Work

Comparative Sequence Analysis of Plant Nuclear Genomes: Microcolinearity and Its Many Exceptions

Year: 2000
Type: article
Abstract: The flowering plants comprise some 250,000 species and are tremendously diverse in growth habit, environmental adaptation, and nuclear genome structure. Plant genomes tend to be large and complex, varying in size from ∼38 Mb (1C) for the crucifer Cardamine amara to >87,000 Mb for Fritillaria assyriaca, a member of the Lilliaceae (Flavell et al., 1974; Bennett and Leitch, 1995). Despite this diversity, plant geneticists have recently found that plants exhibit extensive conservation of both gene content and gene order (Bennetzen and Freeling, 1993). Moreover, comparative genetic analyses have begun to show that different plant species often use homologous genes for very similar functions (Fatokun et al., 1992; Ahn et al., 1993; Paterson et al., 1995; Lagercrantz et al., 1996). The advent of DNA marker technology not only facilitated the rapid generation of detailed plant genetic maps but also allowed map comparisons when common DNA markers were employed. Initial comparisons of the chromosome sets within polyploid species (Helentjaris et al., 1988; Chao et al., 1989) and of diploid genomes between closely related species (Bonierbale et al., 1988; Hulbert et al., 1990) indicated extensive colinearity of genetic maps. These comparisons have been greatly extended, particularly in the grasses (reviewed in Gale and Devos, 1998), revealing significant conservation of gene content and order across plant species that diverged from a common ancestor <50 million years ago (Crepet and Feldman, 1991). These comparative studies show that, within the limits of sequence divergence (i.e., evolutionary time) that permit cross-hybridization, the large majority (>90%) of plant genes have close homologs within most other plant genomes. Gene order is somewhat more variable, however. Comparative genetic maps based on shared restriction fragment length polymorphism probes indicate many large chromosomal rearrangements, some arising specifically during the origin of a particular family of plants (Moore et al., 1995; Gale and Devos, 1998). These large rearrangements are most commonly observed, as one might expect, between distantly related species (Paterson et al., 1996) but can also occur in a relatively short period of time within some lineages (Devos et al., 1993; Zhang et al., 1998). However, there are problems in the interpretation and use of comparative genetic maps based on recombinational mapping of DNA markers. The two chief difficulties arise from marker density and sequence orthology. Because only a subset of DNA probes hybridize efficiently and reveal polymorphism within mapping populations of two different plant species, most genetic map comparisons employ 200 or fewer shared markers. Hence, marker density averages <1 per 10 centimorgans, guaranteeing that small exceptions to colinearity (e.g., inversions of less than a few centimorgans) will not be observed. Even more problematic is the fact that most genes are represented as multiple homologs within a given plant genome, which makes it difficult to determine whether truly orthologous loci are being compared. Orthologous loci are those that show common vertical descent from an ancestral gene, whereas paralogous genes are derived from amplification of the original ancestral gene, an event that may predate or postdate the independent derivation of any two compared species. In many cases, a restriction fragment length polymorphism probe will hybridize to several gel blot bands from each of two species, but only one may be polymorphic and mapped in each species. If they appear to map in colinear positions, then the researcher often assumes that they are orthologous. However, if they map on nonsyntenic linkage groups, the researcher commonly assumes that paralogs have been mapped. Although this circular reasoning often is correct, it can lead to a bias toward promoting colinear cases and ignoring exceptions to colinearity. Further confusing this issue is the fact that many paralogs (derived from tandem duplications) are commonly found within a few centimorgans of each other, thus making mere synteny a less-than-perfect argument for orthology. All of these problems are intensified by comparative maps that rely on small numbers of analyzed progeny. If the accuracy of placement of a particular marker is only within 5 to 10 centimorgans, then it is difficult to determine whether an apparent local exception to colinearity is really due to a small chromosomal rearrangement or to imprecise map positions for the markers involved. Partly because of the limitations of genetic maps based on DNA markers, several laboratories have begun to investigate and compare the structure of small regions of plant genomes by DNA sequence analysis. These studies will provide our first views into the detailed