Cereal genome organisation

Rice as a model for grass and cereal genetics

In the early 1990’s, Arabidopsis was being proposed as the main model for plant science, but at the same time, we proposed and then demonstrated that gene order was conserved between rice and wheat despite their genomes differing greatly in size (Moore et al., 1993; Moore et al., 1994;).  Other cereal genomes were examined, and conservation of gene order (synteny) was found at the genetic and physical level (Moore et al., 1994; Dunford et al., 1995; Foote et al., 1997; Griffiths et al., 2006).  The rice genome is one of the smallest among the grasses and cereals, and in 1995, we demonstrated that rice could be an alternative model for cereals based on this “synteny” because its genome can be divided into groups of genes - a series of genomic building blocks - from which the other larger cereal genomes can be constructed. This analysis exploited mapping data provided by Professor Mike Gale FRS as well as from groups in France, US and Japan.  Comparison of the order of blocks within the different cereal chromosomes revealed that each cereal genome can be derived from the cleavage of a single structure, a hypothetical ‘ancestral’ genome, from which the genomes of present day cereals and grasses have evolved.

‘Lego block’ model of cereal genome organization (view larger image)

 A) The12 chromosomes of the rice genome are dissected into linkage blocks.
B-F) The rice blocks are rearranged to construct the genomes of five economically important grass species. 
G) A hypothetical ancestral chromosome is reconstructed on the basis of the rice blocks.

Comparative mapping of grass and cereal genomes

With Professor Mike Gale FRS, we then developed a circularised consensus map of the genomes of six major grasses (Moore et al., 1995), in which the chromosomes of the six grass species are aligned such that any radius passes through homoeologous (orthologous) genes.  This allows us to exploit the small genome of rice and the extensive resources now available for this model genome, to isolate genes in other cereals by map-based cloning in rice then by homology in the target species.  The “crop circles” diagram shows at a glance the chromosomal region in which a target gene is likely to map. 

‘Crop circles’: Alignment of the genomes of six major grass crop species. A radius taken through a particular gene (eg Ss1 in maize) can be used to predict the position of the orthologous genes in other cereal species.

Molecular characterization of the Ph1 locus

In the early 1990’s it was generally thought that the wheat genome was intractable to analysis due to its size (5-fold larger than the human genome) and complexity (80% of the genome being repetitive), but comparative mapping provided a strategy to clone genes for important traits from the wheat genome, one of which was the Ph1 locus.  The Ph1 locus prevents pairing and recombination of chromosomes that are not true homologues.  This failure in chromosome exchange (introgression) prevents breeders from exploiting the diversity of wild relatives within their breeding programmes, so the ability to turn Ph1 off and on would be of major importance to wheat breeding.

The molecular characterization of the Ph1 locus was achieved using a two-part strategy. The first part of the strategy used wheat deletion lines to physically dissect the Ph1 locus.  Fast-neutron irradiation produced five deletions of the Ph1 locus (Roberts et al., 1999), and the breakpoints for these deletions delimited Ph1 to a location equivalent to a 140kb section of the rice genome. 

The second part of the strategy utilized and extended the comparative mapping of cereal genomes as described above.  The rice genome was exploited to provide markers within the wheat Ph1 region, but additional markers were obtained from a closer relative of wheat, the temperate grass Brachypodium sylvaticum, which we proposed as a potential model for temperate cereals.  Like rice, Brachypodium possesses a small genome and the gene order is highly conserved with that of the wheat genome (Foote et al., 2004; Griffiths et al., 2006).   A Brachypodium BAC library was constructed to facilitate the comparative mapping process (Foote et al., 2004).  Markers generated from the rice and Brachypodium comparative mapping were used to screen a hexaploid wheat BAC library constructed in collaboration with INRA (France) (Allouis et al., 2003), which generated a BAC contig map of the Ph1 region.  BACs forming a minimum tiling path of the region were shotgun sequenced and thirty-six genes were identified.

 

Hexaploid wheat BAC contig map of the Ph1 region and its equivalent region in rice and Brachypodium sylvaticum (Griffiths et al, 2006) - click to view larger image

Ph1 defined to a cluster of Cdk-like genes

Following the molecular characterization of the Ph1 locus, a further five new deletion mutants covering the region were identified, and additional BACs were sequenced and analysed to elucidate the complexity of this locus.  This further analysis, in combination with expression profiling of genes in the region, revealed that Ph1 consists of a multigene family involved in cell cycle control, disrupted by a piece of sub-telomeric heterochromatin.  This multigene family consists of a cluster of Cdk-like genes, which show close homology to Cdk2 in humans and mice, which is involved in preventing non-homologous chromosome pairing at meiosis (Griffiths et al., 2006; Al-Kaff et al., 2007).

 

 

 

Schematic diagram of the deletion mutants and annotated genes in the region containing the Ph1 locus on chromosome 5B compared with the equivalent regions on chromosomes 5A and 5D (Al-Kaff et al, 2007).

Implications for breeding programmes

The discovery that the Ph1 locus contains specific cell-cycle genes suggests that certain chemicals that affect the cell cycle might be used as a temporary off-switch for Ph1. (See chemical modification of chromosome pairing). This would allow efficient introgression of useful genes from wild grasses, without permanently destabilising the polyploid genome. Such an innovation could potentially impact well beyond wheat to other crops.

Other genomic strategies developed

During the course of the Ph1 project, we also developed the first plant chromosome specific library from flow sorted chromosomes (Wang et al., 1992) and cDNA-AFLP display (Money et al., 1996).

John Innes Centre, Norwich Research Park, Colney, Norwich NR4 7UH, UK

Top of Page