Population Dynamics of an Ac-like Transposable Element in Self- and Cross-Pollinating Arabidopsis

Corresponding author: Stephen I. Wright, Institute of Cell, Animal and Population Biology, University of Edinburgh, Ashworth Laboratories, King's Bldgs., W. Mains Rd., University of Edinburgh, Edinburgh, EH9 3JT Scotland., stephen.wright{at}ed.ac.uk (E-mail)

Communicating editor: O. SAVOLAINEN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Theoretical models predict that the mating system should be an important factor driving the dynamics of transposable elements in natural populations due to differences in selective pressure on both element and host. We used a PCR-based approach to examine the abundance and levels of insertion polymorphism of Ac-III, a recently identified Ac-like transposon family, in natural populations of the selfing plant Arabidopsis thaliana and its close outcrossing relative, Arabidopsis lyrata. Although several insertions appeared to be ancient and shared between species, there is strong evidence for recent activity of this element family in both species. Sequences of the regions flanking insertions indicate that all Ac-III transposons segregating in natural populations are in noncoding regions and provide no evidence for local transposition events. Transposon display analysis suggests the presence of slightly higher numbers of insertion sites per individual but fewer total polymorphic insertions in the self-pollinating A. thaliana than A. lyrata. Element insertions appear to be segregating at significantly lower frequencies in A. lyrata than A. thaliana, which is consistent with a reduction in transposition rate, reduction in effective population size, or reduced efficacy of natural selection against element insertions in selfing populations.

TRANSPOSABLE elements (TEs) are mobile self-replicating segments of DNA. Their ability to self-replicate and increase in abundance makes them an important source of spontaneous mutation and cause of genome evolution (KIDWELL and LISCH 2000 ). Population genetics theory predicts that the abundance of TEs in natural populations may be controlled by a balance between the forces of transposition increasing copy number and the action of purifying selection removing insertions from populations (CHARLESWORTH and LANGLEY 1989 ). However, the general importance of purifying selection in controlling TE dynamics and the nature of their deleterious effects in natural populations remain uncertain.

Evidence for the role of weak purifying selection in controlling transposon abundance in euchromatic regions of the genome was documented primarily in natural populations of Drosophila melanogaster, using the techniques of in situ hybridization (MONTGOMERY and LANGLEY 1983 ; CHARLESWORTH and LAPID 1989 ) and RFLP analysis (reviewed in CHARLESWORTH and LANGLEY 1989 ). From these studies TE families have been shown to be represented typically by a small number of copies per genome, despite evidence for much higher rates of transposition than excision. In addition, individual element insertions appear to be maintained at very low frequencies, their presence restricted to only one to a few individuals in population samples. This skewed frequency distribution shows a significant departure from neutral expectation (GOLDING et al. 1986 ; TAJIMA 1989 ) and is consistent with a model of transposition-selection balance, where new insertions are introduced through the transposition process and weak selection prevents them from rising to high frequencies (CHARLESWORTH and CHARLESWORTH 1983 ; CHARLESWORTH and LANGLEY 1989 ; CHARLESWORTH et al. 1992 ). While this model of transposition-selection balance has become well accepted for many families of transposons in D. melanogaster, the exact nature of purifying selection and copy number control on TEs remains an issue of contention (BIEMONT et al. 1997 ; CHARLESWORTH et al. 1997 ).

Although there is strong evidence indicating selective regulation of TE copy number in the large random-mating populations of D. melanogaster, very few population studies are available to examine its relevance in other eukaryotic genomes. In fact, preliminary analysis of retrotransposon families in several other animal species, including other species in the genus Drosophila (HEY 1989 ; BATZER et al. 1996 ; TAKASAKI et al. 1997 ), indicates that transposon insertions may often rise to high frequencies, and the accumulation of many ancient insertions in several taxa (e.g., MURATA et al. 1996 ; SANMIGUEL et al. 1998 ) is suggestive of a neutral process of fixation or loss of TE insertions over evolutionary time. This suggests the importance of detailed comparisons of the patterns of transposon insertion polymorphism in related species with contrasting life histories.