composition and organization of numerous plant genomes. Moreover, such comparative sequence analyses will help determine how and at what rates plant genomes change. Although many studies have focused on the timing and nature of gene evolution in plants (reviewed in Doyle and Gaut, 2000), a lack of genomic sequencing capability has meant that these investigations primarily target gene sequences themselves, without any reference to the different contexts in which these genes are located. The remainder of this article discusses recent findings that have begun to uncover the relationship between plant gene and genome evolution in the nucleus. Over the last several years, a large-scale effort has been under way to sequence the genome of a model plant, Arabidopsis. Arabidopsis was chosen for this ambitious effort because of its exceptionally small genome (∼130 Mb) (Arumuganathan and Earle, 1991), its powerful array of genetic tools, and its outstanding research community (Meinke et al., 1998). Currently, ∼80% of the Arabidopsis genome is completely sequenced, with the rest due to be finished this year, 4 years ahead of schedule. Analysis of the Arabidopsis genome sequence has shown that gene density throughout most of the genome is very high, approximately one gene per 4 to 5 kb, suggesting a total gene content of 25,000 to 30,000 (Kaneko et al., 1999; Lin et al., 1999). Repetitive DNAs are relatively rare, comprising ∼10% of the Arabidopsis genome (Leutweiler et al., 1984). Many of these repeats are transposable elements, found as single copies interspersed between genes (Lin et al., 1999). Most of the repeats, however, are sequestered within centromeric regions, pericentromeric heterochromatin, and knobs. These repeat-rich, gene-poor chromosome segments contain tandem satellite repeats, some apparently involved in centromere function, as well as large quantities of transposable elements (Copenhaver et al., 1999; Fransz et al., 2000; McCombie et al., 2000). Plants other than Arabidopsis have received much less attention at the level of DNA sequence analysis. Several large segments of the rice (Oryza sativa) genome have now been subjected to sequence analysis, revealing at least one gene per 10 kb, with scattered transposable elements making up 25% or less of the total DNA (Chen and Bennetzen, 1996; Han et al., 1999; Tarchini et al., 2000). Two relatively large segments of the maize (Zea mays) genome have been sequenced, a 225-kb segment around the adh1 gene (SanMiguel et al., 1996; Tikhonov et al., 1999) and a 78-kb segment that contains a 22-kD zein gene cluster (Llaca and Messing, 1998). In both maize segments, genic regions were interspersed with large blocks of repetitive DNA. The repetitive DNA was found to be composed mainly of retrotransposons with long terminal repeats (LTRs), of which most were found to be inserted into each other (SanMiguel et al., 1996). The LTR retrotransposon blocks are highly methylated and presumably heterochromatic in most or all adult tissues, and range in size from a few kilobases up to ∼200 kb (Springer, 1992; Bennetzen et al., 1994). Overall, these LTR retrotransposons are estimated to comprise 50 to 80% of the DNA in the maize nucleus (SanMiguel and Bennetzen, 1998). Small, multigene segments of a few other plant species have also been sequenced, including a 60-kb region around the barley mlo locus (Panstruga et al., 1998), a 23-kb region on wheat chromosome 1 (Feuillet and Keller, 1999), a 78-kb segment flanking a sorghum adh gene (Tikhonov et al., 1999), and a 26-kb region of Capsella rubella, a close relative of Arabidopsis (A. Acarkan, M. Rossberg, M. Koch, and R. Schmidt, personal communication). These regions all exhibit high gene densities, with only the occasional single transposable element. Thus, most of the plant nuclear genomes that have been subjected to sequence analysis have a gene and repetitive DNA composition like that of the genic regions of Arabidopsis, whereas maize appears to be exceptional in having genic regions that look more like the pericentromeric heterochromatin of Arabidopsis or the β heterochromatin of Drosophila melanogaster (Holmquist et al., 1998; McCombie et al., 2000). Given that so few plant genomes have been investigated, it is not clear whether the structure and composition of maize genic regions are due to exceptional mechanisms of genome alteration or whether they might be due to the relatively short period of time since most of these LTR retroelements have amplified, a mere 2 to 6 million years (SanMiguel et al., 1998). It is possible that many increases in plant genome size are due to transposable element amplification, perhaps in very narrow temporal windows, while a slow but steady process of genome size reduction counteracts this trend (Petrov et al., 1996; Bennetzen and Kellogg, 1997; SanMiguel and Bennetzen, 1998). One universal observation of all fine-scaled analyses of eukaryotic