Comparative Mapping Between Arabidopsis thaliana and Brassica nigra Indicates That Brassica Genomes Have Evolved Through Extensive Genome Replication Accompanied by Chromosome Fusions and Frequent Rearrangements

Corresponding author: Ulf Lagercrantz, Swedish University of Agricultural Sciences, Box 7080, S-750 07 Uppsala, Sweden., ulf.lagercrantz{at}vbiol.slu.se (E-mail).

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Chromosome organization and evolution in the Brassicaceae family was studied using comparative linkage mapping. A total of 160 mapped Arabidopsis thaliana DNA fragments identified 284 homologous loci covering 751 cM in Brassica nigra. The data support that modern diploid Brassica species are descended from a hexaploid ancestor, and that the A. thaliana genome is similar in structure and complexity to those of each of the hypothetical diploid progenitors of the proposed hexaploid. Thus, the Brassica lineage probably went through a triplication after the divergence of the lineages leading to A. thaliana and B. nigra. These duplications were also accompanied by an exceptionally high rate of chromosomal rearrangements. The average length of conserved segments between A. thaliana and B. nigra was estimated at 8 cM. This estimate corresponds to ~90 rearrangements since the divergence of the two species. The estimated rate of chromosomal rearrangements is higher than any previously reported data based on comparative mapping. Despite the large number of rearrangements, fine-scale comparative mapping between model plant A. thaliana and Brassica crops is likely to result in the identification of a large number of genes that affect important traits in Brassica crops.

ONE important aspect of genome evolution is changes in organization of the DNA caused by duplications and chromosomal rearrangements. Comparative linkage mapping has indicated that many animal and plant genomes have remained surprisingly conserved during evolution (O'BRIEN et al. 1988 ; WHITKUS et al. 1992 ; AHN and TANKSLEY 1993 ; MORIZOT 1994 ; MOORE et al. 1995 ). However, a variation between different evolutionary lineages in the rate of chromosomal evolution is evident, both from cytogenetic work (WILSON et al. 1974 , WILSON et al. 1977 ; BUSH et al. 1977 ; ELDRIDGE and CLOSE 1993 ) and comparative mapping (NADEAU and TAYLOR 1984 ; BARENDSE et al. 1994 ; JOHANSSON et al. 1995 ; GRAVES 1996 ; EHRLICH et al. 1997 ). This pattern has led O'BRIEN et al. 1988 to suggest that karyotypes are generally conservative but occasionally undergo radiation or "shuffles." The cause of these sporadic shuffles is still under debate. In plants, conservation of surprisingly large chromosomal segments has been observed in the Graminae family (MOORE et al. 1995 ), while tomato and chili pepper, in the Solanaceae family, often have been cited as a pair of species that differ by a relatively large number of rearrangements (TANKSLEY et al. 1988 ; PRINCE et al. 1993 ).

Polyploidy is particularly common among plants. It has been estimated that up to 80% of angiosperms are polyploid (MASTERSON 1994 ). Furthermore, recent genetic mapping has revealed cryptic polyploids indicating that the level of polyploidy has been underestimated in several species (HELENTJARIS et al. 1988 ; REINISCH et al. 1994 ; MOORE et al. 1995 ; LAGERCRANTZ and LYDIATE 1996 ). Recent studies have also suggested that polyploidization may be accompanied by rapid genomic change (SONG et al. 1995 ; KENTON et al. 1993 ; CHEN and ARMSTRONG 1994 ; JELLEN et al. 1994 ; LAGERCRANTZ and LYDIATE 1996 ; SHOEMAKER et al. 1996 ; LEITCH and BENNETT 1997 ).

The family Brassicaceae (Cruciferae) is widely distributed and comprises more than 3000 species in approximately 350 genera. The family includes important crops such as Brassica oleracea, B. napus, and B. rapa, as well as the extensively studied model plant Arabidopsis thaliana. The family's major centers of diversity are southwestern and central Asia and the Mediterranean region. Secondary centers of diversity are in the arctic, western North America, and the mountains of South America (PRICE et al. 1994 ).

Species within the family exhibit a continuous range of haploid chromosome numbers from 5 to 15, excluding a large number of known polyploid species with higher chromosome numbers. Previous comparative mapping has indicated that present-day diploid species in the Brassica genus are derived from a hexaploid ancestor (LAGERCRANTZ and LYDIATE 1996 ). Furthermore, preliminary comparisons between Brassica species and A. thaliana have indicated that the A. thaliana genome has a complexity corresponding to one of the three subgenomes of modern diploid Brassica species (LAGERCRANTZ et al. 1996 ; SCHEFFLER et al. 1996 ; CAVELL et al. 1998 ). Comparative mapping data have also indicated a relatively high rate of chromosomal rearrangements in the Brassicaceae family (KOWALSKI et al. 1994 ; LAGERCRANTZ and LYDIATE 1996 ; OSBORN et al. 1997 ). With one exception, previous comparisons between the Brassica and Arabidopsis genomes have been limited to small genomic regions (LAGERCRANTZ et al. 1996 ; SCHEFFLER et al. 1996 ; CAVELL et al. 1998 ). The only across-genome comparison published so far (KOWALSKI et al. 1994 ) was based on a low marker density (the 75 previously positioned B. oleracea loci had an average spacing of one locus per 13 cM; SLOCUM et al. 1990 ). Because of the limitations of incomplete polymorphism in the B. oleracea population, the degree of genome replication in B. oleracea was also underestimated in that study. Consequently, to elucidate the relationship between the uniquely small genome of A. thaliana and the highly replicated Brassica genomes, there is a need for additional comparative mapping data. The apparently high rate of chromosomal rearrangements in the Brassicaceae family also requires an exceptionally high density of markers.

The complete genome sequence of A. thaliana will be available within a few years (ECKER 1998 ), and the function of many of the identified genes will become elucidated. An understanding of the complex relationship between the small genome of A. thaliana and its highly replicated close relatives in the Brassica genus could have a great impact on the understanding of the biology and the improvement of the important Brassica crops.

