«A. Kolodinska Brantestam Faculty of Landscape Planning, Hordiculture and Agricultural Science Department of Crop Science Alnarp Doctoral thesis ...»
Czembor & Czembor, 2000). For instance, the resistance gene for powdery mildew mlo-11 from an Ethiopian landrace has been incorporated in many accessions (Helms Jørgensen, 1992; Piffanelli et al., 2004). The actual amount of exotic material incorporated into the Nordic and Baltic barley genome is not known. This is due to the fact that adaptation of the cultivar to the requirements of modern agriculture and regional conditions (Tigerstedt, 1994; Ortiz et al., 2002) needs several cycles of backcrossing to advanced lines. For example, crosses with the landrace H. vulgare var. ‘laevigatum’ resulted in good mildew resistance from the Ml-v resistance gene (Swanston, 1987). H.vulgare var. laevigatum and the first commercial cultivars carrying the Ml-v mildew resistance gene, ‘Vada’ and ‘Minerva’ (Swanston, 1987), were also included in breeding programmes in the Nordic and Baltic countries (Fig. 8).
Landrace LN5 Landrace LN9 unnamed Local line Hordeum LOC4 UNK3 LOC4 lower Bavaria l. Bavaria cultivar Scandinavia laevigatum LOC4 UNK3 LOC4 UNK3 UNK3 LOC4 Ackerman Danubia Hanna Gull Hanna Hanna Gull
Fig. 8. Pedigree of the Swedish cultivar ‘Simba’ (1975).
9 This is a rare exception of incorporation of an exotic gene source requiring fewer cycles of backcrossing and with a good adaptation to modern agriculture; i.e. with high yield values. However, the genes from this exotic source are negatively affecting malt quality (Swanston, 1987) so that cultivars with H. vulgare var.
laevigatum in their pedigree are mainly used as feed barley.
The wild relatives of barley are a rich source of disease resistance. H. vulgare ssp. spontaneum is a source of useful resistance genes to leaf rust, net blotch, septoria speckled leaf blotch, powdery mildew, barley mosaic virus, scald, etc.
(Jahoor & Fishbeck, 1993; Garvin et al., 1997; Fetch et al., 2003). Another potential resource is H. bulbosum from the secondary genepool (Fig. 1). This species is a rich source of resistance to important fungal and viral diseases (Milhăilescu & Giura, 2001). However, large phenotypic differences exist between modern cultivars and wild barley, which carries a number of undesirable traits in quality and agronomic performance. The success of favourable character transfer can be obtained by applying a refined system of genetic markers (Forester et al., 1997). Such methods were not available in the past and the use of wild relatives in breeding required and still require expensive and laborious prebreeding processes (Lehmann & Bothmer, 1988; Veteläinen, 1994). This is why the use of wild material is restricted in Nordic and Baltic material.
Selection for malting quality has been carried out in the Nordic and Baltic region since the beginning of barley breeding, because beer brewing is economically important (Trolle, 1957; Persson, 1997). The malting quality of barley is a phenotype representing the net effects of a number of interacting component traits (Marquez-Cedillo et al., 2000a; Zale et al., 2000; Ayoub et al., 2002; Prada et al., 2004). The complexity of malting quality improvement has led breeders to work within narrow gene pools. In Nordic and Baltic malting barley there are only a few established lines used in crosses to produce cultivars with good malting quality. For example, excellent malting quality was recognised in the barleys from Moravian ‘Hanna’ at the end of 19th century (Grausgruber et al., 2002). ‘Binder Abed’, which is a selection from ‘Hanna’ barley, was for many years the main barley cultivar in Denmark (Trolle, 1957) and is found in the pedigrees of numerous later Nordic and Baltic cultivars. In later breeding periods ‘Trumph’ became widely used in the Nordic and Baltic malt barley pedigrees (Fischbeck, 1992).
Feed is another important end-use of barley. However, barley cultivars are often developed and selected on the basis of only agronomic and malting-quality characteristics. There is currently no widely accepted set of criteria for determining barley feed quality as there is for malting quality. Therefore barley breeders have been limited in their ability to develop cultivars improved for feed quality characteristics (Bowman et al., 2001). Recent research has identified high starch content, low acid-detergent fibre (ADF), low ruminal dry-matter digestibility (DMD) and large particle size after dry rolling as desirable barley feed-quality characteristics for beef cattle (Hunt, 1996; Surber et al., 2000;
Bowman et al., 2001).