There has been recent theoretical interest in the role of the host breeding system in driving evolutionary changes in TE dynamics, particularly in plant populations (CHARLESWORTH and CHARLESWORTH 1995 ; WRIGHT and SCHOEN 1999 ; MORGAN 2001 ). If many transposon insertions cause recessive deleterious mutations (BIEMONT et al. 1997 ), then greater expression of these deleterious effects due to higher levels of homozygosity may drive transposon abundance and frequency to lower levels in selfing populations (WRIGHT and SCHOEN 1999 ; MORGAN 2001 ). In contrast, if element abundance is regulated by the dominant effects of ectopic recombination between elements (MONTGOMERY et al. 1991 ), the presence of most insertions in a homozygous state may reduce chances for ectopic pairing (MONTGOMERY et al. 1991 ), leading to a relaxation of selection on TE copy number and a potential for rapid increase in element abundance (CHARLESWORTH and CHARLESWORTH 1995 ; WRIGHT and SCHOEN 1999 ; MORGAN 2001 ). Reductions in effective population size in selfing populations, through frequent bottleneck events and/or strong selection at linked loci (MAYNARD SMITH and HAIGH 1974 ; CHARLESWORTH et al. 1993 ; BRAVERMAN et al. 1995 ), could also drive an increase in TE frequency and abundance (BROOKFIELD and BADGE 1997 ; WRIGHT and SCHOEN 1999 ; MORGAN 2001 ), but frequent stochastic loss of element families in selfing lineages is also a possible outcome (WRIGHT and SCHOEN 1999 ). Finally, differences in the breeding system are also expected to cause changes in the selective pressure on TEs themselves, leading to conditions favoring self regulation in genomes with low recombination (CHARLESWORTH and LANGLEY 1986 ), which may also cause reduced levels of insertion polymorphism.

Arabidopsis thaliana is a highly selfing (ABBOT and GOMES 1989 ) species with one of the smallest known genomes in higher plants (LEUTWILER et al. 1984 ). Recent surveys of the diversity and abundance of transposable elements in the Arabidopsis genome have revealed the presence of an extremely diverse array of transposon families, with most present at low copy number (SURZYCKI and BELKNAP 1999 ; LE et al. 2000 ). This low abundance of elements is a strong contrast to many plant species, for which element families may comprise as much as 50% of the genome (SANMIGUEL et al. 1996 ). Despite the low abundance, the high sequence similarity observed in several families of elements (SURZYCKI and BELKNAP 1999 ; LE et al. 2000 ), and the identification of putative empty insertion sites in related ecotypes, (LE et al. 2000 ) suggests that many of these families may have been recently active.

The hobo/Ac/Tam3 (hAT) superfamily of transposons is a widespread group of class II elements, which are known to be responsible for diverse morphological (COEN and CARPENTER 1986 ) and chromosomal (DOONER and BELACHEW 1991 ; SHALEV and LEVY 1997 ; ZHANG and PETERSON 1999 ) mutations. Several families of Ac-like elements have been identified in Arabidopsis, many of which show evidence for current activity (TSAY et al. 1993 ; FRANK et al. 1997 ) and/or recent historical transposition events (FRANK et al. 1998 ; HENK et al. 1999 ; LE et al. 2000 ). The Arabidopsis Ac-like III family (hereafter Ac-III), was recently identified in a survey of TE diversity in the A. thaliana (Columbia) genome project (LE et al. 2000 ). It was classified within the hAT superfamily on the basis of shared sequence and structural similarity of the terminal inverted repeats, the presence of several copies of a putative Ac transposase binding motif (TGGGC), and an 8-bp target site duplication (see HENK et al. 1999 ). Amplification of putative "empty" Ac-III insertion sites in several ecotypes related to Columbia suggested the possibility of recent mobility (LE et al. 2000 ).

