Everything in Its Place

The use of “model” species in biological research is based on the assumption that many of their features are shared among a wide range of related taxa. Thus, it is hoped that many of the genes associated with important traits in crop plants will be identified via homology with their counterparts in Arabidopsis. In addition to a high degree of conservation of individual gene sequences throughout the plant kingdom, comparative genomics has revealed a high degree of conservation in genome structure, or synteny, among closely related taxonomic groups. Synteny, from the Greek syn (together with) and taenia (ribbon), refers to loci contained within the same chromosome. In comparative genomics, it is often used as a synonym for colinearity (which is more properly conserved synteny) and refers to some degree of conservation of gene content, order, and orientation between chromosomes of different species or between nonhomologous chromosomes of a single species.

For example, gene content appears to be highly conserved with a remarkable degree of colinearity among the grasses, including the grain crops rice, wheat, maize, barley, sorghum, and millet. This is despite large differences in genome size (e.g., 430 Mbp in rice compared with 16,000 Mbp in wheat), which appear to be attributable principally to differences in the amounts of repetitive DNA (associated mainly with retroelements) in intergenic regions and polyploidy (Bennetzen and Freeling, 1997). Thus, rice appears to be an excellent model for the grasses (Chen et al., 1997; Devos and Gale, 2000).

Macrosynteny can be explored using genetic resources such as restriction fragment length polymorphism maps. Restriction fragment length polymorphism markers derived from one species are hybridized against genomic DNA from one or more related species to create a comparative map. Conserved macrosynteny can be observed over large genetic distances on linkage maps of closely related taxa, such as species or genera within the same family. Typically, macrosynteny can be explored only among relatively closely related species because of the limits of cross-hybridization of markers. However, the distance between two markers on a genetic map can comprise hundreds of genes, and macrosynteny between two species does not necessarily imply the existence of microsynteny or the conservation of local gene repertoire, order, and orientation. For example, Tarchini et al. (2000) found interruptions in microsynteny between rice and maize and cautioned that the use of rice as a model system for other cereals may be complicated by the presence of rapidly evolving gene families and microtranslocations. Conversely, conserved microsynteny is possible between species that lack obvious signs of macrosynteny.

The investigation of microsynteny requires sequencing and annotation of genomic DNA, enabling direct comparison of the sequences using various computational tools. Thus, the completed Arabidopsis genome sequence and growing lists of genomic resources for other plants have been an incredible boon to comparative genomics research.

Arabidopsis exhibits extensive conserved synteny with species from the closely related genera Brassica (O'Neill and Bancroft, 2000) and Capsella (Acarkan et al., 2000) (all three are in the family Brassicaceae). Comparisons of Arabidopsis with more distantly related species also have shown some degree of synteny. For example, significant synteny was reported between soybean (Leguminosae) and Arabidopsis along the entire length of Arabidopsis chromosome 1 and soybean linkage group A2, and blocks of synteny were found among other chromosomes as well (Grant et al., 2000). Soybean and Arabidopsis are estimated to have diverged from a common ancestor 92 million years ago (MYA), compared with 6 to 10 MYA for Arabidopsis and Capsella and 12 to 20 MYA for Arabidopsis/Capsella and Brassica.

Comparisons of Arabidopsis and rice have revealed some degree of synteny spanning the divide between monocotyledonous and dicotyledonous plants (Paterson et al., 1996). However, microcolinearity between Arabidopsis and rice appears to have eroded to the extent that the Arabidopsis genome will be of limited use for map-based gene prediction and isolation in the grasses (Devos et al., 1999). Due to its relatively small genome size, rice has become the model species for the grasses. The sequence of the rice genome was completed recently by two private companies (Syngenta [Basel, Switzerland] and Myriad Genetics [Salt Lake City, UT]) and is on the road to completion for public access (Eckardt, 2000). It will be interesting to see the extent of conserved synteny between rice and other monocot families.

Arabidopsis and tomato (Solanaceae) diverged from a common ancestor an estimated 112 to 156 MYA, which follows closely the divergence of dicotyledonous from monocotyledonous families, estimated at 130 to 200 MYA. Thus, comparisons of Arabidopsis and tomato should offer a snapshot of evolution since the introduction of dicotyledonous plants and provide information relevant to a wide range of families encompassed within the Arabidopsis-tomato clade, which includes legumes, Curcurbitaceae (melons), Rutaceae (citrus), Salicaceae (poplar), Malvaceae (cotton), Rosaceae, Asteraceae, and others (Ku et al., 2000).

Ku et al. (2000) found an extensive network of synteny between a 105-kb segment of tomato chromosome 2 and Arabidopsis chromosomes 2 to 5. They concluded that the dominating factors in the divergence of genome organization between these species have been repeated rounds of large scale genome duplication followed by selective gene loss. These authors predicted that complicated networks of conserved synteny would be common among higher plant families.