To obtain a more comprehensive picture of the relationships between A. thaliana and the members of the Brassica genus, I applied a large number of previously mapped A. thaliana probes to a highly polymorphic mapping population of B. nigra. This strategy yielded a high marker density (an average spacing of one locus per 3 cM), and a high proportion of replicated loci mapped. Compared to earlier studies, these data allowed a much more detailed comparison comprising the entire A. thaliana genome. The data support that diploid Brassica genomes contain three copies of a basic genome similar in size to the A. thaliana genome, and that chromosome evolution in Brassicaceae seems to involve an exceptionally high rate of chromosomal rearrangements. The comparative data also have implications for the possibilities of transferring genetic resources and information between A. thaliana and Brassica species.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The B. nigra genetic map was developed using a previously described mapping population (LAGERCRANTZ and LYDIATE 1995 ). The population was a highly polymorphic, first backcross population of 88 individuals in which it was possible to score segregation of markers from both the F₁ and (highly heterozygous) recurrent parents. Linkage maps from meioses in the F₁ and recurrent parents were initially developed separately using MAPMAKER 3.0 (LANDER et al. 1987 ; LINCOLN et al. 1992 ) and then integrated using JoinMap (STAM 1993 ). The integration of two linkage maps can introduce ordering errors, particularly if few markers are common to both maps. This potential source of inaccuracy caused few problems in the current analysis as a majority of the loci positioned on the map from the recurrent parent were also positioned on the F₁ map (LAGERCRANTZ and LYDIATE 1996 ). All loci were detected using Southern hybridization analysis as described by LAGERCRANTZ and LYDIATE 1995 , except that washing was carried out at a lower stringency (2x SSC, 0.1% SDS at 65°). Hybridization probes were prepared by PCR or gel isolation of inserts and labeled with [³²P]dCTP.

A set of 160 DNA fragments from the A. thaliana genome were used as RFLP probes. The probes were derived from the following sources (prefix, reference, and source in parentheses): 69 genomic PstI clones (mi, LIU et al. 1996 , Ohio Stock Center); 21 anonymous cDNAs (ve, D. BOUCHEZ, personal communication, Ohio Stock Center); 18 genomic clones from the ARMS set (m, FABRI and SCHAEFFNER 1994 , Ohio Stock Center); 20 anonymous cDNAs (um, MCGRATH et al. 1993 , Ohio Stock Center); 8 anonymous cDNAs (y and tai224, unpublished data, Ohio Stock Center); and six YAC end probes and 3 anonymous cDNAs (L, R, and c, LAGERCRANTZ et al. 1996, Dr. G. Coupland). In addition, the following characterized A. thaliana genes were used: Phytochrome A, B, and C (phy, SHARROCK and QUAIL 1989 , Dr. P. Quail); Athb3, Athb7, Athb13 (Athb, SODERMAN 1996 , Dr. P. Engström); Chalcone synthase (chs, FEINBAUM and AUSUBEL 1988 , Dr. G. Coupland); Ara (MATSUI et al. 1989 , Dr. H. G. Nam); Rps2 (rps2, MINDRINOS et al. 1994 , Dr. F. Ausubel); Gpa1 (gpa1, MA et al. 1990 , Dr. H. Ma); GapC (gapC, SHIH et al. 1991 , Dr. M-C. Shih); Rns1 (rns1, BARIOLA et al. 1994 , Dr. P. Green); Ag (ag, YANOFSKY et al. 1990 ); Ga1 (ga1, SUN and KAMIYA 1994 ); and Pg11 (pg11, GALLANT and GOODMAN, unpublished data). Probes for these were derived from cDNA clones, except for Chs2 (genomic clone) and Ag, Ga1, and Pg11 (PCR amplified from genomic A. thaliana DNA; KONIECZNY and AUSUBEL 1993 ).

Most probes (88%) were mapped in A. thaliana using the recombinant inbred (RI) population of LISTER and DEAN 1993 . The positions of the loci detected by a few of these probes were inferred from physical mapping data. Thirteen A. thaliana expressed sequence tags (ESTs) (ve29, ve30, ve45, ve50, ve53, ve57, y22, y29, y33, y34, y35, y36, and y37) were assigned to YAC clones (ve clones, D. BOUCHEZ, personal communication; y clones, http://genome-www.stanford.edu/ Arabidopsis/ EST2YAC.html/), and the genetic map position was inferred from the position on the RI map of genetically mapped fragments from the same YAC clone. The genetic map positions of all loci on the RI map were extracted from the map published electronically (http://nasc.nott.ac.uk/new ri map.html/).

Nineteen probes were mapped in separate A. thaliana crosses (Ara and PhyB, HAUGE et al. 1993 ; 17 of the um probes, MCGRATH et al. 1993 ). Loci segregating in two crosses were used to infer the relative order of loci mapped in one cross only. Several of the um probes also detect multiple loci in A. thaliana. This complicates the identification of the true orthologous/homeologous loci in B. nigra. In this study, um loci in A. thaliana and B. nigra were considered homeologous if they were positioned next to another homeologous locus pair detected with a non-um probe. Consequently, occasional erroneous classifications of true homeologous locus pairs would result in an underestimate of the number of rearrangements that have occurred since the divergence of the lineages leading to A. thaliana and B. nigra. Data from the um probes were excluded in calculations of the degree of duplication in B. nigra.

Polymorphic loci detected using the A. thaliana probes were positioned on the B. nigra genetic map relative to 288 loci previously mapped using Brassica RFLP probes (LAGERCRANTZ and LYDIATE 1995 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

A total of 160 mapped DNA fragments from the A. thaliana genome revealed 284 polymorphic loci that were incorporated into the existing B. nigra linkage map (Figure 1). With one exception, each B. nigra linkage group contained loci detected with probes from all five A. thaliana chromosomes (Figure 1, Table 1). This pattern indicates that substantial chromosomal rearrangements have occurred since the divergence of the ancestors to A. thaliana and B. nigra.

View larger version (112K): In this window In a new window Download PPT slide   Figure 1. Comparative map of A. thaliana and B. nigra. The eight linkage groups of B. nigra (G1–G8) are represented by vertical lines. Loci on the right of each linkage group were detected using probes previously mapped in A. thaliana. The chromosomal locations of A. thaliana loci detected using these probes are shown with different colors as indicated. Loci on the left of each linkage group were previously located using Brassica RFLP probes (LAGERCRANTZ and LYDIATE 1995 ). Recombination distances are given in Kosambi centimorgans.

To be able to elucidate the true degree of collinearity between the genomes of A. thaliana and B. nigra, the highly duplicated nature of the B. nigra genome has to be fully appreciated. A majority of A. thaliana probes detect a single locus in A. thaliana (CAVELL et al. 1998 ; CHANG et al. 1988 ). In contrast, in B. nigra, the A. thaliana probes detected on average 1.8 polymorphic loci per probe, which is close to the 1.9 loci per probe detected in the same population using Brassica RFLP probes (LAGERCRANTZ and LYDIATE 1996 ). Both those estimates are clearly underestimates of the true level of duplication caused by residual monomorphism (LAGERCRANTZ and LYDIATE 1996 ). Even if the degree of polymorphism is as high as 70% (as estimated in the present study), the probability that all loci in a triplicate set are polymorphic is 34%. Thus, in a hexaploid where virtually every probe should identify three homeologous loci, about one-third of these triplicate loci would be mapped, even in this extremely polymorphic population.

