Theoretical and Applied Genetics

, Volume 131, Issue 3, pp 613–624 | Cite as

A multiple near isogenic line (multi-NIL) RNA-seq approach to identify candidate genes underpinning QTL

  • Ahsan Habib
  • Jonathan J. Powell
  • Jiri Stiller
  • Miao Liu
  • Sergey Shabala
  • Meixue Zhou
  • Donald M. Gardiner
  • Chunji Liu
Original Article


Key message

This study demonstrates how identification of genes underpinning disease-resistance QTL based on differential expression and SNPs can be improved by performing transcriptomic analysis on multiple near isogenic lines.


Transcriptomic analysis has been widely used to understand the genetic basis of a trait of interest by comparing genotypes with contrasting phenotypes. However, these approaches identify such large sets of differentially expressed genes that it proves difficult to isolate which genes underpin the phenotype of interest. This study tests whether using multiple near isogenic lines (NILs) can improve the resolution of RNA-seq-based approaches to identify genes underpinning disease-resistance QTL. A set of NILs for a major effect Fusarium crown rot-resistance QTL in barley on the 4HL chromosome arm were analysed under Fusarium crown rot using RNA-seq. Differential gene expression and single nucleotide polymorphism detection analyses reduced the number of putative candidates from thousands within individual NIL pairs to only one hundred and two genes, which were differentially expressed or contained SNPs in common across NIL pairs and occurred on 4HL. Our findings support the value of performing RNA-seq analysis using multiple NILs to remove genetic background effects. The enrichment analyses indicated conserved differences in the response to infection between resistant and sensitive isolines suggesting that sensitive isolines are impaired in systemic defence response to Fusarium pseudograminearum.


Fusarium pathogens cause economically significant diseases in cereal crops such as Fusarium crown rot (FCR) and Fusarium head blight (FHB). FCR is an insidious and chronic constraint to barley and wheat production worldwide. This disease is predominantly found in many parts of the semi-arid cereals producing regions including Australia (Chakraborty et al. 2006). Several measures have been assessed to minimise damages inflicted by this disease but none of these practices seem to be very effective, as the incidence of FCR has increased recently in Australia and other cereal growing countries. Yield loss due to FCR as high as 13% has been reported in the Pacific Northwest of USA (Smiley et al. 2005) and the most recent survey in Australia found an estimated annual yield loss of $97 million Australian dollars in wheat and barley combined (Murray and Brennan 2009, 2010).

In the pursuit of generating resistant barley varieties, three large-effect quantitative trait loci (QTL) conferring FCR resistance have been reported on the long arms of chromosomes 1H, 3H and 4H, respectively (Chen et al. 2013a, b). The ultimate aim of mapping approaches is to identify and clone the gene or genes underpinning the QTL and to elucidate the mechanism by which these loci contribute improved resistance. In addition, identification and cloning of genes underpinning resistance sources in species with large and complex genomes such as barley remains challenging. In our previous attempt towards the mapping of FCR resistance gene(s), we have developed ten pairs of NILs for a major QTL located on chromosome 4HL conferring FCR resistance (Habib et al. 2016). However, the genetic element underpinning these resistance sources has yet to be identified.

To provide any inference toward the resistance mechanism, mapping approaches must be followed up with complemented functional characterisation. Transcriptomic approaches using RNA-seq have been employed and reported for cereal crops to understand how resistance to various stresses or diseases is mediated. Typically, such studies will compare between two or more genetically distinct lines which also differ significantly in their disease sensitivity (Ding et al. 2011; Kugler et al. 2013; Steiner et al. 2009; Xiao et al. 2013). Isolating which responses contribute tractable resistance or susceptibility proves difficult and often at best narrows the strategy for further functional characterisation. Uniformity in genetic backgrounds can be achieved by developing near isogenic lines (NILs) (Habib et al. 2016; Ma et al. 2012). Compared to traditional genetic populations, NILs offer several advantages for transcriptional analyses due to the minimization of genetic background interference and enhancement of the sensitivity and accuracy of transcriptional analyses (Keurentjes et al. 2007). Several recent studies have compared the host response in wheat NIL pairs carrying resistant or susceptible alleles for major QTL conferring FCR (Ma et al. 2014) or FHB resistance (reviewed in Kazan and Gardiner 2017). These studies have identified numerous genes that are differentially expressed between isolines but a comprehensive understanding of QTL-driven host resistance has remained elusive.

RNA-seq is a powerful approach for detecting differentially expressed genes (DEGs) and novel-expressed genes over a broad dynamic range (Blencowe et al. 2009; Wang et al. 2009). This approach can also be used for detecting single nucleotide polymorphisms (SNPs) in transcribed genes that co-locate with a target locus (Cavanagh et al. 2013). RNA-seq also enables calling expression of genes which are not present within the current annotation (Trapnell et al. 2010); a useful capability when working with relatively incomplete annotations. Transcriptomic profiling enables correlation of expression patterns with phenotypes of interest to infer ‘expression QTL’; an approach adopted widely to study genetic variation underpinning traits in plant species (West et al. 2007; Swanson-Wagner et al. 2009). Such ‘genetical genomics’ approaches have been applied recently to identify eQTL in resistance donor (Sumai-3) derived lines (Samad-Zamini et al. 2017).

Within NILs, SNPs will occur between isolines within regions inherited from different parental donors, i.e., regions still segregating between isolines at the point of selection. SNPs occurring outside the QTL region will obscure accurate identification of candidate genes underpinning the QTL unless these can be filtered out based on available mapping information. Individual NIL pairs will differ in their background genetic variation outside the QTL region as well as the size of the recombinant region harbouring the QTL. Our approach utilises multiple NIL pairs to improve the identification of candidate genes which are differentially expressed and have conserved SNP differences across multiple NIL pairs.

In this work, RNA-seq was employed to analyse the global transcriptional responses to Fusarium infection among three pair of NILs that possess significantly different sensitivity to FCR. The aims of the study are to (1) explore the mechanism through which the 4H locus from the resistant donor contributes resistance by comparing the response to infection between resistant and susceptible isolines and to (2) identify candidate genes occurring within the genomic region harbouring the resistance loci based on their differential expression patterns and SNP complement between ‘R’ and ‘S’ isolines.

The outcomes from this study provide the first insight into the molecular aspect of FCR resistance in barley and will increase our understanding of how resistance to necrotrophic fungal pathogens is mediated in cereal crops. In addition, the value of employing multiple NILs in transcriptomic approaches to improve meaningful SNP and DEG calling is discussed.

Materials and methods

Plant materials

Habib et al. (2016) reported 10 pairs of NILs following the heterogeneous inbred family method for the FCR QTL on chromosome arm 4HL of barley. Each NIL pair was derived from a single plant heterozygous for the QTL based on screening with SSR markers WMS6 and HVM67. Three of these NILs were selected for use in this study, namely NIL_CR4HL_2, NIL_CR4HL_3 and NIL_4HL_6, which were derived from the population of Lockyer//AWCS276/AWCS079. Within each pair, the resistance QTL was fixed in the F8 generation so that ‘R’ isolines were homozygous for the resistance allele from ‘AWCS276’ while ‘S’ isolines did not possess the resistance allele. Based on their previous assessments on the differences in FCR resistance among these NILs, these three pairs of NILs are presented throughout this paper as: NIL1, NIL2 and NIL3 for NIL_4HL_6, NIL_CR4HL_3 and NIL_CR4HL_2, respectively. Seeds from these three pairs of NILs were treated with 2% available hypochlorite solution for 10 min and then were thoroughly rinsed with distilled water four times. Surface sterilised seeds were then placed on three layers of filter paper saturated with water and left to germinate.

