Transcriptome analysis of a barley breeding program examines gene expression diversity and reveals target genes for malting quality improvement

BMC Genomics Transcriptome analysis of a barley breeding program examines gene expression diversity and reveals target genes for malting quality improvement Mara Muoz-Amatrian 0 Yanwen Xiong 0 Mark R Schmitt Hatice Bilgic 0 Allen D Budde Shiaoman Chao Kevin P Smith 0 Gary J Muehlbauer 0 0 Department of Agronomy and Plant Genetics, University of Minnesota , St. Paul, MN 55108 , USA Background: Advanced cycle breeding utilizes crosses among elite lines and is a successful method to develop new inbreds. However, it results in a reduction in genetic diversity within the breeding population. The development of malting barley varieties requires the adherence to a narrow malting quality profile and thus the use of advanced cycle breeding strategies. Although attention has been focused on diversity in gene expression and its association with genetic diversity, there are no studies performed in a single breeding program examining the implications that consecutive cycles of breeding have on gene expression variation and identifying the variability still available for future improvement. Results: Fifteen lines representing the historically important six-rowed malting barley breeding program of the University of Minnesota were genotyped with 1,524 SNPs, phenotypically examined for six malting quality traits, and analyzed for transcript accumulation during germination using the Barley1 GeneChip array. Significant correlation was detected between genetic and transcript-level variation. We observed a reduction in both genetic and gene expression diversity through the breeding process, although the expression of many genes have not been fixed. A high number of quality-related genes whose expression was fixed during the breeding process was identified, indicating that much of the diversity reduction was associated with the improvement of the complex phenotype malting quality, the main goal of the University of Minnesota breeding program. We also identified 49 differentially expressed genes between the most recent lines of the program that were correlated with one or more of the six primary malting quality traits. These genes constitute potential targets for the improvement of malting quality within the breeding program. Conclusions: The present study shows the repercussion of advanced cycle breeding on gene expression diversity within an important barley breeding program. A reduction in gene expression diversity was detected, although there is diversity still present after forty years of breeding that can exploited for future crop improvement. In addition, the identification of candidate genes for enhancing malting quality may be used to optimize the selection of targets for further improvements in this economically important phenotype. - Background Genetic diversity within breeding populations is indispensable for obtaining genetic gains, and consequently for plant breeding progress. Plant breeding that involves crossing elite lines in a closed population is called advanced cycle breeding [1] and it has proved to be successful in achieving genetic gains in major crops such as barley (Hordeum vulgare L.), maize (Zea mays L.), rice (Oryza sativa L.), soybean (Glycine max L. Merr.) and wheat (Triticum aestivum L.) ([2] and references therein). However, over cycles of selection, the genetic variability within breeding populations is reduced, presumably reducing the potential for future gains and increasing genetic vulnerability [3-7]. Therefore, an evaluation of an ongoing breeding program is necessary to gain an understanding of the existing diversity and optimize current and future improvements. At present, few studies have evaluated the variation in genetic diversity within a single breeding program, with examples including barley [3] and wheat [8]. The development of new barley varieties with improved malting quality characteristics is one of the primary aims of the US barley breeding programs due to the economic impact of the malting and brewing sector. Malting quality is a genetically complex phenotype representing a set of component traits, many of which are interrelated [9-11]. Grain protein content, malt extract percentage, ratio of wort soluble protein to total malt protein, diastatic power, a-amylase activity, and wort b-glucan content are some of the most important parameters contributing to malting quality. Qualityrelated QTL reported in the literature have been recently summarized, resulting in 154 QTL associated with 18 quality traits that are located on all barley chromosomes [12], which reveals the genetic complexity of this phenotype. In the U.S., the malting and brewing industry require that new cultivars meet quality parameters specified by maltsters and brewers that generally agree with the ideal commercial malt criteria established by the American Malting Barley Association (AMBA) http://www.ambainc.org/media/AMBA_PDFs/Press_Releases/GUIDELINES.pdf. This, together with the complexity of the phenotype and the high cost of malting quality evaluation, has encouraged breeders to follow a conservative strategy, favoring crosses among closely related elite cultivars with good quality characteristics in order to maintain acceptable malting performance [13,14]. The University of Minnesota six-rowed malting breeding program has used advanced cycle breeding since 1958, developing important malting cultivars like the variety Morex, an archetype for malting quality, and the AMBA recommended six-rowed malting cultivars Robust, Lacey and the recently released Rasmusson http://www.ambainc.org/media/AMBA_PDFs/Pubs/ KYMBV_2010.pdf. This closed population offers an excellent opportunity to study the effect of plant breeding on genetic gain, genetic diversity and phenotypic variation. Condn et al. [3] examined the effect of advanced cycle breeding on allelic diversity and showed a reduction in the number of alleles per locus, from an average of 5.89 to 2.34. This reduction was not uniform across the genome, predictably due to selection pressure on disease-resistance and quality traits. However, only 28% of the total loci studied had been fixed, indicating that there is still genetic variability in the University of Minnesota elite germplasm that can be exploited. Genetic gains during advanced cycle breeding were documented by the same group for most of the 15 agronomic and malting quality traits evaluated [13]. Of the seven traits whose phenotypic variance changed over the four decades of advanced cycle breeding, five showed a significant decrease. Both studies reported that the breeding process generated a germplasm differentiation between the most recent genotypes and their ancestors. Recently, increased attention has been paid to the influence of gene expression differences on phenotypic variation. In Arabidopsis, the Affymetrix ATH1 GeneChip was used to analyze the gene expression diversity between seven pairs of accessions [15]. This study showed that 10-30% of the Arabid (...truncated)