genomes has been the great number of duplicated genes. In some cases, these duplications may be due to ancient polyploidizations that were subsequently obscured by variation in the pairing, segregation, or structure of the different chromosome sets within a single nucleus. For instance, even the simple yeast genome appears to be derived from a primordial tetraploid (Wolfe and Shields, 1997), and ancient duplications, including have been well in the evolution of et al., 1998). It not be that given and for are particularly to Many species, like maize and those in the (Helentjaris et al., 1988; 1998), have an ancient polyploid whereas many other plant species as (reviewed in 2000). mapping and sequence analyses have now shown that even simple plant genomes contain of genetic maps of Arabidopsis, and other species have that chromosome and segments are duplicated et al., et al., 1995). sequence the genome of Arabidopsis appears to be at least suggesting a possible tetraploid et al., 2000). studies also shown that within the ancient tetraploid genome of duplications to genetic mapping studies et al., 1995). Hence, these or genomic duplications can for analyses for as to what is being For instance, sequence comparisons can be between homologous regions within a chromosome the different and rates of that can occur to segments within a single nucleus. in comparisons between species, it may be whether an orthologous or paralogous is being Analysis of sequences derived from two different plant species more and more as the of For instance, sequence comparisons between Arabidopsis and any of the grasses have shown that few colinear segments can be even at the level of genes (Tikhonov et al., 1999; et al., 1999; Devos, personal and Bennetzen, However, these sequences might have diverged so greatly that the of is to the other comparisons between very closely related species, or populations within a species, can only reveal that have been within the last few to of One segment of the rice genome, genes orthologous to the and loci of has been and analyzed on from the and 1998; M. and Bennetzen, The two orthologous regions appear to have for million years with conservation in order and of all genes. and regions very different rates and of sequence however. found that and most those that a simple sequence whereas most DNA diverged approximately more However, transposable elements et al., within the region diverged approximately than the rest of the DNA 1998; M. and Bennetzen, In the to be in sequence so both single and that they within a few million may two have been at a in sequence comparisons of orthologous regions of the genomes of plant species, like maize and sorghum (Tikhonov et al., 1999), that have diverged for million years and of gene sequence polymorphism that and Arabidopsis have for to 10 million years and that they diverged from the to million years ago (A. Acarkan, M. Rossberg, M. Koch, and R. Schmidt, personal communication). In a 26-kb the genes were and in the order in the two species. However, one of the genes a tandem in its divergence from Arabidopsis. In this time conservation was for but number and were highly The sequences within were most but orthologous short of sequence (A. Acarkan, M. Rossberg, M. Koch, and R. Schmidt, personal communication). Further studies may indicate whether these short segments are related to any of an analyses of and rice regions the most comparisons of orthologous regions in and sorghum appear to have diverged to million years whereas common ancestor diverged from a shared ancestor with rice million years ago and regions reveal sequence conservation these species. In of the DNAs at a in one species can be found in the orthologous in species that is this distantly related (Chen et al., 1998; Tikhonov et al., 1999). However, in a of kb the regions of rice and and found that all of the genes were in both order and In this one was found to by a particular suggesting that these genes might have diverged in (Chen et al., 1998). In to the in the regions of and regions have significant gene rearrangements (Tikhonov et al., 1999; Tarchini et al., 2000). In a of maize and Tikhonov and found that genes were shared in colinear but the maize region two apparent that genes (Tikhonov et al., 1999) The rice adh1 region colinearity with the maize or sorghum adh1 region the adh1 gene (Tikhonov et al., 1999; Tarchini et al., 2000), suggesting that adh1 to its nonsyntenic in rice as a comparisons between plant the between and have not many mapping studies that some colinear regions might between Arabidopsis and the grasses (Paterson et al., but a more detailed the rice and Arabidopsis genomes not for significant of colinearity (Devos et al., 1999). genome sequence comparisons also have not any genetic colinearity across the divergence (Tikhonov et al., 1999; Devos, personal and Bennetzen, However, all of these studies may be by of and