Traditional barley breeding skills have served the agricultural industry very well over many years and are likely to retain a significant role in the future. However, 10 with the employment of novel technologies, the speed and accuracy of selection are enhanced (Swanston & Ellis, 2002). Today, successful results in breeding are achieved through increased knowledge about the barley genome and its diversity.
The barley genome Barley is a model genome system for Triticeae species, because it has a selffertile, diploid (2n=2x=14) genetic system that has advantages for studies of phenotypic expression of genes and development of homozygous material. The barley chromosomes are homologous to those of cultivated wheat which is polyploid and inbreeding, and of rye which is diploid and outbreeding (Hori et al., 2003). The genome of barley is approximately 5 000 million base pairs (Mbp) in size (Yu et al., 2000). It has been estimated that the average distance between barley genes is 240 kb (based on an estimated genome size of 5,000 Mb and 21000 genes) (Dubcovsky et al., 2001). However, the average gene density in some genome regions has approximately one gene every 20 kb, which is 12 times higher than the expected genome average (Panstruga et al., 1998; Dubcovsky et al., 2001; Druka et al., 2002). A very heterogeneous distribution of recombination rates is found along individual chromosomes. Recombination is mainly confined to a few relatively small areas interspersed with large segments in which recombination is severely suppressed (Künzet et al., 2000). A high correlation between marker density and recombination frequency implies that most recombination events occur in gene rich regions corresponding to small chromosomal areas (Künzel et al., 2000). Panstruga et al. (1998) conclude that grass genomes are characterized by compositional compartmentalization with gene islands (also termed ‘gene space’) of 100-200 kb. The base paire composition of these gene-rich regions is not significantly different from the average genome base pair composition in Triticeae (Dubcovsky et al., 2001).
Most genes are present in physically small gene-rich regions. Some genes are highly repetitive, such as the 18S-5.8S-25S ribosomal RNA (rRNA) genes and intergenic spacer, together called the rDNA, occurring at the nucleolus organizing regions (NORs) loci on the chromosomes. Major sites of rRNA genes involve hundreds or thousands of copies of the tandem repeat unit, which is about 10 kb long (Linde-Laursen et al., 1997). The gene-rich regions are interspersed by large chromosomal blocks mainly containing repetitive sequences (Barakat et al., 1997).
Repetitive DNA comprises 70% of the genome (Barakat et al., 1997). These repetitive DNA-sequence motives, typically 2 to 10 000 bp long, are repeated hundreds or even thousands of times in the genome (Linde-Laursen et al., 1997).
For example, the BARE-1 retrotransposons in barley are present in 16.6 x 103 copies and more than 6 x 104 single long terminal repeats (Vicient et al., 1999).
Dubcovsky and his co-workers (2001) showed that a large difference in size between the rice and barley colinear gene regions was mainly due to the insertion of different layers of retroelements in the intergenic regions. Retrotransposons, because of their abundance, diversity and widespread distribution, make a major contribution to both the shape and size of plant genomes (Vershinin et al., 2002).
Vicient et al. (1999) suggested that the genome increases and genetic polymorphism in dry environments might be adaptive in the genus Hordeum and associated along with either propagation of BARE-1 or inheritance of new copies.
11 In this regard, transcription or transposition of various retrotransposons is linked to genes for biotic and abiotic stress tolerance (Wessler 1996; Takeda et al., 1998).
Modern tools for barley breeding and their implications Molecular maps and chromosome library The use of molecular techniques as diagnostic tools to assist the conventional breeding process demands the construction of linkage maps. Markers evenly spaced on the chromosomes are then used to scan the genome in order to identify associations of markers and traits. Molecular mapping of the barley genome has been facilitated by the development of molecular markers, the ability to develop doubled haploid (DH) lines, the availability of numerous mutants and cytogenetic stocks, particularly the barley-wheat additional lines, and the recent development of large insert libraries. The comprehensive molecular marker linkage maps have provided a powerful and important tool for identifying quantitative trait loci (QTL) for agronomic and qualitative traits in barley (Kleinhofs & Han, 2002).
Molecular linkage maps of cereals are being improved rapidly by adding new types of markers, by merging different species-specific maps and by comparative mapping of markers between related genomes. Efficient use of the resulting dense maps, therefore, requires detailed insights into the relationship between genetic and physical distances (Künzel et al., 2000).
