Barley Inflorescence Architecture

Abstract

Cultivated barley, Hordeum vulgare ssp. vulgare, is the fourth most abundantly grown cereal in the world (www.fao.org/faostat) and is long associated with human civilisations. Although most barley grain grown today is destined for animal feed and malting, barley remains an important source of primary calories in many parts of the world. Increasing barley yield in the face of challenges posed by increasing world population and climate change is a major goal of current research efforts. Grain is the ultimate product of inflorescence development and maturation. As such, understanding the genetics underlying inflorescence architecture in barley and then learning how to apply this knowledge to manipulate inflorescence development are important steps towards improving yield. The barley reference genome sequence represents an invaluable resource to support the identification and functional characterization of genes controlling inflorescence architecture. Resolving the relationships between gene and inflorescence traits are critical to support breeding as well as to provide insight about fundamental questions in cereal developmental biology. In this chapter, we first provide an overview of inflorescence development in cereals, highlighting the transitions in meristem identity associated with species-specific architectures. From here, we describe the development of key morphological features associated with the barley spike, spikelet, floret and grain, while discussing the identification and functions of genes which regulate their development. We also discuss those genes whose variation contributed to architectural changes during domestication and those with yield potential. Lastly, we describe environmental control of inflorescence development, with special attention to flowering time and the agronomic environment.

Appendix

Genetic and Genomic Resources for Gene Cloning and Functional Analysis

A robust genetic map is the starting point for a good physical map, which in turn is crucial for identification of physical contigs harboring genes of interest in positional cloning studies. Over the past two decades, many barley genetic linkage maps became available based on different marker systems, including RFLPs (Graner et al. 1991; Kleinhofs et al. 1993), SSRs (Varshney et al. 2007), DArTs (Wenzl et al. 2006), ESTs (Potokina et al. 2008; Stein et al. 2007), SNPs (Close et al. 2009; Rostoks et al. 2005), and a combination of different marker systems (Hearnden et al. 2007; Marcel et al. 2007). A major advance was the development of the barley genome zippers by positioning the gene-based barley 454 sequence reads in synteny with the genes of rice, Brachypodium, and Sorghum (Mayer et al. 2011). These syntenic 454 reads were positioned in between the gene-based SNP markers reported in the consensus map developed by Close et al. (2009). A dense genetic map-based on Morex × Barke RIL population (3973 SNP markers) was developed by Comadran et al. (2012) and (IBSC 2012). Mascher et al. (2013a) further refined the Morex × Barke RIL map by introducing a strategy based on sequencing progenies of Morex × Barke of population (POPSEQ) that allowed de novo production of a genetically anchored linear assembly (an ultra dense genetic map with 4.3 million SNPs). Ariyadasa et al. (2014) developed a genome-wide physical map of barley by map-based sequencing (4.9 Gb representing 96% of the physical length). A draft genome sequence of barley was made available by the International Barley Sequencing Consortium (IBSC 2012). Further A high-quality barley reference genome assembly was developed by chromosome conformation capture mapping (HiC) to derive the linear order of sequences across the pericentromeric regions. Such resources greatly facilitate high-resolution trait mapping, gene isolation, and comparative genome or transcriptome studies. The availability of gold standard reference genome, physical maps and marker systems paved the way for several NGS-based fast-forward genetic analysis and gene cloning strategies in barley such as mapping-by-sequencing (Liu et al. 2014; Mascher et al. 2014) exome capture (Mascher et al. 2013b; Russell et al. 2016), and MutChromSeq (Sánchez-Martín et al. 2016). Gene cloning was also accelerated due to robust genetic analyses such as Genome-Wide Association scans (GWAS) benefitting from variation across germplasm (Comadran et al. 2012; Houston et al. 2015; Houston et al. 2013; Ramsay et al. 2011; Youssef et al. 2017).

Gene expression patterns can often reveal insight about possible gene function and links to mutant phenotypes. A high-quality RNA seq data set from 16 different tissues (including six inflorescence tissues) is available for all genes in barley and the expression data can be visualized through the BARLEX database (Colmsee et al. 2015). Apart from this, developmental stage (from SAM until AP) and position specific (LSM, CSM, and IM) reference transcriptome data has been generated from Bowman (Schnurbusch et al., Unpublished), which will provide the expression dynamics of inflorescence developmental genes across time and space.

Once the gene underlying a phenotype is identified, the further functional analysis is essential to substantiate the gene function. Towards this measure, barley has a rich collection of chemical and ionizing ration induced mutants defective in their inflorescence development. All of these mutants are genetically characterized (Lundqvist 2009; Scholz and Lehmann 1961) and have been systematically maintained in geneticist and breeder collections at the National Small Grain Collection, Aberdeen, ID (http://www.ars.usda.gov/Main/docs.htm?docid=2922), IPK Genebank and the Nordic Genetic Resource Center, Alnarp, Sweden (https://www.nordgen.org/bgs/index.php). Several of these mutants have been backcrossed to (BW001 to BW979) a near-isogenic background in the genotype Bowman (Franckowiak et al. 1985) making their functional evaluation less cumbersome. Druka et al. (2011) performed SNP genotyping of all Bowman backcrossed mutants, which delimited the genomic introgression responsible for the mutant phenotype. The historical mutant resources together with SNP introgression data would be a wonderful starting point for the functional studies and positional isolation of responsible gene/s in these mutants.

Most of the historical barley developmental mutants identified in the past have more than one mutant allele by which the underlying gene function can be validated. However, for genes being isolated by conventional bi-parental genetic analyses or GWAS, and for those mutants lacking enough number of mutant alleles, it is necessary to validate identified gene candidates using reverse genetic approaches such as TILLING (Targeting Induced Local Lesions IN Genomes) or transgenics. As of today, four different TILLING populations are available for barley. These include (i) TILLMore, a sodium azide induced six-rowed TILLING population from Morex (Talamè et al. 2008). (ii) Two-rowed EMS (Ethyl Methyl Sulfonate) induced Barke TILLING population (Gottwald et al. 2009). (iii) Two-rowed EMS-induced Optic TILLING population (Caldwell et al. 2004) The other unpublished resources include EMS-induced TILLING lines from two-rowed cultivar Haruna Nijo (Sato et al. unpublished). These reverse genetic tools shall enable functional evaluation of genes in mutants of interest.

Testing genes of interest by transgenics is a common functional genetics approach (Hensel et al. 2011). Several transgenic approaches were established for barley by which a preferred gene can be complemented, downregulated, knocked-out or over-expressed. Complementation is a straight forward approach by which a mutated gene can be complemented by the functional wild-type form of the gene. The quantitative effect of a gene can be analyzed through approaches like RNA interference (RNAi), Virus-Induced Gene Silencing (VIGS) (Hein et al. 2005; Holzberg et al. 2002), which function through homology-dependant post-transcriptional gene silencing. However, these approaches suffer from potential off-target effects (gene family members). This problem can be circumvented through a recent technology called Transcriptional Activator-Like Effector Nucleases (TALENs), which allows sequence specific in planta gene targeting with typically no off-target effects in the genetic background of the host genome (Boch et al. 2009; Joung and Sander 2013; Zhang et al. 2013). This technology has already been proven successful in barley (Gurushidze et al. 2014; Wendt et al. 2013). Recently, the RNA-guided Cas9 system (CRISPR/Cas) has been proven successful to knockout genes in barley (Lawrenson et al. 2015). The comprehensive set of available genetic and genomic resources discussed will be useful to isolate and characterize genetic determinants underlying various developmental processes in barley.