FCR inoculation and assessment

A highly aggressive Fusarium pseudograminearum isolate (CS3096) collected in northern New South Wales and maintained in the CSIRO collection (Akinsanmi et al. 2004) was used in this study. The inoculum preparation, inoculation and FCR assessments were performed as described by Li et al. (2008). Briefly, inoculum was prepared on ½ strength potato dextrose agar (12 g potato dextrose broth with 16 g technical agar per litre) (Becton, Dickenson Difco™ Potato Dextrose Broth, Sparks, MD, USA). Inoculated plates were kept for 12 days at room temperature before the mycelium in the plates were scraped and discarded. After a further 7- to 12-day incubation under a combination of cool white and black fluorescent lights with 12-h photoperiod, the spores were then harvested by adding ~ 1 mL of double distilled water to the agar surface, agitating using a spreader rod and pipetting off the spore suspension. The concentration of spore suspension was adjusted to 1 × 106 spore mL−1 and then used directly for inoculation or stored at − 20 °C until needed. Tween 20 was added (0.1% v/v) to the spore suspension prior to use.

Seven seedlings were used in each of the biological replications and seedlings (4 days post-germination) were inoculated with F. pseudograminearum isolate (F. pseudograminearum inoculation) or distilled water (mock inoculation) following the protocol described previously (Chen et al. 2015). Briefly, seedlings (4 days post-germination) were immersed in either spore suspension (F. pseudograminearum inoculation) or in a water (mock inoculation) for 1 min and 2 seedlings were planted into a 5 cm square punnet (Rite Grow Kwik Pots, Garden City Plastics, Australia) containing sterilised University of California mix C (50% sand and 50% peat v/v). The punnets were arranged in a randomised block design in a controlled environment facility (CEF). The settings for the CEF were: 25/16(± 1) °C day/night temperature and 65%/85% day/night relative humidity, and a 14-h photoperiod with 500 mol m−2 s−1 photon flux density at the level of the plant canopy. Samples were harvested by cutting the shoot bases (4 cm) at 4 days post inoculation (dpi) and snap-frozen in liquid nitrogen and kept at − 80 °C until processed. The sampling time point was selected based on a previous RNA-seq study (Ma et al. 2014).

RNA extraction, library construction and sequencing

Samples were crushed into fine powder in 1.5 μL microcentrifuge tubes using sterilised metal beads and an Oscillating ball mill MM400 (Retsch GmbH, Germany). Total RNA was extracted using a RNeasy plant mini kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions, using RLC buffer and including the optional on-column DNase-I digestion. The concentration of each RNA sample was determined by the absorbance (Abs) at 260 nm and quality by observing DNA contamination at 280 nm and protein/salt contamination at 230 nm using Nanodrop-1000 Spectrophotometer. The degradation and contamination of all RNA samples were assessed by running total RNA on 1% agarose gels. RNA samples (10 μg each) were sent to Australian Genome Research Facility Ltd (Parkville, Victoria, Australia) and sequencing was performed using the Illumina HiSeq-2000 to produce 100-bp paired-end reads. Two technical replications were run for each of the 32 (16 for NIL1 and 8 each for NIL2 and NIL3) RNA-seq libraries. The RNA sequencing files have been made available at the National Centre for Biotechnology Information (NCBI) with the accession number of PRJNA392021.

RNA-seq analyses

A graphical overview of the experimental design for differential gene expression analysis and SNP calling has been provided (Fig S1). All the commands used for trimming raw data and analysing trimmed reads are supplied as a supplementary file (File S1). Briefly, raw reads were trimmed using SolexaQA ++ v3.1.3 with minimum Phred quality value of 30 and minimum final read length of 70 bp. The Tuxedo RNA-seq analysis pipeline (Trapnell et al. 2012) was used to map filtered reads with the annotated genome assembly of the barley variety ‘Morex’ (Mascher et al. 2017). FastQC (version 0.11.2) was used as a preliminary check that the PHRED scores were acceptable. TopHat2 (version 2.0.13) was used for read alignment with default parameters applied. During alignment, a maximum of two substitutions were allowed and multi-aligned reads were discarded. Two additional mismatches were allowed for the first 12 bp of the reads to account for Illumina sequencing artefacts.

Differential gene expression analysis using the Tuxedo pipeline

To measure the level of expression, the quantification of transcript abundance in the samples was conducted with Cufflinks v2.0.2 (Roberts et al. 2011). Assemblies were produced separately for each of the 32 libraries and then parsimoniously merged with the reference genome annotation using Cuffmerge v2.0.2. Changes in the relative abundance of transcripts between mock and treatments were estimated using Cuffdiff, which calculates the number of fragments per kilobase of exon per million reads mapped (FPKM) for each transcript and summarises them for each group of transcripts (Mortazavi et al. 2008).

Prior to the expression analysis, technical replicates for each genotype-treatment sample per NIL were merged together. In total, two pairwise comparisons between genotypes were conducted. These are summarised throughout the paper in the following way: SM_v_RM and SI_v_RI, where ‘M’ for ‘Mock’; ‘I’ for F. pseudograminearum infection; ‘S’ for the susceptible isoline and ‘R’ for the resistant isoline. DEGs were determined with as adjusted p value threshold of ≤ 0.05 and log2 expression fold change of ≥ 1 or ≤ − 1 or ‘inf’ (where the FPKM value in one condition is zero and the other is not).

Genes responsive to FCR infection were identified by two pairwise comparisons between treatments: SM_v_SI and RM_v_RI. The responsive genes after F. pseudograminearum infection compared with mock were identified following the same method as DEGs: threshold of FDR ≤ 0.05 and the absolute value of log2 fold change ≥ 2 or ≤ − 2 or ‘inf’.

Validation of differential gene expression patterns by qRT-PCR

Four transcripts (MLOC_12581.1, MLOC_67531.6, MLOC_10149.1 and MLOC_71136.2) were arbitrarily selected from the identified DEGs between ‘R’ and ‘S’ isolines for validating the RNA-seq data (Fig. S2). Quantitative real-time PCR (qRT-PCR) was used for validation using the actin gene as the reference (Liu et al. 2012). Primers were designed using the software Primer-BLAST ( and listed in Table S1. F. pseudograminearum inoculation, tissue sampling and RNA extraction were done using the methods as described before and three biological replications in two separate wells (technical replication) were used. For synthesising cDNA and analysing their expression, we followed the methods reported by Ma et al. (2013b). The average values from the two technical replications were used for each biological replicate. The relative fold changes were calculated using the comparative CT method (2−∆∆CT).

Identification of SNPs and validation by resequencing

To study differentially expressed alleles that are associated with FCR tolerance, SNPs were detected in each accession by comparing the transcript sequences to the Morex reference assembly. For each accession, all four replicate libraries (R1 reads only) were concatenated after removing low-quality sequences to generate the deepest and widest possible transcriptome representation. Each of the four concatenated files was mapped to the Morex assembly using Biokanga align (Stephen et al. 2012).

