1709American Journal of Botany 91(10): 1709–1725. 2004.THE EVOLUTION OF NUCLEAR GENOME STRUCTURE INSEED PLANTS1ELIZABETHA. KELLOGG2,4ANDJEFFREYL. BENNETZEN32Department of Biology, University of Missouri–St. Louis, One University Boulevard, St. Louis, Missouri 63121 USA; and3Department of Genetics, University of Georgia, Athens, Georgia 30622 USAPlant nuclear genomes exhibit extensive structural variation in size, chromosome number, number and arrangement of genes, andnumber of genome copies per nucleus. This variation is the outcome of a set of highly active processes, including gene duplicationand deletion, chromosomal duplication followed by gene loss, amplification of retrotransposons separating genes, and genome rear-rangement, the latter often following hybridization and/or polyploidy. While these changes occur continuously, it is not surprising thatsome of them should be fixed evolutionarily and come to mark major clades. Large-scale duplications pre-date the radiation ofBrassicaceae and Poaceae and correlate with the origin of many smaller clades as well. Nuclear genomes are largely colinear amongclosely related species, but more rearrangements are observed with increasing phylogenetic distance; however, the correlation betweenamount of rearrangement and time since divergence is not perfect. By changing patterns of gene expression and triggering genomerearrangements, novel combinations of genomes (hybrids) may be a driving force in evolution.Key words: duplication; polyploidy; retrotransposon.Plant nuclear genomes are enormously variable. Chromo-some number, the degree of gene clustering, and chromosomesize can all differ by as much as an order of magnitude, evenbetween closely related species. Some variation is generatedso rapidly that two different allelic versions of a chromosomalsegment (otherwise known as haplotypes) can be dissimilar ingene content and arrangement even within a single plant spe-cies like maize (Fu and Dooner, 2002). Hence, plant nucleargenomes vary sufficiently to serve as powerful differentiatingfactors. Some changes clearly mark particular lineages of seedplants, such as the large inversions and translocations that arefound within some clades of grasses (Gale and Devos, 1998).Other changes, such as polyploidy and most gene duplications/deletions, are so frequent that they occur independently inmultiple lineages. Recent studies have begun to characterizethe natures, rates, and mechanisms of these various types ofchromosomal rearrangement, thereby providing our first de-tailed insights into how these changes contribute to currentevolved states and how they may be used in phylogenetic anal-ysis.This article will describe the standard structural patterns inthe nuclear genomes of seed plants and show how these havebeen conserved over the last 100 million years of angiospermevolution. The natures, mechanisms, and frequencies of spe-cific chromosomal rearrangements will be described. Genomesize variation and genome duplication will be discussed insome detail. Through this presentation, we hope to provide acomprehensive view of the current understanding of plant nu-clear genome structure and evolution and indicate future di-rections of this field of study. Much of this review will focuson grasses (Poaceae), primarily because so much comparativegenome structure information is available within this family.Comparative sequence analysis has been undertaken in or-1Manuscript received 8 January 2004; revision accepted 15 June 2004.The authors thank the editors of this special issue for inviting them tocontribute this article. We also thank Simon Malcomber for help with theillustrations and Jonathan Wendel, Loren Rieseberg, and Associate EditorsJeff Palmer and Mark Chase for comments that greatly improved the manu-script.4E-mail: [email protected] regions of the barley, maize, rice, sorghum, andwheat genomes (reviewed in Bennetzen and Ramakrishna,2002), thereby providing a large data set for the characteriza-tion of local genome evolution. Comparative recombinationalmaps have also been generated for these species, as well asfor pearl millet, sugarcane, foxtail millet, rye, and a few others(Gale and Devos, 1998). Hence, sufficient data are availableonly in the grasses for the comprehensive characterization ofplant nuclear genome structure and evolution. Comparativedata are rapidly accumulating for other families, however, no-tably for Brassicaceae, and we expect that the next severalyears will see a tremendous increase in information on howgenomes evolve.Many other important aspects of nuclear genome evolutionwill not be covered here, although fascinating new data areaccumulating. For example, the overall base composition ofgenomes varies and affects codon usage patterns. Codon usagepatterns and percentage GC in onion are more similar to Ar-abidopsis than to rice, suggesting (although hardly proving)that the high GC content of the grasses may be derived (Kuhlet al., 2004). Intron size and number vary among plants, butthe correlates of this variation are not well known. In animals,small genome size correlates with smaller introns, but this re-lationship does not seem to hold in plants (see for example,Dubcovsky et al., 2001; Wendel et al., 2002). Base composi-tion, rates of substitution, and size and number of introns areall important aspects of genome evolution, and we hope thatmore data and analyses will be forthcoming on all topics inthe next few years.STRUCTURE OF SEED PLANT GENOMESGross genome structure—All angiosperms contain relative-ly complex nuclear genomes with genes scattered across mul-tiple chromosomes. Other plants are less well characterized,but their large genome sizes (see Bennett and Leitch, 2003,for a comprehensive presentation of plant genome sizes) in-dicate that most nonflowering plant genomes are equally com-plex. In even the smallest genomes, like Arabidopsis (about140 Mb), more than 20% of the DNA is composed of variousrepetitive elements (Arabidopsis Genome Initiative, 2000).1710 [Vol. 91AMERICANJOURNAL OFBOTANYThese repeats include transposable elements and various typesof simple tandem repeats, including satellite DNA and simplesequence repeats (SSRs). In most angiosperms, transposableelements, especially long terminal repeat (LTR) retrotranspo-sons, comprise the vast majority of this repetitive DNA (re-viewed in Bennetzen, 2002a).Chromosome numbers are highly variable in the floweringplants and do not generally relate
View Full Document
Unlocking...