Previous mapping using Brassica RFLP probes has suggested that the entire B. nigra genome appears to consist of large, duplicated segments, with the majority of RFLP loci detecting three dispersed, homeologous loci (LAGERCRANTZ and LYDIATE 1995 , LAGERCRANTZ and LYDIATE 1996 ). The present study supports these findings. The mapping population used by LAGERCRANTZ and LYDIATE 1995 , LAGERCRANTZ and LYDIATE 1996 was also used in the present study, and the additional 284 loci based on A. thaliana probes corroborated the previously identified triplicated segments comprising virtually the whole B. nigra genome (data not shown). These data support the hypothesis that B. nigra is descended from a hexaploid ancestor.

How does this triplicated nature of the B. nigra genome correspond to the structure of the A. thaliana genome? Because of the high frequency of rearrangements (see below), collinear segments are quite short in many cases. However, a few large chromosomal segments have remained largely intact since the divergence of Arabidopsis and Brassica, and for these segments, which occur as single copy in A. thaliana, there are strong indications that they occur in three homeologous copies in B. nigra (Figure 2). At least 40 cM of the top of A. thaliana chromosome 5 is present in three homeologous copies on linkage groups G2, G5, and G8 in B. nigra. Two of the B. nigra segments appear collinear with the A. thaliana segment, while the third segment on G5 contains a large inversion compared to that of A. thaliana.

View larger version (30K): In this window In a new window Download PPT slide   Figure 2. Large A. thaliana chromosomal segments correspond to triplicated homeologous segments in B. nigra. Detailed comparative linkage maps of selected segments of the A. thaliana genome. A 40-centimorgan segment of A. thaliana chromosome 5 corresponds to three homeologous tracts on G2, G5, and G8 in B. nigra. The bottom half of A. thaliana chromosome 2 is homeologous to segments on B. nigra G1, G6, and G8. Forty centimorgans of the top of A. thaliana chromosome 3 corresponds to triplicated segments, one on B. nigra G1 and two on G7. The numbers connected with lines to loci on A. thaliana chromosome 2 indicate linkage group location of homologous loci in B. nigra. Those B. nigra loci are not adjacent on their respective linkage groups and, thus, are not connected by vertical lines (representing synthenic segments) in the figure.

Furthermore, large portions of A. thaliana chromosome 2 seem to correspond to three homeologous segments in B. nigra (Figure 2). The bottom 40 cM of A. thaliana chromosome 2 corresponds to a contiguous 50-cM tract on B. nigra G6, while the homeologous segments on B. nigra G1 and G8 are interrupted by segments from other A. thaliana chromosomes (Figure 1 and Figure 2). Loci homologous to those from the top half of A. thaliana chromosome 2 are scattered mainly on the three linkage groups in B. nigra: G3–G5.

In addition, a 40-cM segment of A. thaliana chromosome 3 corresponds to three homeologous tracts in B. nigra (Figure 2). The three B. nigra tracts are all associated with tracts homeologous to A. thaliana chromosome 1 segments (depicted in gray in Figure 2). In two of the B. nigra tracts, the A. thaliana chromosome 3 homeologous segments are interrupted by the tract homeologous to chromosome 1, and for the third tract, an inversion seems to have placed the chromosome 1 homeologous tract at the end of the linkage group. Triplicated homeologous copies are also discernible for more limited regions of the A. thaliana genome (data not shown), but as the homeologous regions get smaller, it becomes progressively more difficult to detect triplicated copies, if they exist, because of lack of polymorphic loci.

Although it is not possible to identify triplicated structures in B. nigra corresponding to the entire A. thaliana genome, probes from the different A. thaliana chromosomes detected a similar level of duplication, ranging from 1.4 to 1.9 loci per probe (Table 1; F₄₁₃₇ = 1.8, P > 0.1), indicating that none of the A. thaliana chromosomes are significantly over- or underrepresented in duplicate homeologous copies in B. nigra. To get a more complete description of replicated regions corresponding to different A. thaliana segments, an even higher density of markers is needed.

The scattered distribution of markers from different A. thaliana chromosomes on all B. nigra linkage groups suggests that a very large number of chromosomal rearrangements have occurred since the divergence from a common ancestor. Simply counting the number of breakpoints that are needed to account for markers from different A. thaliana chromosomes that are adjacent on B. nigra linkage groups yields 79 breakpoints (Figure 1). This calculation does not account for the fact that several blocks of loci syntenic with particular A. thaliana chromosomes are not collinear. For example, on B. nigra G2, neither the block on top homeologus to A. thaliana chromosome 1 nor the large, contiguous block homeologous to A. thaliana chromosome 5 are collinear with their A. thaliana counterparts. In both cases, inversions either in the Arabidopsis or the Brassica lineage are needed to account for the different order of homeologous loci.

To obtain a more detailed picture of the number and types of rearrangements that have occurred during the evolution of Arabidopsis and Brassica from a common ancestor, an attempt was made to reconstruct ancestral chromosomal segments by invoking a minimal number of rearrangements resulting in segments collinear with tracts in the A. thaliana genome. The B. nigra linkage groups show a typical pattern of relatively large blocks of markers from particular A. thaliana chromosomes interrupted by a few markers from one or more other A. thaliana chromosomes. This distribution can, to a large extent, be explained by a limited number of inversions, as illustrated for linkage groups G1 and G8 in Figure 3. Invoking six inversions on B. nigra G8 results in four syntenic blocks, two of which are apparently contiguous and collinear with A. thaliana.

View larger version (101K): In this window In a new window Download PPT slide   Figure 3. Tentative models for the evolution of B. nigra G1 and G8 (see text for details). The present-day B. nigra linkage groups are shown on the left. , inversions; , additional chromosome breaks (fissions of translocations) that are necessary to explain the differences between A. thaliana and B. nigra linkage groups; T, loci where the A. thaliana homologue is located close to a telomere; C, positions that in A. thaliana correspond to positions close to a centromere. The color of the different loci indicate the chromosomal location of the A. thaliana locus detected with the corresponding probe, as in Figure 1.

The other two blocks (homeologous to segments from A. thaliana chromosomes 3 and 4) are also collinear, but they probably lack an internal piece of the segment, as indicated by an additional breakpoint in Figure 3. This conclusion is based on the fact that the A. thaliana homologues to loci flanking the breakpoints are separated by a large segment, and that within this A. thaliana segment there are at least subsegments that correspond to three homeologues on other B. nigra linkage groups.