For SNPs identification, the trimmed sequence read files for all four biological replicates in each genotype were pooled (‘M’ and ‘I’ read files were also pooled together in the case of NIL1) and used for alignment. The alignment of reads to the reference sequences was performed with a maximum of two mismatches per read. SNPs between the resistant and susceptible isolines were identified using the Biokanga snpmarkers sub-process (Stephen et al. 2012) with a minimum 80% score (the percentage of a given nucleotide at a SNP position is at least 80% in the resistant or susceptible isolines).

Four genes with SNPs, AK_252954, AK_369386, Morex_Contig_47222 and Morex_Contig_244003, were selected for validation by resequencing. Primers were designed using Primer-BLAST ( and listed in Table S1. Genomic DNA from all the isolines was extracted using the hexadecyltrimethylammonium bromide (CTAB) method (Murray and Thompson 1980). PCR amplification and sequencing were conducted based on the methods described by Ma et al. (2013a) with annealing temperatures ranging from 55 to 60 °C depending on the primers (Table S1). Each of the four SNPs identified were confirmed by Sanger sequencing (Fig. S3).

Gene annotation and GO term enrichment analysis

BLAST, mapping and annotation steps were performed using standard running parameters within BLAST2GO (Conesa et al. 2005). DEGs from NIL1 SM_vs_SI and RM_vs_RI were separated into up-regulated and down-regulated genes and used as individual test sets for enrichment analysis using Fisher’s exact testing (p value < 0.05). In a similar manner, DEGs from NIL1, NIL2 and NIL3 SI_vs_RI comparisons were divided into genes expressed more highly in R isoline or S isoline and were used as individual test sets for enrichment analysis using the same parameters.


RNA-seq was performed to measure the transcriptome changes within a single NIL pair following F. pseudograminearum- or mock inoculation. The numbers of genes that were upregulated in response to Fusarium detected from the ‘R’ lines was 1737 (RM_v_RI) and from ‘S’ lines were 1377 (SM_v_SI). A smaller number of genes were down-regulated by Fusarium; 199 and 234 from the ‘R’ and ‘S’ isolines, respectively (Fig. 1a). Following F. pseudograminearum infection (compare to those in the mock), 859 were differentially expressed in both ‘R’ and ‘S’ isolines, whereas 1077 and 752 were differentially expressed only in ‘R’ and ‘S’ isolines, respectively. The differential expression results were verified through qRT-PCR analysis against four genes that were selected from the DEGs between the ‘R’ and ‘S’ isolines. The expression patterns of these four genes assessed by qRT-PCR were consistent with those obtained from the RNA-seq analysis (Fig. S2).
Fig. 1

Differentially expressed genes (DEGs) within resistant and susceptible isolines of NIL1 following F. pseudograminearum infection and mock treatment (SM_v_SI and RM_v_RI). a An overview of the number of DEGs. b A Venn diagram showing the numbers of up- and down-regulated genes in the resistant isolines compared with those in the susceptible isolines. DEGs were determined with the threshold of FDR ≤ 0.05 and the absolute value of log2 fold change ≥ 1 or ≤ − 1 or ‘inf’ (the value of one comparative object is zero and the other one is not). Symbols are: ‘R’ for resistant isolines; ‘S’ for susceptible isolines, ‘M’ for Mock inoculation and ‘I’ for F. pseudograminearum infection

Gene ontology enrichment analysis of Fusarium responsive genes reveals highly different responses to infection within resistant and susceptible isolines

To detect how global transcriptional response to Fusarium infection differs between resistant and susceptible isolines, gene ontology (GO) term enrichment analysis was performed on genes upregulated during infection from RM_vs_RI and SM_vs_SI comparisons. The response to infection in the resistant isoline was enriched for multiple defence-related processes including defense response to fungus (GO:0050832). Multiple defence related to biosynthesis of compounds producing physical barriers including lignin catabolic process (GO:0046274), lignin biosynthetic process (GO:0009808), isoprenoid biosynthetic process (GO:0008299) and cutin biosynthetic process (GO:0010143) were over-represented. Also, genes associated with pathways for production of phytoalexins, reactive oxygen species and signalling molecules including aromatic amino acid family catabolic process (GO:0009074), cinnamic acid biosynthetic process (GO:0009800), oxoacid metabolic process (GO:0043436), phenylpropanoid catabolic process (GO:0046271), cinnamic acid biosynthetic process (GO:0009800), phenylalanine ammonia-lyase activity (GO:0045548) and beta-glucosidase activity (GO:0008422) were also enriched.

By contrast, the response in the NIL1 susceptible isoline yielded notably fewer defence-associated terms with only defense response to fungus (GO:0050832) and oxoacid metabolic process (GO:0043436) enriched. Localised signalling terms were also enriched including protein serine/threonine kinase activity (GO:0004674) and transmembrane receptor activity (GO:0099600).

Using a multi-NIL approach for differential expression analysis reduces the number of candidate genes but is highly sensitive to the degree of infection

To test the utility of multi-NIL method for transcriptome analysis following F. pseudograminearum infection, we used three NIL pairs for analysing transcriptional differences between them. A total of 1294 genes were expressed to a significantly higher level in ‘R’ isolines, of which 61 were differentially expressed in NIL1, 36 were in NIL2 and 1231 were in NIL3. Also, a total of 2097 genes expressed more highly in ‘S’ isolines were identified, of which 607 were expressed in ‘S’ isolines of NIL1, 43 were in NIL2 and 1734 in NIL3 (Fig. 2a).
Fig. 2

Differentially expressed genes (DEGs) between resistant and susceptible isolines among NILs following F. pseudograminearum infection. a An overview of the number of DEGs. b A Venn diagram showing the numbers of up- and down-regulated genes in the resistant isolines compared with those in the susceptible isolines. DEGs were determined with the threshold of FDR ≤ 0.05 and the absolute value of log2 fold change ≥ 1 or ≤ − 1 or ‘inf’ (the value of one comparative object is zero and the other one is not)

Venn diagram analysis of up-regulated genes failed to reveal any DEGs commonly expressed among all three NILs, whereas for down-regulated genes, 16 were commonly expressed (Fig. 2b). One possible driver of the large differences observed in DEG abundance between NILs could be the degree of infection within individual NIL pairs. To assess the degree of infection, F. pseudograminearum biomass was estimated by determining the proportion of fungal reads present within NIL pair RNA-seq read files. To this end, read files were aligned against the F. pseudograminearum genome assembly (Gardiner et al. 2012) using Tophat2 with default parameters applied. The proportion of fungal reads differed significantly between isolines with the abundance consistently greater in susceptible isolines relative to resistant isolines for each NIL (Figure S4). F. pseudograminearum abundance also differed significantly between NILs. These patterns were consistent with the number of DEGs identified between R and S isolines among the three NILs (Fig. 2). For genes differentially expressed between R and S isolines in at least two NILs, 279 co-differentially expressed genes were identified with 11 of these genes located to a region spanning ~ 9.8 Mbp within the previously mapped QTL region on 4H (Fig. 3).
Fig. 3