only analyses of the sequences of a and a genome will of this Despite great in both size and plant genomes exhibit some but exceptional such as et al., 1996; et al., 1999), et al., 1998; Fransz et al., 2000; McCombie et al., 2000), and may have very similar in very different plant species. regions also exhibit some as in of of and the indicate genes that are homologous to or genomic genes from plants and other species. The within the show including and and apparent The between the two regions with extensive sequence The has been greatly by of the numerous blocks of LTR retrotransposons (SanMiguel et al., represented by the is based on and analyses by Tikhonov et yeast In all sequence analyses to plant genes have been found to be with small It is not for two in to an region of <1 In most cases, genes not appear to be by or of in the of duplicated from the numerous of transposable elements are found in genic of these like and some are found or within genes et al., 1995; et al., 1996; Tikhonov et al., 1999). like most highly repetitive LTR appear to with other repetitive DNAs et al., 1996; SanMiguel et al., 1996; et al., 1998; McCombie et al., 2000). have of the for the in genome composition or organization It is clear that the size variation is an of transposable element amplification, but it is not clear some species to have these more often than may be a of or it might be that some species from or are less to genome size increases (Bennetzen and Kellogg, It is however, that all comparative investigations to have this great variation in genome size to be with any on genic colinearity in orthologous for the of Plant the indicate genes represented by vertical each and of The region an LTR retrotransposon The and of these blocks appear to be between genomes and perhaps between regions of the of the model are in the simple sequence is represented by an Comparative studies of orthologous regions have that is often even between distantly related species. might be colinearity is most between closely related species. For instance, colinearity is between the regions of maize and sorghum than it is between those regions from of these species and rice (Tikhonov et al., 1999). However, there are some significant exceptions to this In genetic map comparisons within the and have shown that wheat and barley exhibit colinearity than wheat with the fact that is a relative (Devos et al., 1993). Thus, some lineages of plants appear to have more of rearrangement than analyses of the of large chromosomal rearrangements have for a for the generation and of such during the evolution of the grasses (Paterson et al., with some exceptions (Devos et al., 2000; Zhang et al., 1998). However, it is not at all clear whether small rearrangements might occur at or comparisons of the of large of orthologous regions et al., 1995; et al., and of genomic sequences (Chen et al., 1998; Tikhonov et al., 1999) reveal small rearrangements, and tandem duplications or appear to be particularly Thus, it appears that small rearrangements are of more than are large chromosomal It is not clear whether small local and large chromosomal rearrangements be by the or not there are different it is that an a high of small rearrangements might not a high of large rearrangements or It is that not the or of genome at any in any family of species. few plant genomes have been at the DNA sequence and these only in a few Hence, it is to how and genomes have these However, a few have in the of plant genome structure and of chromosome number and genome size in closely related plant species have long in genome structure within a short evolutionary time Comparative sequence analyses of genes and of repetitive DNAs have shown that these at different with the genes being more highly 1998; SanMiguel et al., 1998). few like some are for rapid (reviewed in and and thus may more than some However, the of many repeats and the of apparently most of these sequences to even more than genes under The variation in genome size and repetitive DNA content (Flavell et al., that of these has much on plant the with in genome size and repetitive DNA content might a under some and and 1995). that the of a great array of transposable elements be under of in which with the genome be of in a plant that a 1984). However, it has been that it is difficult to how such a genomic be without being for of and that it is that a number of rearrangements a even in the of a 2000). Moreover, closely related plant species show in genome but not in gene or the fact that model that in gene and order be the of the genomic In it to that the DNAs and tandem repeats that up most of the repetitive DNA in genomes are or that primarily because they have and and of genome including large and inversions and have been to be of these of rearrangement can greatly Hence, will toward the of these at a that is to One the gene content of the plant genome to be under the of