In this study, we utilized an amplified fragment length polymorphism (AFLP)-based (VOS et al. 1995 ) technique, transposon display (KORSWAGEN et al. 1996 ; WAUGH et al. 1997 ; VAN DEN BROECK et al. 1998 ), to examine the distribution, abundance, and levels of insertion polymorphism of the Ac-III transposon family in natural populations of A. thaliana and its self-incompatible, highly outcrossing (KARKKAINEN et al. 1999 ) relative, A. lyrata.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Transposon display—element family and primer design:
The technique of transposon display involves the digestion of total genomic DNA with a frequent-cutting restriction enzyme, the ligation of adaptors to the digested fragments, and labeled PCR amplification of this digestion-ligation template using a TE-specific primer and an adaptor-specific primer. Since the restriction sites flanking transposon insertions are located at variable positions, this allows for the visualization of individual element insertions on a polyacrylamide gel, and variation among individuals in the sizes of these bands allows for an assessment of the patterns of transposon-based polymorphism in natural populations.

Sequence information for the Ac-III elements was accessed from the A. thaliana transposable element database at McGill University (http://www.tebureau.mcgill.ca; see LE et al. 2000 ), and several additional elements from recently released genomic clones were also identified through BLAST searches of GenBank (National Center for Biotechnology Information (NCBI); http://www.ncbi.nlm.nih.gov/blast/). All nine elements identified to date from the genome sequence data were aligned using the program Pileup (University of Wisconsin Genetics Computing Group, Version 10.0), and conserved subterminal regions were identified. Two overlapping degenerate primers were designed, on the basis of sequence alignments, for nested PCR using transposon display [ep1, 5' GGTTCGGTTA(A/T)TCGGTTAGC(G/T)G 3'; ep2, 5' G(C/A)TTCGGTTCGGTTA(A/T)TCGGTTAG 3']. By examining the frequent-cutter restriction map of the Ac-III elements, the four-cutter restriction endonuclease NlaIII was selected for use in transposon display, since no sites were observed between the primer sequence and the end of the element. Restriction mapping of A. thaliana chromosomes 2 and 4 suggested that the size distribution of NlaIII restriction fragments was suitable for transposon display (97% of restriction fragments fall between 4 and 1000 bp). An NlaIII adaptor was designed on the basis of the vectorette used by KORSWAGEN et al. 1996 (NlaIII 503, 5' CAAGGAGAGGACGCTGTCTGTCGAAGGTAAGGAACGGACGAGAGAAGGGAGA 3'; NlaIII 504, 5' TCTCCCTTCTCGAATCGTAACCGTTCGTACGAGAATCGCTGTCCTCTCCTTGCATG 3'), and two nested adaptor-specific primers were also constructed (ap1, 5' CGAATCGTAACCGTTCGTACGAGAATCGCT 3'; ap2, 5' GTACGAGAATCGCTGTCCTC 3'; KORSWAGEN et al. 1996 ). Using the available A. thaliana genomic sequence (93% of the total genome), the expected sizes of transposon display bands within the A. thaliana Columbia ecotype could be predicted by identifying the nearest NlaIII site to element insertions.

Population samples:
A total of 21 A. lyrata individuals and 29 A. thaliana individuals from a wide range of geographical locations were analyzed in this study (Table 1). The standard A. thaliana ecotype Columbia was also used as a positive control to confirm specificity of the technique. All A. lyrata samples were derived from maternal seed families that were field collected by various researchers. A. thaliana individuals were either obtained from field-collected seed, the progeny of field-collected seed that were selfed for one generation, or from the Arabidopsis Biological Resource Center (ABRC) or the Nottingham Arabidopsis Stock Center (NASC; see Table 1). One individual per seed family was analyzed for both species.