Numbers of expressed genes containing SNPs. A Venn diagram showing the numbers of unique and common SNPs between resistant and susceptible isolines among NILs following F. pseudograminearum infection

Enrichment analysis reveals biological processes differ significantly between resistant and susceptible isolines

Gene ontology enrichment analysis was also performed to identify differences in the response to infection between isolines. NIL3 exhibited the greatest degree of infection both in terms of number of differentially expressed host genes as well as fungal biomass accumulation. Within NIL3, the resistant isoline exhibited defence-related GO terms including plant-type hypersensitive response (GO: 0009626), detection of stimulus (GO: 0051606), defense response to bacterium (GO:0042742), cellular response to stress (GO:0080135) and response to chitin (GO:0010200). A strong inference was made for activation of salicylic acid-mediated systemic acquired resistance with salicylic acid biosynthetic process (GO:0009697), salicylic acid-mediated signalling pathway (GO:0009862), systemic acquired resistance (GO:0009862) and regulation of plant-type hypersensitive response (GO:0010363) terms all enriched. Evidence for the regulation of systemic acquired resistance and hypersensitive response was also found with enrichment of negative regulation of defence response (GO:0031348), regulation of programmed cell death (GO:0043067), regulation of hydrogen peroxide metabolic process (GO:0010310) and regulation of reactive oxygen species metabolic process (GO:2000377). Several terms associated with terpenoid production were enriched including isopentenyl diphosphate metabolic process, isoprenoid biosynthetic process (GO:0006720), isoprenoid metabolic process (GO:0006720), terpenoid biosynthetic process (GO:0016114) as well as tetraterpenoid biosynthetic process and tetraterpenoid metabolic process (GO:0016108), carotenoid metabolic process (GO:0016116) and carotenoid biosynthetic process (GO:0016117). The resistant isoline was also enriched for many growth and related terms such as photosynthetic electron transport in photosystem I (GO:0009773), chlorophyll biosynthetic process (GO:0015995) and also sugar metabolism like pentose-phosphate shunt (GO:0006098), starch biosynthetic process (GO:0019252), response to sucrose (GO:0009744) and response to fructose (GO:0009750).

For the sensitive isoline, enriched terms included defence response to fungus (GO:0050832), response to wounding (GO:0009611) and chitinase activity (GO:0004568), indicating preliminary perception and response to the pathogen was occurring in the ‘S’ isoline. However, classical defence response terms indicating activation of systemic acquired resistance or induced systemic resistance were absent. Terms associated with detoxification of mycotoxins and other xenobiotic compounds including quercetin 3-O-glucosyltransferase activity (GO:0080043), quercetin 7-O-glucosyltransferase (GO:0080044) activity and glutathione transferase activity (GO:0004364) were enriched. Several defence-related metabolite pathways were also enriched including lignin metabolic process (GO:0009808), phenylpropanoid biosynthetic process (GO:0009699) and phenylpropanoid metabolic process (GO:0009698), flavonoid metabolic process (GO:0009812) and flavonoid biosynthetic process (GO:0009813). The patterns of gene ontology enrichment observed indicate the resistant isolines deploy a stronger and earlier systemic defence response potentially leading to the reduced fungal biomass and increased resistance observed.

Identifying inter-isoline SNPs conserved across NIL pairs inferred putative candidate genes underpinning the 4H FCR resistance QTL

A total of 4566 homozygous SNPs were detected between isolines for the three NIL pairs, of which 2354 were found in NIL1, 1993 in NIL2 and 1379 in NIL3 with 493 common for all NILs. All 493 common SNPs were located within the target QTL interval (Fig. 4). SNPs were identified within an interval encompassing the QTL region of ~ 21 Mbp in NIL1, ~ 32 Mbp in NIL2 and ~ 20 Mbp in NIL3. In contrast, the interval in which common SNPs across all three NILs were identified spanned a physical distance of ~ 6.3 Mbp. Observation of genes consistently differentially expressed within at least two NILs identified a region spanning ~ 9.8 Mbp enclosing the QTL. These results demonstrate the utility of DEG and SNP calling across multiple NILs to improve the resolution for mapping of the underlying QTL region.
Fig. 4

Physical distribution of genes a differentially expressed in response to Fusarium infection, b differentially expressed between R and S isolines under infection; and c SNPs in the region surrounding the QTL. For each diagram, surrounding markers with their genetic distances are provided as a point of reference. a Physical distribution of genes differentially expressed in response to Fusarium infection in the R isoline only (left), both R and S isolines (middle) and the S isoline only (right). b Physical distribution of genes differentially expressed between R and S isolines under infection in NIL1 (far left), NIL2 (inner left), NIL3 (inner right) and the consensus map which shows only genes found to be differentially expressed in two or more NIL pairs. c Physical distribution of SNPs in NIL1 (far left), NIL2 (inner left), NIL3 (inner right) and the consensus map which shows only SNPs common to all three NILs. Black boxes in b and c indicates the regions defined by DEGs and SNPs within individual NILs compared to the consensus region where common DEGs and SNPs were identified

Putative candidate genes underpinning the 4HL resistance source

One hundred and forty-four genes could be inferred as potential candidate genes underpinning the 4HL resistance QTL due to their location within the revised ~ 6.3 Mbp flanking region. From these, four protein encoding genes presented as highly interesting candidates based on differential expression patterns and SNP inclusions (HORVU4Hr1G083930, HORVU4Hr1G084740, HORVU4Hr1G085140, HORVU4Hr1G085750) (Table S2). These genes are annotated as an Alkane 1-monooxygenase/Omega-hydroxylase, predicted transporter (major facilitator superfamily), remorin, C-terminal region (Remorin_C) and hydroxymethylglutaryl-CoA synthase/hydroxymethylglutaryl coenzyme alpha-condensing enzyme, respectively. All four were up-regulated in response to infection only in the ‘S’ isoline of NIL1 with log2 fold changes of 2.43, 2.42, 2.38 and 3.51, respectively. HORVU4Hr1G084740 was also expressed to a significantly higher level in the ‘S’ isoline under infection in NIL3 and HORVU4Hr1G085750 was expressed more highly in both NIL1 and NIL3 ‘S’ isolines. Three of these candidate genes contained SNPs resulting in non-synonymous codon changes with five amino acid changes in HORVU4Hr1G083930, three amino acid changes in HORVU4Hr1G084740 and five amino acid changes in HORVU4Hr1G085750. In addition, a non-protein coding gene was upregulated in response to infection in the NIL1 ‘S’ isoline (log2 fold change of 3.92), was expressed more highly under infection in the S isoline in NIL3 (log2 fold change of 3.27) and contained 15 SNPs.

Any such gene, bearing one or more SNPs, displaying different expression levels between resistant and sensitive lines and responding differently between resistant and sensitive lines, presents as a candidate with great potential and illustrates the power of a multi-NIL RNA-seq approach to both select probable candidate genes and provide functional insight into the role they might play. Further fine-mapping approaches will be required to identify the genetic element underpinning the 4HL locus.


