The goal of this project was to exploit ancient Bere barley gene pools to improve the ability of barley to acquire and utilize nutrients from the soil more efficiently:
1. Identify Mn efficiency candidate regions by exome capture sequencing using phenotypic bulks of a segregating landrace x elite F2 inter-cross population.
2. Discover transcriptional and molecular responses, which are regulated under Mn deficiency and recovery conditions in the barley landraces.
3. Characterise and quantify the physiological effects of downstream Mn-responsive genes and molecular factors important for Mn acquisition and allocation in crosses, landraces and elite cultivars grown in marginal soil.
Results
Despite COVID restrictions, limited access to the Institute and the termination of the project after 12 months, significant and valuable contributions to research in this area were achieved, with many of the defined tasks and deliverables completed.
WP1 Manganese efficiency candidate region was identified on 6H by classical genetic mapping of an F2 segregating population and confirmed by bulk segregant analysis (Task 1.1 and D1)
Task 1.1 was built on previous field trials, conducted during the fellow’s recent research visit at JHI. Because of this striking phenotypic contrast, crosses between these landraces and elite cultivars were developed. To identify potential candidate regions, two approaches were pursued:
1) Bi-parental F2 mapping
Over 1,200 F2 individual plants were phenotyped using non-destructive chlorophyll a fluorescence as a proxy for manganese efficiency. Leaf samples were harvested for both DNA extraction and inductively coupled plasma-mass spectrometry (ICP-MS) at the University of Copenhagen. Furthermore, spikes from each F2 plant were harvested and stored for fine scale mapping.
DNA was extracted from a subset of 288 F2 leaf samples that had been stored and genotyped using the barley 50K SNP assay. From the 44,000 gene-based SNPs assayed, 16,340 were polymorphic in the population of F2 lines, and these were genetically mapped. Combining the phenotypic data, both fluorescence (Fv/Fm) and ICP data (Mn), a QTL analysis performed in GenStat ((VSN International (2021)) which identified a significant association on 6H, spanning 36.5 cM to 87.2 cM, with a peak candidate at 64.4cM (Figure 1).
2) Bulked Segregant Analysis (BSA) based on phenotypically contrasting pools.
From the Fv/Fm and the ICP data, we identified pools of 24 ‘low’ and 24 ‘high’ lines to use in the pooling strategy. Genomic DNA was extracted from each line and equal amounts combined into 3 x 8 subsets for high and low phenotypic values. The 6 pooled DNA samples were sent for whole genome sequencing (Novogene Europe). Data was first quality checked and mapped to the reference barley MorexV3 assembly. Variant calling for high versus low pools over the 6 samples identified over 19 million SNPs. Following stringent filtering, alleles with frequency differences (AFD) between the pools were identified ranging from 209,468 at the lowest quality (QUAL 100) and lower AFD (0.6) to only a single SNP at highest quality (800) and highest AFD of 1.0. In order to identify potential candidate SNPs, this data was plotted along chromosomes to determine specific locations. Figure 2 clearly highlights a region on 6H which is differentiating between contrasting pools. This region was identified in a similar location to the region identified in the F2 segregating mapping with the 50K SNPs. With a high quality and an AFD of 0.8 84 SNPs were identified physically between 351-361 Mbp with a sharp peak at 358 Mbp.
WP2: Capture and validation of the transcriptome
A microarray experiment was performed using hydroponics-grown barley seedlings, examining the global gene expression response to varying manganese levels (10 nM, 100 nM & 1,000 nM) in Bere compared to control KWS Irina. Total RNA was extracted from 48 tissue (leaf & root) samples (2 genotypes x 3 Mn treatments x 2 tissues x 4 biological replicates) and hybridised to Agilent barley 60k microarrays. Samples were processed with established procedures using in-house facilities at the Hutton. Data quality analysis revealed that several of the biological replicates were outliers and removal of these replicate samples meant that full balanced statistical analysis of the data was not possible, limiting the value of the dataset. However, gene expression profiles of selected candidate genes identified from previous experiments were examined.
WP3: Characterisation of crosses generated, and haplotypes identified in WP1
Due to both COVID restrictions at the Institute and the termination of the fellowship, characterisation planned in Task 3.1 which was due to commence in month 3 of the project did not start. Task 3.2 was scheduled for later in the project and therefore was not performed.
WP4: Obtaining Transferable skills and Dissemination & Communication activities
During the 12 months of the project, regular management and planning meetings were held, using video conferencing and when permitted face to face, with colleagues at the James Hutton Institute. An online R course provide by BioSS was attended and used to analyse the data generated in WP1. Throughout this period, skills in both genetic mapping and QTL analysis were attained and evidenced by the identification of candidate QTL for manganese efficiency. Furthermore, with the generation of the WGS data, bioinformatics skills have been increased and a greater understanding of the type of data and analysis. As part of Task 4.2 a perspective manuscript title ‘Heritage genetics: Signatures of local adaptation of barley landraces to marginal soils’ has been drafted and is in the process of internal review and will be submitted before the end of December 2021