It should be pointed out that there are other possible scenarios for the chromosomal evolution than those presented in Figure 3, but the fact that a single inversion often places scattered makers, not only in a syntenic block, but also in a collinear position, makes the proposed scenario attractive. The estimates of the number of rearrangements are shown in Table 2. The estimates range from 8 to 20 rearrangements per B. nigra linkage group, resulting in a total of 112 rearrangements. These estimates should be regarded as maximum estimates based on the present data as all interruptions of synteny are not necessarily the result of chromosomal rearrangements. There are a number of segments in Figure 1 that are defined by a single locus. An alternative explanation to the occurrence of such single loci disrupting an otherwise syntenic segment could be the transposition of a duplicated small segment or even partial transcripts.

To reduce the potential bias caused by single deviant loci, the number of rearrangements and the length of conserved segments were also estimated from map distances between the outmost markers of conserved segments (NADEAU and TAYLOR 1984 ). Using this method, a single deviant locus will only affect the estimate if it disrupts an otherwise collinear segment. In the present data, however, the single deviant loci are always located at the breakpoint between different segments (Figure 1). Even when a single deviant locus is flanked by loci homologous to the same A. thaliana chromosome (e.g., as for ve23b about two-thirds down on G2; Figure 1), this locus does not interrupt collinearity as it is located at an inversion breakpoint. The estimate of the conserved length using mapping data from B. nigra is 8 cM. This estimate is based on 41 conserved segments, including two or more markers. An estimate of the number of chromosomal rearrangements based on these data requires the number of chromosomes in the last common ancestor of A. thaliana and B. nigra. This number is not known, but it will only affect the estimate marginally. Assuming tentatively that the ancestor to B. nigra has 15 chromosomes (three copies of the A. thaliana genome) results in an estimate of 87 rearrangements. Assuming conservatively that the ancestor to B. nigra has 8 chromosomes results in a marginally higher estimate of 94 rearrangements. In comparisons of the number of rearrangements with other species (see DISCUSSION), an estimate of 90 rearrangements was used.

Several loci that map close to the telomeres on A. thaliana chromosomes have homologous loci mapping internally on B. nigra linkage groups. The positions of some of those loci in B. nigra suggest that direct telomere-telomere fusion might have been important in the restructuring of Brassica genomes. In the hypothesized scenario for G8 (Figure 3), there are at least three positions where adjacent blocks corresponding to two different A. thaliana chromosomes are joined by loci that have homologues close to the telomeres in A. thaliana.

The positions of centromeres on the genetic map of A. thaliana have been reported recently (ROUND et al. 1997 ). In Figure 3, homologous positions to some of the A. thaliana centromeres have been inferred through the positions of B. nigra homologues to centromere-linked loci in A. thaliana. The proposed locations of the regions that are homologous to A. thaliana centromeres are close to the breakpoints of chromosomal rearrangements in all cases (Figure 3). In at least two cases, however, the positions of flanking loci indicate that the breakpoint has not been in the centromere itself. On G1, mi19b and mi291c map in a cluster of loci from different A. thaliana chromosomes. The centromere is located between mi19 and mi291 in A. thaliana, suggesting that the chromosome break resulting in the different structures in A. thaliana and B. nigra took place outside these flanking markers. Similarly, on B. nigra G8, the homologous segment to A. thaliana chromosome 4 includes y29b and mi306a mapping closely together. Assuming that the mapping data in A. thaliana are correct, the homologues to these two loci flank the centromere.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Replication in Brassica genomes:
The A. thaliana genome is one of the smallest among higher plants, with an estimate of 145 million bp (Mbp, ARUMUGANATHAN and EARLE 1991 ) distributed on five chromosomes in the haploid complement. In the Brassicaceae family, chromosome numbers and DNA content show a large variation, from a low in A. thaliana to 19 chromosome pairs and 1235 million bp per haploid complement in B. napus. It has been suggested that the small genome size of A. thaliana results from an exceptionally low amount of repetitive DNA and high gene density (MEYEROWITZ 1992 ; BEVAN et al. 1998 ). The present study and other comparative mapping data now suggest that these attributes might be typical for most Brassicaceae species, and that the differences in genome content, to a large extent, are caused by different degrees of whole-genome replication. Comparative mapping among the three diploid species—B. nigra, B. oleracea, and B. rapa—supports the hypothesis that these genomes were descended from a hexaploid ancestor with three copies of a rearranged unit genome still discernible (LAGERCRANTZ and LYDIATE 1996 ). There is also good evidence that B. napus is an amphidiploid arisen from the hybridization of the two diploid species B. oleracea and B. rapa or close relatives (U 1935; PARKIN et al. 1995 ).

In the present study, the few large segments that have remained largely intact since the divergence of the ancestors of Brassica and Arabidopsis are present in three homeologous copies in B. nigra, but in a single copy in A. thaliana. In addition, there are a number of smaller segments of the A. thaliana genome that have three homeologous copies in B. nigra. Although it was not possible to detect three homeologous copies of every single segment of the A. thaliana genome, the present data support that the A. thaliana genome is similar in complexity to the triplicated unit genome of the diploid Brassica species. Furthermore, the B. nigra genome (0.97 pg/diploid nucleus; ARUMUGANATHAN and EARLE 1991 ) is approximately three times larger than that of A. thaliana (0.3 pg/diploid nucleus; ARUMUGANATHAN and EARLE 1991 ). Parallel comparisons of limited portions of the A. thaliana and B. napus genomes (SCHEFFLER et al. 1996 ; OSBORN et al. 1997 ; CAVELL et al. 1998 ) are congruent with the present data, showing an average of six homeologous copies of A. thaliana segments in the amphidiploid B. napus. KOWALSKI et al. 1994 also found evidence for triplication of some homeologous copies of A. thaliana segments in the B. oleracea genome, but a genome-wide triplication was not suggested. The lower marker density and limited polymorphism in this earlier study probably led to a systematic underestimation of genome replication in Brassica.

Assuming that the lineage leading to the present-day diploid Brassica species has indeed gone through a triplication of the genome, these replications must have been accompanied by a number of chromosome fusion events to reduce the chromosome number. It is not likely that the common ancestor of Arabidopsis and Brassica had a considerably lower number than A. thaliana. If the common ancestor also had 5 chromosomes and this genome was triplicated, the chromosome number had to be reduced from 15 to 8 in the lineage leading to B. nigra.