Transposon display analysis:
Genomic DNA was extracted using a modified version of DELLAPORTA et al. 1983 , with the addition of a phenol:chloroform:IAA extraction after the first precipitation. Genomic DNA (20–50 ng) was digested with 2 units of NlaIII, 1x NEB buffer 4 (New England BioLabs, Beverly, MA), and 1 mg/ml bovine serum albumin in a total volume of 10 µl for at least 2 hr. Digested DNA was then ligated to the vectorette overnight at 16°, using 5 units of T4 DNA ligase, 15 pmol adaptor, and 1x T4 DNA ligase buffer in a total volume of 25 µl. The ligation reaction was then diluted to a total volume of 100 µl for use as template in PCR. Preselective PCR amplification was carried out in a total volume of 25 µl, using 3 µl of the digestion-ligation mixture, 10 pmol of NlaIII adaptor-specific primer (ap1), and 10 pmol of element-specific primer (ep2), 1.5 mM MgCl₂, 0.2 mM dNTPs, and 1 unit Amplitaq DNA polymerase (Perkin-Elmer, Foster City, CA). PCR reactions were run on a Perkin-Elmer PCR model 9700 device for 20 cycles consisting of 1 min denaturing at 94°, 1 min annealing at 65°, and 1 min extension at 72°. This PCR reaction was diluted 1/100 and then used as template for selective PCR amplification under the same reaction conditions, except that 10 pmol of the nested element-specific primer and 2 pmol of the IRD700-labeled (LiCor, Lincoln, NB) nested adaptor-specific primer were used. Selective amplification products were then denatured for 5 min at 95°, rapidly cooled on ice, and run on a 66 cm, 5.5% denaturing polyacrylimide gel (3.75% acrylimide, 7 M urea, 1x Tris-borate EDTA), and bands were read using the LiCor DNA sequencer Long ReadIR4200 system. Bands were sized using the LiCor IRD700-labeled 50 to 700-bp molecular weight ladder, and their presence or absence was scored for each individual. Band scoring was checked with the assistance of the program Cross Checker (version 2.91, Dr. J. B. Buntjier, 1999). For a subset of samples, banding patterns were rechecked by repeating the process of digestion, ligation, and PCR. All amplifications were conducted twice from the same digestion-ligation reaction, and inconsistencies were examined. Several band sizes that were inconsistently present were excluded from subsequent analysis.

To assess the relative importance of insertion polymorphism vs. nucleotide and indel variation in producing polymorphic bands, selective amplifications were also run using the unlabeled adaptor-specific primer, and reaction products were cloned into the pCR 2.1 vector using the TA cloning kit (Invitrogen, Carlsbad, CA). Clones were then sequenced using the LiCor DNA sequencer Long ReadIR4200 according to the manufacturer's protocol. Sequences were analyzed by BLAST searches against GenBank (NCBI; http://www.ncbi.nlm.nih.gov/blast/) for sequence similarity and through multiple alignments using Pileup (University of Wisconsin Genetics Computing Group, version 10.0).

Data analysis:
A table of presence/absence for each insertion site was generated for analysis of patterns of insertion polymorphism. For all analyses, bands of the same size were assumed to be identical, while bands of distinct size were assumed to be independent insertions. Evidence supporting these assumptions, and the implications of any violations, are addressed in the DISCUSSION.

To test for a role for purifying selection in driving patterns of insertion polymorphism, the minimum ² method of CHARLESWORTH and CHARLESWORTH 1983 (Appendix 3) was applied. This method estimates the parameters of the probability distribution for element frequency x, using the ß-distribution

(1)

where = , ß = 4N_e( + s), N_e is the effective population size, u is the rate of transposition, T is the total number of insertion sites, is the equilibrium element copy number, is the rate of excision, and s is the strength of selection against element insertions. measures the effects of repeated transposition into the same insertion site, and ß provides an estimate of the strength of forces removing insertions from natural populations (CHARLESWORTH and LANGLEY 1989 ). If the number of sites is effectively infinite, Equation 1 reduces to

(2)

(CHARLESWORTH and CHARLESWORTH 1983 ). The parameter estimation method uses the estimated frequency distribution of individual insertion sites per haploid genome and finds the values of and ß, which minimize the deviation between observed and expected insertion frequencies. From in situ hybridization studies of Drosophila, insertion frequencies were estimated directly from haploid chromosomes (CHARLESWORTH and LANGLEY 1989 ). However, because the current study examines diploid chromosomes, and transposon display does not distinguish between heterozygous and homozygous insertions, estimates of allele frequencies, element copy number per haploid genome, and the insertion site frequency profile cannot be obtained directly from the banding patterns. In A. thaliana, estimates of outcrossing rates from natural populations indicate that populations are almost 100% selfing (ABBOT and GOMES 1989 ), and all insertions were thus assumed to be homozygous for the purposes of allele frequency and copy number estimation. Estimates of insertion site frequencies per haploid genome could then be estimated directly from transposon display. For A. lyrata, a modified version of Equation A9 in CHARLESWORTH and CHARLESWORTH 1983 was used to generate expected numbers of insertions present in m diploid genomes using the assumption of Hardy-Weinberg equilibrium and was compared to the observed values. Specifically, let g_i be the number of insertion sites present in i diploid individuals (i = 1, 2, ... m). The expected value of g_i is, assuming Hardy-Weinberg equilibrium and from CHARLESWORTH and CHARLESWORTH 1983 ,