Molecular marker linkage maps have been developed using restriction fragment length polymorphism (RFLP), amplified fragment length polymorphism (AFLP), simple sequence repeat (SSR), isozyme and protein markers (Kleinhofs et al., 1993; Becker et al. 1995; Langridge et al., 1995; Sherman et al., 1995; Waugh et al., 1997). All these mapping efforts have resulted in the location of several thousands of different molecular markers to the barley genome. These data have been used to merge several maps (Sherman et al., 1995; Kleinhofs & Han, 2002).
All barley maps appear to be co-linear and easy to integrate with only minor differences in genetic distances between markers. In addition, a barley morphological-marker linkage map containing over 200 loci has been constructed (Franckowiak, 1997) and is integrated with the molecular marker maps (Qi et al., 1996; Kleinhofs & Graner, 2001). With a more extensive effort to merge mapping information from different mapping populations, the North American Barley Genome Mapping Project (NABGMP) introduced the BIN Map concept (Kleinhofs & Graner, 2001). Using the ‘Steptoe’ x ‘Morex’ (SM) map as a base, the barley genome was divided into approximately 10 cM intervals or Bins, allowing the placement of many markers on different maps in their appropriate Bins. The map integrated RFLP, AFLP and SSR markers, which are mapped in independent linkage studies, by allocating them to 99 evenly spaced BIN groups.
However, the average resolution on the physical distance by 1 172 markers in the barley genome is about 4 500 kb/marker in the present map. There is still a large gap between the physical distance and the resolution of the high-density map in barley. Constructing high-density molecular maps in barley has revealed uneven distributions of markers and a strong clustering of markers around the centromere (Becker et al., 1995; Qi et al., 1998; Ramsay et al., 2000). Clustering of markers at centromeric regions was also observed on the barley integrated map (Qi et al., 1996).
12 A bacterial artificial chromosome (BAC) library has been constructed for the cultivar ‘Morex’ using the cloning enzyme HindIII (Yu et al., 2000). The library contains 313 344 clones with an average insert size of 106 kbp. BAC libraries are useful for examining genomic structure and positional cloning of genes (Song et al., 1995). Moreover, the barley BAC library is an important tool for developing physical maps of the gene-rich regions (Yu et al., 2000). High-resolution maps have been prepared for several barley genome regions (Kleinhofs & Han, 2002), for example lig, mlo, wx, Rpg1 and Mla (Jorgensen & Jensen, 1979, Rosichan et al.; 1979; Konishi, 1981; DeScenzo et al., 1994; Kilian et al., 1997; Simons et al., 1997).
Quantitative trait loci and marker assisted selection Mapping and marker development have progressed in recent years for both barley and two related Hordeum species. Establishment of synthetic relationships among major grass species, especially at the micro level, has made possible ‘map based cloning’ of genes from a large genome such as barley by utilising the results of genetic and physical mapping of the small genomes (Kleinhofs & Han, 2002).
A goal of the molecular mapping activity in barley is to locate quantitative trait loci (QTLs) for map-based cloning or to find associated markers for molecular marker-assisted selection (MAS) to supplement ongoing breeding programmes (Hoffman & Dahleen, 2002). The agronomic performance of crop varieties is mainly influenced by complex quantitative traits, for example, yield and quality components (Pillen et al., 2003). QTLs are often clustered and a discriminating marker could be located in a region associated with a number of traits. For example, the AFLP marker E36M47.M162 on chromosome 5 is situated in a region associated with plant height, heading data, malt extract, grain protein, diastatic power and the Kohlbach index (Hoffman & Dahleen, 2002). QTL analysis identifies chromosome regions, linked molecular markers, gene effects and QTL x E (environment) and QTL x QTL interactions for a given trait (Zale et al., 2000).
QTL mapping in barley has received world-wide attention. The North American Barley Genome Mapping Project has focussed on ‘Steptoe’/‘Morex’, ‘Harrington’/TR306, ‘Harrington’/’Morex’ and some other populations. European researchers have studied the ‘Blenheim’/E224/ 3 population and Australian researchers have concentrated on the ‘Chebec’/‘Harrington’, ‘Clipper’/‘Sahara’ and ‘Galleon’/‘Haruna Nijo’ mapping populations (Zale et al., 2000). More than 750 QTLs are identified (Hayes at al., 2003).
An important and practical question is whether QTLs are conserved among genotypes of the same species. This question is especially important for MAS (Clancy et al., 2003). Traditionally, researchers have looked at single genes and QTLs for disease resistance, morphological markers and many agronomic traits.