In this issue of The Plant Cell, Rossberg et al. (pages 979–988) report a remarkable degree of conserved microsynteny between Arabidopsis and tomato based on the sequencing and annotation of a 57-kb region of tomato chromosome 7 and a comparison of this region with the Arabidopsis genome. The tomato sequence chosen included the coding regions for five genes, which were used to search the Arabidopsis genome for similar sequences. Homologous sequences for all five of the genes were found within a 30-kb region corresponding to Arabidopsis chromosome 1. The intergenic regions were greatly expanded in tomato, and three of the five genes appeared in inverted orientation relative to Arabidopsis. The 30-kb region on chromosome 1 was found to be part of a larger segment that shows evidence of duplication on chromosome 3 (Blanc et al., 2000; The Arabidopsis Genome Initiative, 2000). However, only one of the seven genes in the 30-kb region (gene C) had a counterpart in the duplicated region on chromosome 3. These data are consistent with the hypothesis that the region of Arabidopsis chromosome 1 that retains synteny with tomato chromosome 7 underwent duplication followed by extensive gene loss within the Arabidopsis genome after divergence from the common ancestor with tomato.

Rossberg et al. also investigated Capsella for synteny with the 30-kb region of Arabidopsis chromosome 1 and identified a 27-kb contiguous overlapping sequence (contig) from eight cosmid clones that exhibited almost complete microcolinearity with the Arabidopsis sequence. Capsella homologs were found in this contig for all seven Arabidopsis genes located in the 30-kb region, all seven genes were in the same order and orientation, and the intergenic regions were of similar sizes. Interestingly, gene C from Arabidopsis chromosome 1 had greater similarity to Capsella gene C than to the homologous gene on Arabidopsis chromosome 3. Hybridization experiments suggested that Capsella also contains another region of homology that may be analogous to the region on Arabidopsis chromosome 3, indicating that duplication of the region that contains gene C most likely occurred before the divergence of the Arabidopsis and Capsella genera.

The widespread occurrence of gene duplication and the consequent proliferation of large gene families in plants leads to difficulties in determining orthology between species. Orthologous genes, or orthologs, are genes from different species that are derived from a common ancestor, whereas paralogs are genes within a species that arose from a duplication event. Conserved microsynteny is indicative of orthologs, although Bennetzen (2000) cautioned that microcolinearity can be deceiving in this respect because gene duplications often occur over short distances and can be followed by loss of the original parent gene. Thus, orthology can be defined unambiguously only if a high degree of colinearity is observed in the flanking regions of putative orthologs. These distinctions are important not only for establishing evolutionary histories of plant taxa but also for making assumptions about gene function. A high degree of gene sequence homology between individual genes from different species together with conserved microsynteny suggests true functional orthology, whereas a lack of colinearity (even if associated with a highly conserved individual coding region) is more likely to be indicative of a divergence of gene function and/or a rapidly evolving gene family. For example, Rossberg et al. found that one of the five genes on tomato chromosome 7, gene D, which has similarity to WRKY transcription factors, has numerous homologs in other regions of the Arabidopsis genome that were not found to be colinear with the segment identified on Arabidopsis chromosome 1. Thus, analyses of microsynteny also may help to characterize rapidly evolving gene families.

Analyses of many different eukaryotic genomes have revealed that polyploidization and gene duplication are widespread phenomena among eukaryotes. A number of simple diploid genomes, including those of yeast and Arabidopsis, appear to be derived from ancient polyploids. Numerous analyses of the Arabidopsis genome indicate that extensive “genome shuffling” has occurred, characterized by several rounds of large scale duplication followed by gene loss (Blanc et al., 2000; Grant et al., 2000; Ku et al., 2000). The work of Rossberg et al. provides additional evidence for this hypothesis. In fact, it has been suggested that all eukaryotes are derived from ancient polyploids, which had a tendency to evolve to a diploid state through sequence diversification and chromosomal rearrangement (Grant et al., 2000). Polyploidy is particularly widespread among extant plant species relative to other eukaryotes. Comparative genomics may begin to provide answers regarding the evolution and maintenance of the various states of ploidy observed among plants and other eukaryotes.

The work by Rossberg et al. fills an important gap in our knowledge of plant comparative genomics by demonstrating a remarkable degree of conserved microsynteny between distantly related (Arabidopsis and tomato) as well as closely related (Arabidopsis and C. rubella) dicotyledonous species. This study, together with that of Ku et al. (2000), suggests that segments of microcolinearity can be exploited to identify orthologous genes in Arabidopsis and distantly related dicots.