In this study, RNA-seq analysis was performed using three pairs of NILs for a large-effect locus on chromosome arm 4HL conferring FCR resistance in barley. The gene ontology term (GO term) analysis was performed to identify which molecular pathways and processes responded differently between resistant and sensitive isolines and revealed striking differences in responses between isolines. Analysis of genes differentially expressed between resistant and susceptible isolines under infected conditions was conducted across three NIL pairs within the QTL region, though only two NIL pairs were considered informative due to phenotypic different in the NIL2 pair. Finally, SNP analysis was performed to find genes with SNP differences between all three pairs of isolines. Combining DEG and SNP analyses across multiple NILs reduced the thousands of potential candidate genes identified for individual pairs down to 144 plausible candidates, thus demonstrating the advantage of taking a multi-NIL approach for identifying candidate genes putatively underpinning QTL for highly quantitative traits.

Similar to RNA-seq analysis of wheat NILs (Ma et al. 2014), the different genetic backgrounds of each NIL pair have contributed to significant differences in the response to F. pseudograminearum infection. While individual NIL pairs contribute large sets of DEGs and SNPs due to differing genetic background effects, this study demonstrated combining multiple NILs is effective in greatly reducing the number of candidate genes underpinning QTL.

The disparity in DEG calling results between NILs, particularly influenced by infection levels in NIL2, highlighted the importance of ensuring even levels of infection and overall similar phenotypic appearance across NIL pairs used in such studies. NIL2 displays a dwarf genotype (Fig. S5) and had far fewer DEGs compared to the other two NILs. It has been reported before in several studies that dwarf genotypes tend to show less symptoms of FCR than tall genotypes (Chen et al. 2014; Li et al. 2009; Liu et al. 2010). In these studies, it has been claimed that the dwarfing gene increases cell density by predominantly reducing cell lengths and thus, the higher cell densities of shorter plants could be responsible to their better FCR resistance. Later, the histological and qRT-PCR analysis against two pairs of NILs for the dwarf barley genotypes supported the argument that the development of FCR in the dwarf isolines were indeed slower than that in the tall isolines (Bai and Liu 2015).

We identified a total of 2688 FCR responsive genes in NIL1 (‘mock’ controls were not used in assessing the other two NILs). The gene ontology analysis revealed striking differences between expression patterns observed in ‘R’ and ‘S’ isolines. Gene enrichment analysis indicates the resistant isoline may be deploying a defence response based on salicylic acid-mediated systemic defence signalling, deposition of structural barriers such as cutin, suberin and lignin to limit pathogen ingress, and production of anti-fungal metabolites. Similar responses have been previously implicated in defence responses against F. pseudograminearum in hexaploid wheat (Desmond et al. 2008; Powell et al. 2016). The strong inference toward salicylic acid-mediated defence response is interesting since systemic acquired resistance is generally associated with resistance to biotrophic pathogens and not necrotrophs since deployment of hypersensitive responses can be exploited by necrotrophic pathogens to promote host cell death. Studies on the effect of salicylic acid on Fusarium head blight have indicated salicylic acid provides a measure of resistance to Fusarium graminearum infection (Makandar et al. 2012; Qi et al. 2012). The number of enriched terms associated with the regulation of hypersensitive response perhaps suggests that the resistant isoline is able to utilise systemic acquired resistance mechanisms to inhibit infection without leading to widespread cell death responses which would favour the pathogen. In contrast, the susceptible isoline response yielded relatively few defence-related GO term enrichments and lacked any inference of systemic defence signalling. Such differences in response might explain the increased resistance observed in the ‘R’ isoline and may also suggest the gene(s) underpinning the 4H QTL may regulate the systemic defence response against Fusarium infection. Interpretation of transcriptomic responses often rest on the assumption that mRNA abundance correlates with protein abundance and activity which is often not the case. Therefore, further observation of the differences in responses between isolines using proteomic or ribosome-associated mRNA profiling approaches would be useful (Sablok et al. 2017). Future work would benefit from the inclusion of the mock-inoculated samples across multiple NIL pairs to isolate differences between resistant and susceptible isolines driven by the QTL rather than genetic background.

We hypothesised that the use of multiple pairs of NILs might allow the identification of better defined sets of candidate genes underlying the targeted locus, so we used three pairs of NILs to identify SNPs and DEGs. As expected, the use of multiple pairs of NILs decreased the number of candidate genes underpinning the 4H QTL based on DEG and SNP calling appreciably. From the results of this study, we conclude that SNP calling provides superior resolution compared to DEGs since a smaller revised QTL interval was identified using SNPs (6.3 versus 9.8 Mbp) and also since all common SNPs identified were mapped within the previously identified QTL region where a relatively small proportion of common DEGs mapping into the QTL interval.

Gene expression patterns are highly influenced by physiological state at time of sampling; as such the set of DEGs identified between ‘R’ and ‘S’ isolines will vary greatly across NIL pairs. Therefore, taking a multiple NIL approach for DEG analysis is still beneficial since it captures a greater number of DEG events between isolines and allows removal of genetic background influences where these are only found in a single NIL pair. Instead, we recommend common DEGs should be inferred where common in at least two NILs rather than when only identified in all NIL pairs. The variability in DEG patterns could be further compounded when undertaking similar approaches to characterise QTL in polyploid species such as hexaploid wheat. Phenomena such as homoeolog expression bias and homoeolog induction bias during infection may confound results if insufficient care is applied during read mapping to ensure stringent alignment to individual homoeologs (Powell et al. 2017). The use of multiple pairs of NILs allowed the identification of better defined sets of candidate genes underpinning the targeted locus. These genes are now being used as markers in fine-mapping the FCR locus on chromosome arm 4HL based on a NIL-derived population.

We also identified 5726 homozygous SNPs with 493 SNPs common between all three NILs. In support of our hypothesis, all 493 identified common SNPs were mapped within the previously identified QTL interval. The range of their positions was found from 111 to 128 cM, where the targeted FCR locus also resides (Chen et al. 2013a). Functional annotations of these genes provide insights into responses regulated by the 4HL locus and will inform future efforts toward fine-mapping the FCR locus in barley. By comparing across multiple NIL pairs, SNP analysis refined the QTL region to a physical region of ~ 6.3 Mbp which encompassed 144 genes. From this set of candidate genes underpinning the QTL, four genes appeared particularly interesting based on SNP inclusion, expression patterns and function. These four genes are functionally diverse but each has functions potentially related to plant–pathogen interactions. Alkanes play an important role in plants in cuticle development and wax deposition (Aarts et al. 1995) which form barriers to pathogen infection. Transporter proteins have a known role in mediating resistance to fungal pathogens with an ABC transporter encoded by the resistance gene Lr34 providing tractable adult plant resistance to leaf, stem and stripe rust as well as powdery mildew in hexaploid wheat (Krattinger et al. 2009). Remorin proteins have been previously shown to inhibit the ingress of potato mosaic virus in tomato (Raffaele et al. 2009), have been implicated in increased susceptibility to the oomycete, Phytophthora infestans (Bozkurt et al. 2014) and play an important role in mediating plant interactions with beneficial microbes (Lefebvre et al. 2010). Hydroxymethylglutaryl-CoA synthase functions in isoprenoid biosynthesis through production of the precursor isopentenyl pyrophosphate which is in turn metabolised into terpene, sterol and carotenoid compounds (Nagel et al. 2014; Pankratov et al. 2016). These classes of compounds possess known anti-microbial properties and were functionally enriched in the ‘R’ isoline.