The present comparative mapping data support such a reduction in chromosome numbers through chromosome fusions. In B. nigra, homologues to loci that in A. thaliana are located close to a telomere often map internally on B. nigra linkage groups (Figure 3). B. nigra G8 comprises six such homologous loci, indicating six interstitial telomeric sites. Preliminary data (J. FAHLESON, T. AXELSSON and U. LAGERCRANTZ, unpublished results) indicate that at least some of these sites actually contain sequences hybridizing to the telomeric repeat from A. thaliana (RICHARDS and AUSUBEL 1988 ). Furthermore, comparative mapping of B. oleracea (n = 9), B. nigra (n = 8), and B. rapa (n = 10) support that changes in chromosome numbers caused by chromosome fusion or fission are frequent and also have occurred recently in the Brassicaceae family (LAGERCRANTZ and LYDIATE 1996 ).

Rapid chromosomal evolution in Brassica genomes:
A number of comparative analyses of genomes within the animal and plant kingdoms have suggested that the rate of chromosomal rearrangements is surprisingly low in most cases (NADEAU and TAYLOR 1984 ; O'BRIEN et al. 1988 ; MOORE et al. 1995 ). However, the present study indicates that the evolution of genomes in the Brassicaceae family involves an unusually high rate of chromosomal rearrangements. Different methods for estimating the amount of rearrangements since the divergence of Arabidopsis and Brassica lineages resulted in figures ranging from 79 to 112. To compare the A. thaliana-B. nigra data with previously published data, I used the estimate of 90 rearrangements obtained using the method of NADEAU and TAYLOR 1984 . The number of rearrangements in other species was also recalculated from published data using the same method (Table 3, Figure 4). This method has shown to be remarkably good, even with small sample sizes; modest genetic maps and possible mapping errors do not seem to influence results dramatically (EHRLICH et al. 1997 ).

View larger version (9K): In this window In a new window Download PPT slide   Figure 4. Estimated number of chromosomal rearrangements differentiating various animal species pairs and plant species pairs vs. approximate divergence times. All estimates are based on comparative linkage mapping. , comparisons between species in the Brassicaceae family; , comparisons including Mus; , all other comparisons. The straight line is estimated from linear regression excluding data from Brassicaceae () and Mus ().

Estimates of the divergence times between different species and genera in the Brassicaceae family vary widely. Divergence time ranging from 10 million years, based on paleopalynological data (MULLER 1981 ), to 35 million years, based on DNA sequence data from the rbcL gene (R. PRICE, personal communication), have been suggested for the lineages leading to Arabidopsis and Brassica. Even if we conservatively assume a divergence time of 35 million years, the A. thaliana-B. nigra comparison reveals an exceptionally high rate of chromosomal rearrangements (Table 3, Figure 4). Among previous comparisons, humans and mice are among those that have diverged most rapidly (EHRLICH et al. 1997 ). The estimate of the length of conserved segments from COPELAND et al. 1993 of 8.8 cM results in an estimate of 144 rearrangements between man and mouse (EHRLICH et al. 1997 ). Assuming a divergence time of rodents and primates of 114 million years (JANKE et al. 1994 ) yields an estimate of 0.63 rearrangements per million years. Even in comparison to the divergence of mouse and man, A. thaliana-B. nigra, with approximately 3 rearrangements per million years, stands out as having experienced an extreme rate of chromosomal repatterning.

KOWALSKI et al. 1994 obtained an estimate of at least 26 rearrangements since the divergence of A. thaliana and B. oleracea. This estimate is considerably lower than that obtained in the present study. A considerable part of this difference most likely results from a lower marker density in the previous study. Comparative mapping data between limited chromosomal regions of A. thaliana and B. napus support an extremely high rate of rearrangements for some chromosomal regions (OSBORN et al. 1997 ). A 14-cM tract on the B. napus linkage group N2 corresponds to segments from four different A. thaliana chromosomes. If such highly rearranged regions are common, as also indicated in the present study, a marker density of 15 cM [as for B. oleracea markers on the A. thaliana map of KOWALSKI et al. 1994 ] is likely to underestimate the true level of rearrangements.

Because of the poor data on divergence times in Brassica, it is not possible to conclude if the high rate of chromosomal divergence is typical of Brassicaceae species that have diverged more recently than A. thaliana and B. nigra. Sequence data from the large subunit of rubisco suggest that B. nigra diverged from B. rapa/oleracea ~20 mya and B. rapa diverged from B. oleracea 1 mya (R. PRICE, personal communication). If these data are reasonably correct, the rate of chromosomal rearrangements in these lineages does not seem to be higher that in most other plant and animal species (Table 3, Figure 4).

Why has the rate of chromosomal repatterning been so high between A. thaliana and B. nigra?
Population structure and recent polyploidization are probably important factors contributing to the rapid rearrangements of Brassicaceae chromosomes. The present study and other data (LAGERCRANTZ and LYDIATE 1996 ) suggest that modern diploid species of Brassica and related genera have descended from a common hexaploid ancestor and are, thus, degenerate hexaploids. It seems likely that the replications have occurred through amphidiploidization the same way as novel amphidiploids such as B. napus, B. juncea, and B. carinata are derived from hybridization between different diploids (U 1935; PARKIN et al. 1995 ; AXELSSON et al. 1998 ). Allopolyploidization is likely to result in an increase in aberrant meiotic pairing and translocations among homeologous chromosomes. Studies of resynthesized B. napus plants show that intergenomic translocations are surprisingly frequent between chromosomes from B. oleracea and B. rapa (LYDIATE et al. 1993 ; U. LAGERCRANTZ and D. LYDIATE, unpublished results).

Chromosome fusions after duplication also might have resulted indirectly in an increase of the frequency of rearrangements. As discussed above, polyploidization in the Brassica lineage has probably been followed by an extensive reduction in chromosome number through chromosome fusion events. These fusions have apparently resulted in interstitially located telomere repeats [ITRs (TTTAGGG)_n]. There are several independent data suggesting that such ITRs may be particularly prone to recombination, breakage, and fragility (HASTIE and ALLSHIRE 1989 ; MEYNE et al. 1990 ; BARNETT et al. 1993 ; BERTONI et al. 1994 ; SLIJEPCEVIC et al. 1996 ). The position of ITRs in the B. nigra genome (J. FAHLESON, T. AXELSSON and U. LAGERCRANTZ, unpublished results) supports their involvement in chromosomal rearrangements. The ITRs that have been positioned are almost exclusively located at the breakpoints between conserved blocks in the B. nigra genome.