(3)

T is then estimated in the usual way (CHARLESWORTH and CHARLESWORTH 1983 ) as T = , where is estimated by the sum of the Hardy-Weinberg estimates of insertion frequency,

(4)

and z_i is equal to the frequency of the "null" allele for the ith insertion site.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Consistency and specificity of transposon display in Arabidopsis:
Several lines of evidence indicate that the technique of transposon display provides a specific and repeatable visualization of Ac-III transposon insertions in Arabidopsis. First, virtually all insertion site sizes predicted from the available Arabidopsis genomic sequence were observed in analysis of the Columbia ecotype genomic DNA (Fig 1A). Only the largest predicted insertion site, 1444 bp, was not observed, suggesting the presence of an upper size limitation on the ability to amplify and visualize insertion sites. Estimates of numbers of insertion sites per individual should thus be considered as minimum estimates. However, given that analysis of NlaIII restriction fragment sizes in the Arabidopsis genome suggests that <5% of digested fragments should be >1 kb, this should not result in extreme underestimates. Second, replicated digestion-ligations from the same genomic DNA and repeated amplifications from the same digestion-ligation reaction resulted in consistent banding patterns (Fig 1A). Sequencing of a subset of amplification products from both species also allowed for confirmation that many observable bands were derived from Ac-III transposon insertions, through sequencing of the termini of the elements (e.g., Fig 1B). Although several amplification products that did not correspond to element insertions were also sequenced, these fragments did not appear on transposon display gels. Finally, in some cases, sequence similarity between the regions flanking the insertion and regions of the Columbia genome lacking the element provides evidence that novel bands correspond to new transposition events (e.g., Fig 1B).

View larger version (31K): In this window In a new window Download PPT slide   Figure 1. Characteristic results from the application of transposon display to the Ac-III family in Arabidopsis. (a) Transposon display patterns for four individuals from natural populations and the A. thaliana ecotype Columbia. Three replicates for each individual from natural populations are shown; the first two represent two different amplifications from the same digestion-ligation, and the third is derived from an independent digestion-ligation reaction. Lanes 1–3, A. lyrata GM-1; 4–6, A. lyrata GM-2; 7–9, A. thaliana Y2-1; 10–12, A. thaliana Y2-2. Lane 13, A. thaliana ecotype Columbia. Asterisks indicate the sizes of insertion sites predicted from the Columbia genome sequencing project. The triangle indicates the size of the insertion site for which the sequence is shown in b. Numbers represent the molecular weight, in base pairs. (b) Sequence analysis of cloned transposon display fragment. Top sequence: 5' end of cloned fragment. Middle sequence: 3' end of Ac-III transposon. Bottom sequence: region of Arabidopsis thaliana (Columbia) genome with high sequence similarity to cloned flanking region, with the accession number for the genomic clone given. ep1, element-specific primer used in transposon display; IR, inverted repeat.

Ac-III transposon insertions in Arabidopsis:
Amplification of the Ac-III element family was successful in all populations of both A. thaliana and A. lyrata analyzed in this study (Fig 2, Table 2). The vast majority of insertion sites identified in both species appeared to be polymorphic, although several bands of identical size appeared to be shared between both species, possibly representing ancient fixed insertions (Fig 2, Table 2). These latter sites were present only faintly and somewhat inconsistently in A. lyrata, while they were predominantly present as intense bands in A. thaliana. While a total of 46 polymorphic insertion sites were identified in A. lyrata, only 24 were identified in A. thaliana (Table 2). Despite the lower total number of polymorphic insertion sites, A. thaliana populations exhibited a slightly higher average number of insertion sites per individual and a higher estimated equilibrium number of insertions per individual, (Table 2).

View larger version (45K): In this window In a new window Download PPT slide   Figure 2. Example of among-population variability of transposon display patterns. (a) Arabidopsis lyrata. (b) A. thaliana.

