Genetics of Fusarium (F. pseudograminearum) crown rot resistance in barley

Fusarium crown rot (FCR) is a severe and chronic disease of barley and wheat. The disease is predominantly found in many parts of the semiarid regions worldwide. Growing resistant cultivars is an effective way to manage crown rot. Several quantitative trait loci (QTL) conferring FCR resistance have been detected in barley. As each of these loci conferring only partial resistance, this project was to investigate the effect of gene pyramiding on overall resistance. For this purpose, we developed and assessed two barley doubled haploid populations segregating for three large-effect QTL located on the long arms of chromosomes 1H, 3H and 4H, respectively. Significant effects were detected for each of the three QTL in both populations. Lines with any combination of two resistant alleles gave, on average, significantly better resistance than those with a single resistant allele only, and lines with resistant alleles from all three QTL gave the least FCR symptom. However, wide variations in FCR severity were detected for lines belonging to each of the groups with different numbers of resistant alleles. Significant effects of plant height on FCR were detected in both populations, and a significant association between heading date and FCR severity was also detected in one of the populations. We also found that the effects of a given resistant allele decreased with the increase in the number of resistant alleles. Overall, results from this study demonstrated that gene pyramiding can be an effective approach in improving FCR resistance and those lines with all three resistant alleles could be valuable for breeding programs. Most of the works on FCR resistance stopped after mapping the QTL. However, QTL mapping provides only limited resolution for a targeted locus due to the heterogeneity in genetic backgrounds in mapping populations. Thus molecular markers obtained from such studies can often be reliably used to tag a targeted QTL. One of the approaches to obtain populations with uniform genetic backgrounds is to develop and exploit a series of near isogenic lines (NILs). As part of our long term objectives to develop diagnostic markers and investigate functions of FCR resistance genes in cereals, we have developed 10 pairs of NILs for a major QTL conferring FCR resistance in barley. The locus locates on the long arm of chromosome 4H and the presence of the resistance allele reduced FCR severity by between 32.8 and 63.4% with an average of 43.9% across these NILs. Histological and quantitative PCR analyses confirmed that the rates of Fusarium infection and disease development were much lower in the resistant isolines compared with those in the susceptible isolines. The results from this study would facilitate efforts in cloning and functional analyses of genes conferring resistance to FCR. Also the availability of these NILs offers an excellent genetic resource for transcriptional analysis. RNA sequencing (RNA-seq) has become a powerful tool for transcriptome analysis which is not only highly sensitive and efficient for identifying differentially expressed genes (DEGs) but can also be used for detecting single nucleotide polymorphisms (SNPs) in transcribed genes that co-locate with a target locus when combined with genomic and genetic analysis. We used three sets of NILs (197 Gb sequences) to examine transcriptional changes associated with FCR resistance locus located on the long arm of chromosome 4H. Owing to the lack of annotated genes in barley genome currently, we also used the expressed reads from our RNAseq data to find out the unannotated genes. When considered both annotated and unannotated genes, a total of 2,359 genes were expressed to a significantly higher level, of which 103 were expressed in NIL1, 28 were in NIL2 and 2,228 were in NIL3. Also a total of 4,074 down-regulated annotated and unannotated DEGs were identified, of which 947 were expressed in 'S‘ isolines of NIL1, 120 were in NIL2 and 3,007 in NIL3. It has been hypothesized in this experiment that the use of multiple sets of NILs would allow the identification of better defined sets of candidate genes underlying the targeted locus. When compared among the three NILs, only two DEGs commonly expressed among all NILs, whereas for down-regulated genes, 40 were commonly expressed. Also a total of 10,141 induced genes (7,650 up- and 2,491 down-regulated) were detected between the two isolines of NIL1 following Fp-infection and mock treatments. Of the 76 commonly expressed genes containing SNPs (SNP-EGs) across the NILs, 73 were mapped to the long arm of chromosome 4H where the targeted QTL resides. Functional annotation of the transcripts indicated that several of these identified DEGs and SNP-EGs were involved in host-pathogen interactions. These results showed that the multi-NIL approach is a powerful tool in transcriptomic analysis and the DEGs and SNPs identified here will improve the accuracy of fine mapping to develop markers for breeding programs