Assuming no selective advantage of chromosomal rearrangements (LANDE 1979 ), the rate of chromosomal rearrangements depends on mutation rate and random fixation rate. The effect of replication provides an explanation as to the generation of a relatively high number of new rearrangements in Brassicaceae genomes. However, a key step in karyotype evolution is the fixation of newly arisen chromosomal rearrangements. Translocations and inversions are generally deleterious when heterozygous, but have normal fitnesses when homozygous (WHITE 1973 ). The fixation of such rearrangements requires small, isolated populations and is aided greatly by self-fertilization (LANDE 1979 ). Many wild Brassicaceae species occupy marginal fragmentated habitats, such as maritime cliffs (SNOGERUP et al. 1990 ; MITHEN et al. 1995 ). This distribution is likely to result in small deme sizes and high turnover rates of local populations. The exposed habitat is likely to lead to relatively frequent local extinction and recolonization. Local fixation is favored by small, reproductively isolated demes and selfing. Once established in a deme, a negatively heterotic gene arrangement can spread in homozygous form through a subdivided population by random local extinction and recolonization. Although the population structure of Brassicaceae species in the past is largely unknown, the present-day population structure of many species suggests an explanation as to the high rate of chromosomal rearrangements evident in some lineages of Brassicaceae. A combination of an enhanced level of chromosomal mutations caused by genome replication and a population structure characterized by small deme size may have contributed to the exceptionally high rate of rearrangements observed between A. thaliana and B. nigra.

Additional comparative mapping studies within the Brassicaceae, also including species closely related to A. thaliana, and more precise estimates of divergence times between species within Brassicaceae will shed more light on the rapid chromosomal evolution observed in the present study. Has rapid chromosome evolution been restricted mainly to the polyploid Brassica species, and were rearrangements mainly confined to a short period after polyploidization? Have rearrangement frequencies been higher in species where chromosome numbers have been reduced as a result of chromosomal fusions, or is the frequency of rearrangements mainly an effect of population structure?

Practical implications:
Obviously, the highly replicated nature of Brassica genomes must be acknowledged. It is likely that many important traits in Brassica species are controlled by duplicated genes originating from previous whole-genome replications. Identification of such duplicate genes would facilitate the understanding of the genetics and the improvement of various agronomic traits.

There are also good prospects for utilization of the rich source of biological information and genetic resources emanating from A. thaliana research. Even though the rearrangements have been frequent since the divergence of Arabidopsis and Brassica, the average length of conserved segments between A. thaliana and B. nigra was estimated at 8 cM. Thus, mapping a Brassica gene to an interval of <10 cM is often likely to allow the identification of the homeologous collinear region in A. thaliana. It should be kept in mind that some regions of the genome are considerably more rearranged, which will require much more detailed mapping. There is also a lack of data on the amount of fine-scale rearrangements that are not detected using comparative linkage mapping data. Such local rearrangements could obviously complicate the identification of homologous genes in A. thaliana and Brassica solely on the basis of their map position.

Still, with the prospect that a large proportion of the genes in A. thaliana will soon be identified, further fine-scale comparative mapping in the Brassicaceae family is likely to result in the identification of a large number of genes that affect important traits in different Brassica crops.

	ACKNOWLEDGMENTS

I thank L. Andersson, T. Axelsson, and two anonymous reviewers for valuable comments. I am also grateful to the Arabidopsis Biological Resource Center and the people listed in MATERIALS AND METHODS for kindly sending DNA clones. This work was supported by grants from the Swedish Natural Science Research Council and the Carl Trygger Foundation.

Manuscript received March 27, 1998; Accepted for publication July 24, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AHN, S. and S. TANKSLEY, 1993  Comparative linkage maps of the rice and maize genomes. Proc. Natl. Acad. Sci. USA 90:7980-7984[Abstract/Free Full Text].

AHN, S., J. ANDERSON, M. SORRELLS, and S. TANKSLEY, 1993  Homeologous relationships of rice, wheat and maize chromosomes. Mol. Gen. Genet. 241:483-490[Medline].

ARUMUGANATHAN, K. and E. D. EARLE, 1991  Nuclear DNA content of some important plant species. Plant Mol. Biol. Rep. 9:208-218.

AXELSSON, T., C. BOWMAN, D. LYDIATE and U. LAGERCRANTZ, 1998 RFLP mapping indicates conserved A and B genome linkage groups in Brassica juncea. Proceedings, Plant Genome VI, p. 133.

BARENDSE, W., S. ARMITAGE, L. KOSSAREK, A. SHALOM, and B. KIRKPATRICK et al., 1994  A genetic linkage map of the bovine genome. Nat. Genet. 6:227-235[Medline].

BARIOLA, P., C. HOWARD, C. TAYLOR, M. VERBURG, and V. JAGLAN et al., 1994  The Arabidopsis ribonuclease gene RNS1 is tightly controlled in response to phosphate limitation. Plant J. 6:673-685[Medline].

BARNETT, M., V. BUCKLE, E. EVANS, A. PORTER, and D. ROUT et al., 1993  Telomere directed fragmentation of mammalian chromosomes. Nucleic Acids Res. 21:27-36[Abstract/Free Full Text].

BERTONI, L., C. ATTOLINI, L. TESSERA, E. MUCCIOLO, and E. GIULOTTO, 1994  Telomeric and nontelomeric (TTAGGG)_n sequences in gene amplification and chromosome stability. Genomics 24:53-62[Medline].

BEVAN, M., I. BANCROFT, E. BENT, K. LOVE, and H. GOODMAN et al., 1998  Analysis of 1.9 Mb of contiguous sequence from chromosome 4 of Arabidopsis thaliana.. Nature 391:485-488[Medline].

BUSH, G., S. CASE, A. WILSON, and J. PATTON, 1977  Rapid speciation and chromosomal evolution in mammals. Proc. Natl. Acad. Sci. USA 74:3942-3946[Abstract/Free Full Text].

CAVELL, A., D. LYDIATE, I. PARKIN, C. DEAN, and M. TRICK, 1998  A 30 centimorgan segment of Arabidopsis thaliana chromosome 4 has six collinear homologues within the Brassica napus genome. Genome 41:62-69[Medline].

CHANG, C., J. L. BOWMAN, A. W. DE JOHN, E. S. LANDER, and E. M. MEYEROWITZ, 1988  Restriction fragment length polymorphism linkage map for Arabidopsis thaliana.. Proc. Natl. Acad. Sci. USA 85:6856-6860[Abstract/Free Full Text].

CHEN, Q. and K. ARMSTRONG, 1994  Genomic in situ hybridization in Avena sativa.. Genome 37:607-612.

COPELAND, N., N. JENKINS, D. GILBERT, J. EPPIG, and L. MALTAIS et al., 1993  A linkage map of the mouse: current applications and future prospects. Science 262:57-66[Abstract/Free Full Text].