In total the sequences of 15 (30%) insertion sites in A. lyrata and 6 (22%) in A. thaliana populations were determined, through the cloning and sequencing of amplification products from several individuals and from shared insertion sites to elements present in the Columbia genome project (Table 3). Nine of the sequenced flanking regions (57%) showed very high (90%) sequence similarity to genomic regions that were sequenced in A. thaliana. All of these insertion sites were in noncoding regions, including both introns and intergenic regions, with several inserted into other repetitive sequences, including a non-LTR retrotransposon (Table 3). All identified flanking regions appeared to be unique, suggesting that a significant proportion of distinct bands were the result of independent insertion events. Insertion sites showed similarity to physically distant regions across all five A. thaliana chromosomes, providing no evidence for local transposition events, which were observed for maize Ac-like transposons (e.g., BANCROFT and DEAN 1993 ).

Patterns of polymorphism:
In A. thaliana, the majority of insertion sites were present at intermediate to high frequency (Fig 2 and Fig 3). In contrast, a large fraction of the insertion sites were segregating at low frequency in A. lyrata, with many insertion sites present in only one or a few individuals (Fig 2 and Fig 3). To test for a significant difference among species in the insertion site frequency distribution, a 2 x 2 contingency table was constructed, similar to the approach of SAWYER et al. 1987 . This table compares the numbers of insertion sites unique to a single individual ("singletons") to higher frequency insertions between the two species and can be used to test the null hypothesis of an equal proportion of singleton insertions in each species. In A. lyrata, the number of singletons is 23, compared to 26 nonsingletons. In contrast, in A. thaliana, there are 3 singletons compared with 24 nonsingletons. Under Fisher's exact test, the contingency analysis shows a highly significant deviation from random expectation (P < 0.01), reflecting the much higher number of low-frequency insertions in the A. lyrata sample in comparison to A. thaliana.

View larger version (19K): In this window In a new window Download PPT slide   Figure 3. Frequency distributions of element insertions in (a) A. lyrata and (b) A. thaliana. Open bars, estimated insertion site frequency distribution using the transposon display patterns. Solid bars, predicted site occupancy profile assuming a ß-distribution, where the fixed insertion site class is excluded in the estimation. Hatched bars, predicted site occupancy profile assuming a ß-distribution, where the fixed insertion class is included. Estimations of the parameters of the ß-distribution are as described in the text.

To test for departures from neutral expectations of insertion frequency distributions, the parameters and ß were estimated for each species. For both data sets, was not significantly different from 0, and ² values increased as was increased from 0. This suggests that the number of insertion sites are effectively infinite, and estimates of ß were subsequently made under the assumption that = 0. Table 4 shows the estimates of ß and the corresponding minimum ² values for both species under two sets of conditions: (a) including and (b) excluding the potentially ancient high frequency insertions. Excluding these insertions is justified if they are in fact ancient and fixed, since they do not contribute to current population dynamics in each species. ² values were obtained after pooling classes for which expected values were <5. In all cases, the insertion site profiles do not show a significant deviation from expected by the minimum ² parameter estimates. Because of the small total number of insertion sites and the error associated with the estimation of the occupancy distribution from dominant markers, the estimates are subject to a high variance. However, under either scenario, the A. lyrata population data show a significant deviation from the hypothesis that = 0 and ß = 1, which is consistent with a significant role for purifying selection in driving patterns of insertion frequency. In contrast, for the A. thaliana insertion site distribution, the estimated minimum ² values of ß are <1, suggesting that the effects of genetic drift dominate over selection. Estimates of ß for A. lyrata when the potentially ancient insertions are excluded are of the same order of magnitude as was observed for several element families in D. melanogaster (CHARLESWORTH and LANGLEY 1989 ), suggesting that selection coefficients may be similar. Fig 3 shows a comparison of the expected insertion site occupancy profile under the estimated parameters to that calculated from the data.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Our study represents the first detailed investigation of the levels and patterns of insertion polymorphism of a class II transposon family in natural plant populations. The high levels of polymorphism of element insertions and the diversity of sequenced flanking regions strongly suggest that the Ac-III transposon family has been active in both species since their divergence. This allows for a detailed comparison of the population dynamics of the same transposon family in related selfing and outcrossing species.