The observation that genes within the refined QTL region which responded to infection only did so in the sensitive isolines was surprising since genes conferring increased resistance would usually be expected to be induced by infection. However, these genes remain interesting candidates as the infection responsive alleles in the sensitive isoline may promote susceptibility. Future mapping and characterisation of genes within the QTL region will determine whether increased resistance observed is due to inclusion of a resistance gene from the resistant parent or removal of a susceptibility allele from the sensitive parent.

RNA-seq provides useful data to both map loci underpinning important crop traits and dissect the role these genes play within expression of these traits as well as identify novel QTL. Bulk segregant RNA-seq approaches have identified QTL underpinning important agronomic traits such as kernel row number in maize (Liu et al. 2016a, b), pod number in canola (Ye et al. 2017) and nitrogen stress tolerance in sorghum (Gelli et al. 2017).

RNA-seq has also proven useful for identifying candidate genes underpinning QTL. One study utilising a four-way multi-parent advanced generation inter-cross (MAGIC) population identified two candidate genes underpinning a dormancy QTL (Barrero et al. 2015). PM19-A1 was implicated due to strong differential expression between dormant and non-dormant lines while PM19-A2 presented as an interesting candidate based on inclusion of SNPs resulting in amino acid changes within the encoded protein. In cotton, an integrated approach using both fine-mapping and RNA-seq implicated four genes within the fine-mapped region as interesting candidates based on different transcriptional response in high-quality versus low-quality lines (Liu et al. 2016a, b).

RNA-seq analysis of Fusarium head blight in wheat examined near isogenic lines differing in presence of Fhb1 treated with deoxynivalenol and infected with Fusarium graminearum (Hofstad et al. 2016). This approach identified infection responsive genes specifically associated with the Fhb1 locus but was not able to successfully identify Fhb1 itself. In barley, a transcriptomic approach was applied to near isogenic lines differing in resistance QTL from a resistant cultivar, Chevron (Huang et al. 2016). DEG patterns indicated resistant NILs possessing either the 2Hb8 or 6Hb7 QTL enacted an earlier and stronger classical defence response; however, this approach did not refine the QTL region or identify a reduced set of candidate genes. In pursuit of the genes underpinning these resistance QTL, further fine-mapping appears necessary to reduce the number of potential candidates prior to functional characterisation approaches.

While RNA-seq approaches have proven useful for identification of candidate genes and understanding the molecular mechanism underpinning the effect of QTL, it remains to be shown that such approaches can in fact correctly identify the causal gene. In any case, RNA-seq is inadequate in itself to isolate and describe the function of the underlying gene but instead presents as a useful complementary tool to guide further fine-mapping approaches and inform attempts at functional characterisation. In this study, RNA-seq analysis enabled mapping of the resistance QTL to a considerably narrower region than previously identified and identified genes of interest based on expression patterns and inclusion of non-synonymous SNPs. Whether any of the genes of interest identified within this analysis in fact underpin the QTL remains to be seen; however, we posit that undertaking multi-NIL RNA-seq is effective in reducing the potential genomic region enclosing the QTL and identifying candidate genes within this which underpin the QTL effect.

Author contribution statement

SS, MZ, DMG and CL conceived the original screening and research plans; AH performed the experimental work; JS, and ML provided technical assistance to AH; AH and JJP analysed the data; AH and JJP wrote the article with contributions from all the authors; SS, MZ DMG, and CL supervised and complemented the writing.



Work reported in this publication was partially funded by the Grains Research and Development Corporation, Australia (Project no. CFF00010). AH is grateful to University of Tasmania, Australia, and Khulna University, Bangladesh, for financial supports during the tenure of his Ph.D. studentship.

Supplementary material

122_2017_3023_MOESM1_ESM.docx (2.1 mb)
Supplementary material 1 (DOCX 2153 kb)