any gene will be in a and the of but loci also greatly in gene number might also plant particularly if these increases are genic an in gene number without a alteration in genic the exception of those in that might the two chromosome sets in an In an all genes are in multiple so that one of an gene into a without total of the original gene gene copies may be if the genetic is with the polyploid or if the genetic is to the the of an of a gene may be of if it the divergence of In a polyploid within the last years, each of the and genomes appears to one of an orthologous locus et al., Hence, with a few exceptions (Devos et al., the and genomes within wheat have not extensive gene or rearrangement since polyploid In more ancient like many between the chromosome sets have begun to of these rearrangements (Moore et al., but most appear to be with small local Tikhonov et that genes in the sorghum adh region were from the orthologous maize However, by analysis, these genes were found to be in the maize It will be particularly to if the genes are colinear within the maize a possible model for the structure of the orthologous regions of a diploid genome and the tetraploid genome of a related species. Because there are copies of each gene in the there is for or divergence of each of these genes to In the all genes are and all are colinear in the but in one have to investigate both orthologous segments from the tetraploid to such The apparent lack of between the genomes of Arabidopsis and the grasses (Bennetzen et al., 1998; et al., 1999; Tikhonov et al., 1999) is a observation that colinearity. If many genomes have extensive colinearity 50 million years or then colinearity be completely a time that is only a (Crepet and Feldman, It is difficult to how colinearity be on a it is to that a very high level of genome rearrangement in very divergence from in a level of colinearity et al., 1999). high rates of rearrangement may have specifically in the of Arabidopsis and its close Further DNA sequence comparisons between Arabidopsis and other between these two et al., 2000). for Orthologous Gene in a and are indicated by and common In this all of the ancestral genes are in the diploid region but genes apparent have been in generation of the tetraploid genome and one gene from the tetraploid that copies of all of the genes are in the but some genes (e.g., and are now only as diploid It is possible to for or genes. In the of some plant for instance, linkage can rapid evolution of by a process of et al., 1993; in and 2000). However, this process of the of even the thus for of very to that lack genes of the may for the fact that many genes not appear in orthologous species et al., 1998). However, the linkage of genes that have a particular be a that is as a single Although more are can now that genomic in plants will be a for plant gene and The most use for genomic is in the of genes by chromosome within small colinear genomes of species (Bennetzen and Freeling, 1993). However, the of this not on genome colinearity but on local at reference genes are located. because of the high of small rearrangements between distantly related species, this has to However, several are very close to and even of can provide a number of DNA probes to loci et al., Comparative sequence analysis can provide a for sequence In comparisons of genomes that have diverged 5 million years or it appears that only the genes are Hence, any sequences that are shared between two orthologous regions are to be genes et al., 1996; Tikhonov et al., 1999), even when they are not by a or by to any other gene or For the long the most and use of comparative sequencing and mapping will be to determine the nature of the functions that one species different from (Bennetzen and Freeling, Because colinear map positions a of the can in the sequence and of a locus with the different of and of a particular orthologous gene in two or more different species. compared across the plant the of a single orthologous gene can be will reveal not only how a gene can but also what functions it may in the nature of evolution and the of all of the genes in a Comparative sequence analyses have begun to uncover the and of plant genome structure. Several are to the of repetitive including with genes and with is apparent in comparisons of many plant but there are also many small rearrangements, including of transposable elements and duplications or of occur without most The timing of these may be (SanMiguel et al., 1998), perhaps related to the environmental or of the exceptions to colinearity are much more than are the large rearrangements by and genetic maps. Plant genome can be by plant geneticists and genetic the most to be from will be the of that will for clear between the nature and of plant genome
Language: English
Subfield: Genetics
Open Access status: bronze
OpenAlex ID: W2159077216
FWCI: 14.48
Cites: 80
Cited by: 279
Related to: 9