effects influencing different traits under LW or in both watering conditions were, respectively, observed on chromosomes 4, 8, and 12 and on chromosomes 1, 3, 4, 5, and 8. The analysis of genes annotated in the Nipponbare reference sequence and included in the regions associated to traits related to plant morphology, grain yield, and physiological parameters allowed the identification of genes that were demonstrated to affect the respective traits. Among these, three (OsOFP2, Dlf1, OsMADS56) and seven (SUI1, Sd1, OsCOL4, Nal1, OsphyB, GW5, Ehd1) candidate genes were, respectively, identified to co-localize with LW-specific associations and associations in common between the two water treatments. For several LW-specific MTAs, or in common among the two treatments, positional co-localizations with previously identified QTLs for rice adaptation to water shortages were observed, a result that further supports the role of the loci identified in this work in conferring adaptation to LW. The most robust associations identified here could represent suitable targets for genomic selection approaches to improve yield-related traits under LW.

Key messageRice breeding programs based on pedigree schemes can use a genomic model trained with data from their working collection to predict performances of progenies produced through rapid generation advancement.AbstractSo far, most potential applications of genomic prediction in plant improvement have been explored using cross validation approaches. This is the first empirical study to evaluate the accuracy of genomic prediction of the performances of progenies in a typical rice breeding program. Using a cross validation approach, we first analyzed the effects of marker selection and statistical methods on the accuracy of prediction of three different heritability traits in a reference population (RP) of 284 inbred accessions. Next, we investigated the size and the degree of relatedness with the progeny population (PP) of sub-sets of the RP that maximize the accuracy of prediction of phenotype across generations, i.e., for 97 F5–F7 lines derived from biparental crosses between 31 accessions of the RP. The extent of linkage disequilibrium was high (r 2 = 0.2 at 0.80 Mb in RP and at 1.1 Mb in PP). Consequently, average marker density above one per 22 kb did not improve the accuracy of predictions in the RP. The accuracy of progeny prediction varied greatly depending on the composition of the training set, the trait, LD and minor allele frequency. The highest accuracy achieved for each trait exceeded 0.50 and was only slightly below the accuracy achieved by cross validation in the RP. Our results thus show that relatively high accuracy (0.41–0.54) can be achieved using only a rather small share of the RP, most related to the PP, as the training set. The practical implications of these results for rice breeding programs are discussed.Electronic supplementary materialThe online version of this article (10.1007/s00122-017-3011-4) contains supplementary material, which is available to authorized users.