  1. Aarts M, Keijzer CJ, Stiekema WJ, Pereira A (1995) Molecular characterization of the CER1 gene of arabidopsis involved in epicuticular wax biosynthesis and pollen fertility. Plant Cell 7:2115–2127CrossRefPubMedPubMedCentralGoogle Scholar
  2. Akinsanmi O, Mitter V, Simpfendorfer S, Backhouse D, Chakraborty S (2004) Identity and pathogenicity of Fusarium spp. isolated from wheat fields in Queensland and northern New South Wales. Crop Pasture Sci 55:97–107CrossRefGoogle Scholar
  3. Bai Z, Liu C (2015) Histological evidence for different spread of Fusarium crown rot in barley genotypes with different heights. J Phytopathol 163:91–97CrossRefGoogle Scholar
  4. Barrero JM, Cavanagh C, Verbyla KL, Tibbits JF, Verbyla AP, Huang BE, Rosewarne GM, Stephen S, Wang P, Whan A (2015) Transcriptomic analysis of wheat near-isogenic lines identifies PM19-A1 and A2 as candidates for a major dormancy QTL. Genome Biol 16:93CrossRefPubMedPubMedCentralGoogle Scholar
  5. Blencowe BJ, Ahmad S, Lee LJ (2009) Current-generation high-throughput sequencing: deepening insights into mammalian transcriptomes. Genes Dev 23:1379–1386CrossRefPubMedGoogle Scholar
  6. Bozkurt TO, Richardson A, Dagdas YF, Mongrand S, Kamoun S, Raffaele S (2014) The plant membrane-associated REMORIN1.3 accumulates in discrete perihaustorial domains and enhances susceptibility to phytophthora infestans. Plant Physiol 165(3):1005–1018CrossRefPubMedPubMedCentralGoogle Scholar
  7. Cavanagh CR, Chao S, Wang S, Huang BE, Stephen S, Kiani S, Forrest K, Saintenac C, Brown-Guedira GL, Akhunova A (2013) Genome-wide comparative diversity uncovers multiple targets of selection for improvement in hexaploid wheat landraces and cultivars. Proc Natl Acad Sci 110:8057–8062CrossRefPubMedPubMedCentralGoogle Scholar
  8. Chakraborty S, Liu C, Mitter V, Scott J, Akinsanmi O, Ali S, Dill-Macky R, Nicol J, Backhouse D, Simpfendorfer S (2006) Pathogen population structure and epidemiology are keys to wheat crown rot and Fusarium head blight management. Australas Plant Pathol 35:643–655CrossRefGoogle Scholar
  9. Chen G, Liu Y, Ma J, Zheng Z, Wei Y, McIntyre CL, Zheng Y-L, Liu C (2013a) A novel and major quantitative trait locus for Fusarium crown rot resistance in a genotype of wild barley (Hordeum spontaneum L.). PLoS One 8:e58040CrossRefPubMedPubMedCentralGoogle Scholar
  10. Chen G, Liu Y, Wei Y, McIntyre C, Zhou M, Zheng Y-L, Liu C (2013b) Major QTL for Fusarium crown rot resistance in a barley landrace. Theor Appl Genet 126:2511–2520CrossRefPubMedGoogle Scholar
  11. Chen G, Yan W, Liu Y, Wei Y, Zhou M, Zheng Y-L, Manners JM, Liu C (2014) The non-gibberellic acid-responsive semi-dwarfing gene uzu affects Fusarium crown rot resistance in barley. BMC Plant Biol 14(1):22CrossRefPubMedPubMedCentralGoogle Scholar
  12. Chen G, Habib A, Wei Y, Zheng Y-L, Shabala S, Zhou M, Liu C (2015) Enhancing Fusarium crown rot resistance by pyramiding large-effect QTL in barley. Mol Breed 35:1–8CrossRefGoogle Scholar
  13. Conesa A, Götz S, García-Gómez JM, Terol J, Talón M, Robles M (2005) Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21:3674–3676CrossRefPubMedGoogle Scholar
  14. Desmond OJ, Manners JM, Schenk PM, Maclean DJ, Kazan K (2008) Gene expression analysis of the wheat response to infection by Fusarium pseudograminearum. Physiol Mol Plant Pathol 73(1–3):40–47CrossRefGoogle Scholar
  15. Ding L, Xu H, Yi H, Yang L, Kong Z, Zhang L, Xue S, Jia H, Ma Z (2011) Resistance to hemi-biotrophic F. graminearum infection is associated with coordinated and ordered expression of diverse defense signaling pathways. PLoS One 6:e19008CrossRefPubMedPubMedCentralGoogle Scholar
  16. Gardiner DM, McDonald MC, Covarelli L, Solomon PS, Rusu AG, Marshall M, Kazan K, Chakraborty S, McDonald BA, Manners JM (2012) Comparative pathogenomics reveals horizontally acquired novel virulence genes in fungi infecting cereal hosts. PLoS Pathog 8:e1002952CrossRefPubMedPubMedCentralGoogle Scholar
  17. Gelli M, Konda AR, Liu K, Zhang C, Clemente TE, Holding DR, Dweikat IM (2017) Validation of QTL mapping and transcriptome profiling for identification of candidate genes associated with nitrogen stress tolerance in sorghum. BMC Plant Biol 17:123CrossRefPubMedPubMedCentralGoogle Scholar
  18. Habib A, Shabala S, Shabala L, Zhou M, Liu C (2016) Near-isogenic lines developed for a major QTL on chromosome arm 4HL conferring Fusarium crown rot resistance in barley. Euphytica 209:555–563CrossRefGoogle Scholar
  19. Hofstad AN, Nussbaumer T, Akhunov E, Shin S, Kugler KG, Kistler HC, Mayer KF, Muehlbauer GJ (2016) Examining the transcriptional response in wheat near-isogenic lines to infection and deoxynivalenol treatment. Plant Genome. PubMedGoogle Scholar
  20. Huang Y, Li L, Smith KP, Muehlbauer GJ (2016) Differential transcriptomic responses to Fusarium graminearum infection in two barley quantitative trait loci associated with Fusarium head blight resistance. BMC Genom 17:387CrossRefGoogle Scholar
  21. Kazan K, Gardiner DM (2017) Transcriptomics of cereal-Fusarium graminearum interactions: what we have learned so far. Mol Plant Pathol. Google Scholar
  22. Keurentjes JJ, Bentsink L, Alonso-Blanco C, Hanhart CJ, Blankestijn-De Vries H, Effgen S, Vreugdenhil D, Koornneef M (2007) Development of a near-isogenic line population of Arabidopsis thaliana and comparison of mapping power with a recombinant inbred line population. Genetics 175:891–905CrossRefPubMedPubMedCentralGoogle Scholar
  23. Krattinger SG, Lagudah ES, Spielmeyer W, Singh RP, Huerta-Espino J, McFadden H, Bossolini E, Selter LL, Keller B (2009) A putative ABC transporter confers durable resistance to multiple fungal pathogens in wheat. Science 323:1360–1363CrossRefPubMedGoogle Scholar
  24. Kugler KG, Siegwart G, Nussbaumer T, Ametz C, Spannagl M, Steiner B, Lemmens M, Mayer KF, Buerstmayr H, Schweiger W (2013) Quantitative trait loci-dependent analysis of a gene co-expression network associated with Fusarium head blight resistance in bread wheat (Triticum aestivum L.). BMC Genom 14:1CrossRefGoogle Scholar
  25. Lefebvre B, Timmers T, Mbengue M, Moreau S, Hervé C, Tóth K, Bittencourt-Silvestre J, Klaus D, Deslandes L, Godiard L (2010) A remorin protein interacts with symbiotic receptors and regulates bacterial infection. Proc Natl Acad Sci 107:2343–2348CrossRefPubMedPubMedCentralGoogle Scholar
  26. Li X, Liu C, Chakraborty S, Manners JM, Kazan K (2008) A simple method for the assessment of crown rot disease severity in wheat seedlings inoculated with Fusarium pseudograminearum. J Phytopathol 156:751–754CrossRefGoogle Scholar
  27. Li H, Zhou M, Liu C (2009) A major QTL conferring crown rot resistance in barley and its association with plant height. Theor Appl Genet 118:903–910CrossRefPubMedGoogle Scholar
  28. Liu Y, Yang X, Ma J, Wei Y, Zheng Y, Ma H, Yao J, Yan G, Wang Y, Manners J (2010) Plant height affects Fusarium crown rot severity in wheat. Phytopathology 100:1276–1281CrossRefPubMedGoogle Scholar
  29. Liu Y, Ma J, Yan W, Yan G, Zhou M, Wei Y, Zheng Y, Liu C (2012) Different tolerance in bread wheat, durum wheat and barley to Fusarium crown rot disease caused by Fusarium pseudograminearum. J Phytopathol 160:412–417CrossRefGoogle Scholar
  30. Liu C, Zhou Q, Dong L, Wang H, Liu F, Weng J, Li X, Xie C (2016a) Genetic architecture of the maize kernel row number revealed by combining QTL mapping using a high-density genetic map and bulked segregant RNA sequencing. BMC Genom 17:915CrossRefGoogle Scholar