The insertion site frequency distributions are consistent with the importance of the action of purifying selection on transposon insertions in outcrossing populations; segregating insertions in A. lyrata were present at lower frequencies than expected on the basis of neutral expectations. The few high-frequency bands identified in A. lyrata were of similar size to insertions identified in A. thaliana and may thus be potentially ancient. In contrast to the patterns observed in the outcrossing A. lyrata, most insertions were segregating at intermediate to high frequencies in the selfing A. thaliana and did not show evidence of a strong departure from neutral expectations. Virtually all insertions identified in a given population in A. thaliana were shared with other populations and present in other individuals from the same population. Although differences in element abundance between the species were less pronounced, there was a slightly higher number of insertion sites per individual in A. thaliana.

One limitation of our study concerns the population sampling: A. thaliana collections represent a more extensive worldwide sample, while A. lyrata is more concentrated, with more individuals sampled from fewer populations. This approach was taken because diversity is structured primarily between selfing populations, making it likely for most insertions to be fixed within populations. However, this contrasting population structure between selfers and outcrossers makes a direct test of population genetics models difficult, since demographic differences between the species may be substantial, and a worldwide sample from A. thaliana may represent a population that is out of equilibrium (SAVOLAINEN et al. 2000 ). Nevertheless, more extensive sampling of A. thaliana would be expected to inflate levels of transposon diversity and lower their average frequency, while the opposite pattern was observed in this study. Similarly, an alternative explanation to selection for the abundance of low-frequency insertions in A. lyrata could be the presence of independent transposition events among diverged populations. Detailed analysis of insertion polymorphism for a large within-population sample would provide an important confirmation of the hypothesis of purifying selection in A. lyrata. Once again, however, given that overall nucleotide divergence between A. thaliana populations appears to be as great as divergence in A. lyrata (SAVOLAINEN et al. 2000 ), the presence of population structure would be unlikely to explain the contrast between species.

The contrasting patterns of insertion polymorphism between the two species are consistent with the hypothesis that the breeding system contributes to the control of transposon dynamics. Several possible forces may be involved in driving this difference. First, a decrease in the abundance of the low-frequency class of insertions is consistent with a reduction in the effectiveness of purifying selection in selfing lineages (WRIGHT and SCHOEN 1999 ). Alternatively, the differences may be caused by a reduction in the transposition rate in A. thaliana. The population data indicate that A. thaliana populations show a reduced number of low-frequency insertions relative to A. lyrata, but there is not a significant increase in the number of high-frequency insertions, and the total number of insertion sites per individual does not differ dramatically between species (Table 2). Transposon display patterns also suggest a reduction in the species-wide number of polymorphic insertions in A. thaliana (Table 2), which has not been observed for nucleotide substitutions (SAVOLAINEN et al. 2000 ). Many or all of the Ac-III elements may lack coding capacity, and their transposition rates may thus be subject to control by other autonomous Ac-like elements. Stochastic loss of some of these autonomous elements in A. thaliana may also be an important force that might reduce transposition rates and drive down levels of insertion polymorphism. For nonautonomous elements, the assumption of a simple linear relationship between the number of elements and transposition rate may be violated.

The use of the technique of transposon display for surveying levels of transposon polymorphism has several weaknesses that may affect interpretation of the results. First, the inability to discriminate between heterozygous and homozygous insertions prevents direct estimation of the relevant population parameters. However, the use of Hardy-Weinberg assumptions for A. lyrata is conservative with respect to the hypothesis of the action of purifying selection, since deleterious selection would be expected to generate a lower frequency of homozygous insertions than predicted. The contrast between the species is also not an artifact of the assumption of complete homozygosity in Arabidopsis; even if the Hardy-Weinberg estimation of allele frequencies is made for the data from A. thaliana, no significant departure from neutral expectation is found (data not shown), and thus the contrast between species remains.