ECKER, J., 1998  Genes blossom from a weed. Nature 391:438-439[Medline].

EHRLICH, J., D. SANKOFF, and J. NADEAU, 1997  Synteny conservation and chromosome rearrangements during mammalian evolution. Genetics 147:289-296[Abstract].

ELDRIDGE, M. and R. CLOSE, 1993  Radiation of chromosome shuffles. Curr. Opin. Genet. Dev. 3:915-922[Medline].

FABRI, C. and A. SCHAEFFNER, 1994  An Arabidopsis thaliana RFLP mapping set to localize mutations to chromosomal regions. Plant J. 5:149-156.

FEINBAUM, R. and F. AUSUBEL, 1988  Transcription regulation of the Arabidopsis thaliana chalcone synthase gene. Mol. Cell. Biol. 8:1985-1992[Abstract/Free Full Text].

GAUT, B. and J. DOEBLEY, 1997  DNA sequence evidence for the segmental allotetraploid origin of maize. Proc. Natl. Acad. Sci. USA 94:6809-6814[Abstract/Free Full Text].

GRAVES, J., 1996  Mammalians that break the rules: genetics of marsupials and monotremes. Annu. Rev. Genet. 30:233-260[Medline].

HASTIE, N. and R. ALLSHIRE, 1989  Human telomeres: fusion and interstitial sites. Trends Genet. 5:326-331[Medline].

HAUGE, B., S. HANLEY, S. CARTINHOUR, J. CHERRY, and H. GOODMAN, 1993  An integrated genetic/RFLP map of the Arabidopsis thaliana genome. Plant J. 3:745-754.

HELENTJARIS, T., D. WEBER, and S. WRIGHT, 1988  Identification of the genomic locations of duplicate nucleotide sequences in maize by analysis of restriction fragment length polymorphims. Genetics 118:353-363[Abstract/Free Full Text].

JANKE, A., G. FELDMAIER-FUCHS, W. KELLEY TOMAS, A. VON HAESELER, and S. PÄÄBO, 1994  The marsupial mitochondrial genome and the evolution of the placental mammals. Genetics 137:243-256[Abstract].

JELLEN, E., B. GILL, and T. COX, 1994  Genomic in situ hybridization differentiates between A/D- and C-genome chromatin and detects intergenomic translocations in polyploid oat species (genus Avena). Genome 37:613-618.

JOHANSSON, M., H. ELLEGREN, and L. ANDERSSON, 1995  Comparative mapping reveals extensive linkage conservation—but with gene order rearrangements—between the pig and the human genomes. Genomics 25:682-690[Medline].

KENTON, A., A. S. PAROKONNY, Y. Y. GLEBA, and M. D. BENNETT, 1993  Characterization of the Nicotiana tabacum L. genome by molecular cytogenetics. Mol. Gen. Genet. 240:159-169[Medline].

KONIECZNY, A. and F. AUSUBEL, 1993  A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J. 4:403-410[Medline].

KOWALSKI, S., T. LAN, K. FELDMANN, and A. PATERSON, 1994  Comparative mapping of Arabidopsis thaliana and Brassica oleracea chromosomes reveals islands of conserved organization. Genetics 138:499-510[Abstract].

LAGERCRANTZ, U. and D. LYDIATE, 1995  RFLP mapping in Brassica nigra indicates differing recombination rates in male and female meiosis. Genome 38:255-264.

LAGERCRANTZ, U. and D. LYDIATE, 1996  Comparative genome mapping in Brassica. Genetics 144:1903-1910[Abstract].

LAGERCRANTZ, U., J. PUTTERILL, G. COUPLAND, and D. LYDIATE, 1996  Comparative mapping in Arabidopsis and Brassica, fine scale genome collinearity and congruence of genes controlling flowering time. Plant J. 9:13-20[Medline].

LANDE, R., 1979  Effective deme size during long-term evolution estimated from rates of chromosomal rearrangements. Evolution 33:234-251.

LANDER, E., P. GREEN, J. ABRAHAMSON, A. BARLOW, and M. DALY et al., 1987  MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174-181[Medline].

LEITCH, I. and M. BENNETT, 1997  Polyploidy in angiosperms. Trends Plant Sci. 2:470-476.

LINCOLN, S., M. DALY and E. LANDER, 1992 Mapping Genes Controlling Quantitative Traits with MAPMAKER/QTL 1.1, Ed. 2. Whitehead Institute Technical Report, Cambridge, MA.

LISTER, C. and C. DEAN, 1993  Recombinant inbred lines for mapping RFLP and phenotypic markers in Arabidopsis thaliana.. Plant J. 4:745-750.

LIU, Y., N. MITSUKAWA, C. LISTER, C. DEAN, and R. WHITTIER, 1996  Isolation and mapping of a new set of 129 RFLP markers in Arabidopsis thaliana using recombinant inbred lines. Plant J. 10:733-736[Medline].

LYDIATE, D., A. SHARPE, U. LAGERCRANTZ, and I. PARKIN, 1993  Mapping the Brassica genome. Outlook Agric. 22:85-92.

MA, H., M. YANOFSKY, and E. MEYEROWITZ, 1990  Molecular cloning and characterization of GPA1, a G protein alpha subunit gene from Arabidopsis thaliana.. Proc. Natl. Acad. Sci. USA 87:3821-3825[Abstract/Free Full Text].

MASTERSON, J., 1994  Stomatal size in fossil plants: evidence for polyploidy in majority of angiosperms. Science 264:421-424[Abstract/Free Full Text].

MATSUI, M., S. SASAMOTO, T. KUNIEDA, N. NOMURA, and R. ISHIZAKI, 1989  Cloning of Ara, a putative Arabidopsis thaliana gene homologous to the Ras-related gene family. Gene 76:313-320[Medline].

MCGRATH, J. M., M. M. JANSCO, and E. PICHERSKY, 1993  Duplicate sequences with a similarity to expressed genes in the genome of Arabidopsis thaliana.. Theor. Appl. Genet. 86:880-888.

MEYEROWITZ, E., 1992 Introduction to the Arabidopsis genome, pp. 102–118 in Methods in Arabidopsis Research, edited by C. KONCZ, N. CHUA and J. SCHELL. World Scientific, Singapore.

MEYNE, J., R. BAKER, H. HOBART, T. HSU, and O. RYDER et al., 1990  Distribution of non-telomeric sites of the (TTAGGG)_n telomeric sequence in vertebrate chromosomes. Chromosoma 99:3-10[Medline].