  31. Liu D, Zhang J, Liu X, Wang W, Liu D, Teng Z, Fang X, Tan Z, Tang S, Yang J (2016b) Fine mapping and RNA-Seq unravels candidate genes for a major QTL controlling multiple fiber quality traits at the T 1 region in upland cotton. BMC Genom 17:295CrossRefGoogle Scholar
  32. Ma J, Yan G, Liu C (2012) Development of near-isogenic lines for a major QTL on 3BL conferring Fusarium crown rot resistance in hexaploid wheat. Euphytica 183:147–152CrossRefGoogle Scholar
  33. Ma J, Jiang Q-T, Zhang X-W, Lan X-J, Pu Z-E, Wei Y-M, Liu C, Lu Z-X, Zheng Y-L (2013a) Structure and expression of barley starch phosphorylase genes. Planta 238:1081–1093CrossRefPubMedGoogle Scholar
  34. Ma J, Jiang Q-T, Zhao Q-Z, Zhao S, Lan X-J, Dai S-F, Lu Z-X, Liu C, Wei Y-M, Zheng Y-L (2013b) Characterization and expression analysis of waxy alleles in barley accessions. Genetica 141:227–238CrossRefPubMedGoogle Scholar
  35. Ma J, Stiller J, Zhao Q, Feng Q, Cavanagh C, Wang P, Gardiner D, Choulet F, Feuillet C, Zheng Y-L (2014) Transcriptome and allele specificity associated with a 3BL locus for fusarium crown rot resistance in bread wheat. PLoS One 9:e113309CrossRefPubMedPubMedCentralGoogle Scholar
  36. Makandar R, Nalam VJ, Lee H, Trick HN, Dong Y, Shah J (2012) Salicylic acid regulates basal resistance to Fusarium head blight in wheat. MPMI 25:431–439CrossRefPubMedGoogle Scholar
  37. Mascher M, Gundlach H, Himmelbach A, Beier S, Twardziok SO, Wicker T, Radchuk V, Dockter C, Hedley PE, Russell J (2017) A chromosome conformation capture ordered sequence of the barley genome. Nature 544:427–433CrossRefPubMedGoogle Scholar
  38. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B (2008) Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods 5:621–628CrossRefPubMedGoogle Scholar
  39. Murray GM, Brennan JP (2009) Estimating disease losses to the Australian wheat industry. Australas Plant Pathol 38:558–570CrossRefGoogle Scholar
  40. Murray G, Brennan J (2010) Estimating disease losses to the Australian barley industry. Australas Plant Pathol 39:85–96CrossRefGoogle Scholar
  41. Murray MG, Thompson WF (1980) Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res 8(19):4321–4326CrossRefPubMedPubMedCentralGoogle Scholar
  42. Nagel R, Berasategui A, Paetz C, Gershenzon J, Schmidt A (2014) Overexpression of an isoprenyl diphosphate synthase in spruce leads to unexpected terpene diversion products that function in plant defense. Plant Phys 164:555–569CrossRefGoogle Scholar
  43. Pankratov I, McQuinn R, Schwartz J, Bar E, Fei Z, Lewinsohn E, Zamir D, Giovannoni JJ, Hirschberg J (2016) Fruit carotenoid-deficient mutants in tomato reveal a function of the plastidial isopentenyl diphosphate isomerase (IDI1) in carotenoid biosynthesis. Plant J 88:82–94CrossRefPubMedGoogle Scholar
  44. Powell JJ, Fitzgerald TL, Stiller J, Berkman PJ, Gardiner DM, Manners JM, Henry RJ, Kazan K (2016) The defence-associated transcriptome of hexaploid wheat displays homoeolog expression and induction bias. Plant Biotechnol J 15:533–543CrossRefPubMedPubMedCentralGoogle Scholar
  45. Powell JJ, Carere J, Fitzgerald T, Stiller J, Covarelli L, Xu Q, Gubler F, Colgrave ML, Gardiner DM, Manners J, Henry RJ, Kazan K (2017) The Fusarium crown rot pathogen Fusarium pseudograminearum triggers a suite of transcriptional and metabolic changes in bread wheat (Triticum aestivum L.). Ann Bot 119:853–867PubMedGoogle Scholar
  46. Qi P-F, Johnston A, Balcerzak M, Rocheleau H, Harris LJ, Long X-Y, Wei Y-M, Zheng Y-L, Ouellet T (2012) Effect of salicylic acid on Fusarium graminearum, the major causal agent of fusarium head blight in wheat. Fungal Biol 116:413–426CrossRefPubMedGoogle Scholar
  47. Raffaele S, Bayer E, Lafarge D, Cluzet S, Retana SG, Boubekeur T, Leborgne-Castel N, Carde J-P, Lherminier J, Noirot E (2009) Remorin, a solanaceae protein resident in membrane rafts and plasmodesmata, impairs potato virus X movement. Plant Cell 21:1541–1555CrossRefPubMedPubMedCentralGoogle Scholar
  48. Roberts A, Pimentel H, Trapnell C, Pachter L (2011) Identification of novel transcripts in annotated genomes using RNA-Seq. Bioinformatics 27:2325–2329CrossRefPubMedGoogle Scholar
  49. Sablok G, Powell JJ, Kazan K (2017) Emerging roles and landscape of translating mRNAs in plants. Front Plant Sci. Google Scholar
  50. Samad-Zamini M, Schweiger W, Nussbaumer T, Mayer KF, Buerstmayr H (2017) Time-course expression QTL-atlas of the global transcriptional response of wheat to Fusarium graminearum. Plant Biotechnol J. PubMedPubMedCentralGoogle Scholar
  51. Smiley RW, Gourlie JA, Easley SA, Patterson L-M, Whittaker RG (2005) Crop damage estimates for crown rot of wheat and barley in the Pacific Northwest. Plant Dis 89:595–604CrossRefGoogle Scholar
  52. Steiner B, Kurz H, Lemmens M, Buerstmayr H (2009) Differential gene expression of related wheat lines with contrasting levels of head blight resistance after Fusarium graminearum inoculation. Theor Appl Genet 118:753–764CrossRefPubMedGoogle Scholar
  53. Stephen S, Cullerne D, Spriggs A, Helliwell C, Lovell D, Taylor J (2012) Biokanga: a suite of high performance bioinformatics applications. Accessed 15 Sept 2017
  54. Swanson-Wagner RA, DeCook R, Jia Y, Bancroft T, Ji T, Zhao X, Nettleton D, Schnable PS (2009) Paternal dominance of trans-eQTL influences gene expression patterns in maize hybrids. Science 326(5956):118–1120CrossRefGoogle Scholar
  55. Trapnell C, Williams BA, Pertea G, Mortazavi A, Kwan G, van Baren MJ, Salzberg SL, Wold BJ, Pachter L (2010) Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat Biotechnol 28:511–515CrossRefPubMedPubMedCentralGoogle Scholar
  56. Trapnell C, Roberts A, Goff L, Pertea G, Kim D, Kelley DR, Pimentel H, Salzberg SL, Rinn JL, Pachter L (2012) Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks. Nat Protoc 7:562–578CrossRefPubMedPubMedCentralGoogle Scholar
  57. Wang Z, Gerstein M, Snyder M (2009) RNA-Seq: a revolutionary tool for transcriptomics. Nat Rev Genet 10:57–63CrossRefPubMedPubMedCentralGoogle Scholar
  58. West MAL, Kim K, Kliebenstein DJ, van Leeuwen H, Michelmore RW, Doerge RW, St. Clair DA (2007) Global eQTL mapping reveals the complex genetic architecture of transcript-level variation in Arabidopsis. Genetics 175:1441–1450CrossRefPubMedPubMedCentralGoogle Scholar
  59. Xiao J, Jin X, Jia X, Wang H, Cao A, Zhao W, Pei H, Xue Z, He L, Chen Q (2013) Transcriptome-based discovery of pathways and genes related to resistance against Fusarium head blight in wheat landrace Wangshuibai. BMC Genom 14:1CrossRefGoogle Scholar
  60. Ye J, Yang Y, Chen B, Shi J, Luo M, Zhan J, Wang X, Liu G, Wang H (2017) An integrated analysis of QTL mapping and RNA sequencing provides further insights and promising candidates for pod number variation in rapeseed (Brassica napus L.). BMC Genom 18:71CrossRefGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2017

Authors and Affiliations

  1. 1.Commonwealth Scientific and Industrial Research Organization Agriculture and FoodSt LuciaAustralia
  2. 2.School of Land and Food and Tasmanian Institute of AgricultureUniversity of TasmaniaHobartAustralia
  3. 3.Biotechnology and Genetic Engineering DisciplineKhulna UniversityKhulnaBangladesh

Personalised recommendations