MINDRINOS, M., F. KATAGIRI, G. YU, and F. AUSUBEL, 1994  The A. thaliana disease resistance gene RPS2 encodes a protein containing a nucleotide-binding site and leucine-rich repeats. Cell 78:1089-1099[Medline].

MITHEN, R., A. F. RAYBOLD, and A. GIAMOUSTARIS, 1995  Divergent selection for secondary metabolites between wild populations of Brassica oleracea and its implications for plant-herbivore interactions. Heredity 75:472-484.

MOORE, G., K. DEVOS, Z. WANG, and M. GALE, 1995  Grasses line up and form a circle. Curr. Biol. 5:737-739[Medline].

MORIZOT, D., 1994  Reconstructing the gene map of the vertebrate ancestor. Anim. Biotech. 5:113-122.

MULLER, J., 1981  Fossil pollen records of extant angiosperms. Bot. Rev. 47:1-142.

NADEAU, J. and B. TAYLOR, 1984  Lengths of chromosomal segments conserved since divergence of man and mouse. Proc. Natl. Acad. Sci. USA 81:814-818[Abstract/Free Full Text].

O'BRIEN, S., H. SEUÁNEZ, and J. WOMACK, 1988  Mammalian genome organization: an evolutionary view. Annu. Rev. Genet. 22:323-351[Medline].

OSBORN, T. C., C. KOLE, I. A. P. PARKIN, A. G. SHARPE, and M. KUIPER et al., 1997  Comparison of flowering time genes in Brassica rapa, B. napus and Arabidopsis thaliana.. Genetics 146:1123-1129[Abstract].

PARKIN, I., A. SHARPE, D. KEITH, and D. LYDIATE, 1995  Identification of the A and C genomes of amphidiploid Brassica napus (oilseed rape). Genome 38:1122-1131.

PATERSON, A., T. LAN, K. REISCHMANN, C. CHANG, and Y. LIN et al., 1996  Toward a unified genetic map of higher plants, transcending the monocot-dicot divergence. Nat. Genet. 14:380-382[Medline].

PRICE, R., J. PALMER and I. AL-SHEHBAZ, 1994 Systematic relationships of Arabidopsis: a molecular and morphological perspective, pp. 7–19 in Arabidopsis, edited by E. MEYEROWITZ and C. SOMERVILLE. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

PRINCE, J., E. POCHARD, and S. TANKSLEY, 1993  Construction of a molecular linkage map of pepper and a comparison of synteny with tomato. Genome 36:404-417.

REINISCH, A., J. DONG, C. BRUBAKER, D. STELLY, and J. WENDEL et al., 1994  A detailed RFLP map of cotton, Gossypium hirsutum x Gossypium barbadense: chromosome organization and evolution in a disomic polyploid genome. Genetics 138:829-847[Abstract].

RICHARDS, E. and F. AUSUBEL, 1988  Isolation of a higher eukaryotic telomere from Arabidopsis thaliana.. Cell 53:127-136[Medline].

ROUND, E., S. FLOWERS, and E. RICHARDS, 1997  Arabidopsis thaliana centromere regions: genetic map positions and repetitive DNA structure. Genome Res. 7:1045-1053[Abstract/Free Full Text].

SCHEFFLER, J., A. SHARPE, H. SCHMIDT, P. SPERLING, and I. PARKIN et al., 1996  Desaturase multigene families of Brassica napus arose through genome duplication. Theor. Appl. Genet. 94:583-591.

SHARROCK, R. and P. QUAIL, 1989  Novel phytochrome sequences in Arabidopsis thaliana: structure, evolution, and differential expression of a plant regulatory photoreceptor family. Genes Dev. 3:1745-1757[Abstract/Free Full Text].

SHIH, M., P. HEINRICH, and H. GOODMAN, 1991  Cloning and chromosomal mapping of nuclear genes encoding chloroplast and cytosolic glyceraldehyde-3-phosphate dehydrogenase from Arabidopsis thaliana.. Gene 104:133-138[Medline].

SHOEMAKER, R., T. OLSON and V. KANAZIN, 1996 Soybean genome organization: evolution of a legume genome, pp. 139–150 in Genomes of Plants and Animals: 21st Stadler Genetics Symposium, edited by J. P. GUSTAFSON and R. FLAVELL. Plenum Press, New York.

SLIJEPCEVIC, P., Y. XIAO, I. DOMINGUEZ, and A. NATARAJAN, 1996  Spontaneous and radiation-induced chromosomal breakage at interstitial telomeric sites. Chromosoma 104:596-604[Medline].

SLOCUM, M. K., S. S. FIGDORE, W. C. KENNARD, J. Y. SUZUKI, and T. C. OSBORN, 1990  Linkage arrangements of restriction fragment length polymorphism loci in Brassica oleracca. Theor. Appl. Genet. 80:57-64.

SNOGERUP, S., M. GUSTAFSSON, and R. VON BOTHMER, 1990  Brassica sect. Brassica (Brassicaceae). I. Taxonomy and variation. Willenowia 19:271-365.

SÖDERMAN, E. 1996 Genes encoding homeodomain-leucine zipper proteins in Arabidopsis thaliana. Ph.D. thesis, Uppsala University, Uppsala Sweden.

SONG, K., P. LU, K. TANG, and T. OSBORN, 1995  Rapid genome change in synthetic polyploids of Brassica and its implications for polyploid evolution. Proc. Natl. Acad. Sci. USA 92:7719-7723[Abstract/Free Full Text].

STAM, P., 1993  Construction of integrated genetic linkage maps by means of a new computer package: JoinMap. Plant J. 3:739-744.

SUN, T. and Y. KAMIYA, 1994  The Arabidopsis GA1 locus encodes the cyclase ent-kaurene synthetase A of gibberellin biosynthesis. Plant Cell 6:1509-1518[Abstract].

TANKSLEY, S., M. GANAL, J. PRINCE, M. DE VICENTE, and M. BONIERBALE et al., 1992  High density molecular maps of the tomato and potato genomes. Genetics 132:1141-1160[Abstract].

TANKSLEY, S., R. BERNATZKY, N. LAPITAN, and J. PRINCE, 1988  Conservation of gene repertoire but not gene order in pepper and tomato. Proc. Natl. Acad. Sci. USA 85:6419-6423[Abstract/Free Full Text].

Genomic analysis in Brassica with special reference to the experimental formation of B. napus and peculiar mode of fertilization. (1935) Jpn. J. Bot. 7:389-452.

WENDEL, J., 1989  New world cottons contain old world cytoplasm. Proc. Natl. Acad. Sci. USA 86:4132-4136[Abstract/Free Full Text].

WHITE, M., 197