The second limitation of the transposon display technique is the possibility that some novel bands arise not from transposition events but from nucleotide substitution leading to restriction site variation or from indel variation. Some of the rare insertion sites may thus not derive from recent transposition events but from polymorphism generated by other classes of mutation. This potential problem is of particular concern with the A. lyrata data, where most insertions are present at low frequencies. However, given the sequence analysis of a significant fraction of the identified insertion sites, the small sizes of the regions amplified, and the relatively small number of insertion sites overall, variability in banding pattern is likely to arise primarily from insertion polymorphism. PCR amplification of many individual insertion sites, using primers specific to flanking regions, could allow for confirmation of this assumption as well as for a more direct examination of the levels of heterozygosity of insertion sites.

Our results provide support for the hypothesis of an effect of the selfing rate on driving transposon dynamics in natural populations. This suggests the importance of examining more element families and other related selfing and outcrossing species to investigate whether such patterns are in fact general and consistently associated with differences in the rate of self-fertilization. Comparisons of the transposition properties of element families is also useful to tease apart the role of selection on elements vs. their hosts in driving changes in population dynamics. Detailed comparisons of the within-vs. between-population patterns of insertion polymorphism are also important for distinguishing the importance of purifying selection and transposition rate. The Arabidopsis genomic sequencing project, as well as the one for rice, provides useful sources of sequence information on transposable elements, and this will allow for further investigation into plant transposon dynamics and the impact of transposons on mutation and evolution.

	ACKNOWLEDGMENTS

We thank Brian Charlesworth and Isabel Gordo for detailed advice and assistance with the data analysis; Chuck Langley for helpful discussion; Brian Charlesworth, Deborah Charlesworth, and Martin Morgan for comments on the manuscript; and Outi Savolainen, Chuck Langley, Massimo Pigliucci, Rodney Mauricio, and Thomas Mitchell-Olds for their generous contributions of seed material. This research was supported by National Sciences and Engineering Research Council (NSERC) operating grants to T.E.B. and D.J.S. and by an NSERC PGSA fellowship to S.I.W.

Manuscript received November 24, 2000; Accepted for publication April 11, 2001.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABBOT, R. J. and M. F. GOMES, 1989  Population genetic structure and outcrossing rate of Arabidopsis thaliana (L.) Heynh. Heredity 62:411-418.

BANCROFT, I. and C. DEAN, 1993  Transposition pattern of the maize element Ds in Arabidopsis thaliana.. Genetics 134:1221-1229[Abstract].

BATZER, M., S. S. ARCOT, J. W. PHINNEY, M. ALEGRIA-HARTMAN, and D. H. KASS et al., 1996  Genetic variation of recent Alu insertions in human populations. J. Mol. Evol. 42:22-29[Medline].

BIEMONT, C., A. TSITRONE, C. VIEIRA, and C. HOOGLAND, 1997  Transposable element distribution in Drosophila.. Genetics 147:1997-1999[Medline].

BRAVERMAN, J. M., R. R. HUDSON, N. L. KAPLAN, C. H. LANGLEY, and W. STEPHAN, 1995  The hitchhiking effect on the site frequency spectrum of DNA polymorphisms. Genetics 140:783-796[Abstract].

BROOKFIELD, J. F. Y. and R. M. BADGE, 1997  Population genetics models of transposable elements. Genetics 109:281-294.

CHARLESWORTH, B. and D. CHARLESWORTH, 1983  The population dynamics of transposable elements. Genet. Res. 42:1-27.

CHARLESWORTH, B. and D. CHARLESWORTH, 1995  Transposable elements in inbreeding and outbreeding populations. Genetics 140:415-417[Medline].

CHARLESWORTH, B. and C. H. LANGLEY, 1986  The evolution of self-regulated transposition of transposable elements. Genetics 112:359-383[Abstract/Free Full Text].

CHARLESWORTH, B. and C. H. LANGLEY, 1989  The population genetics of Drosophila transposable elements. Annu. Rev. Genet. 23:251-287[Medline].

CHARLESWORTH, B. and A. LAPID, 1989  A study of ten families of transposable elements on X chromosomes from a population of Drosophila melanogaster.. Genet. Res. 54:113-125[Medline].

CHARLESWORTH, B., A. LAPID, and D. CANADA, 1992  The distribution of transposable elements within and between chromosomes in a population of Drosophila melanogaster. I. Element frequencies and distribution. Genet. Res. 60(2):103-114[Medline].

CHARLESWORTH, B., M. T. MORGAN, and D. CHARLESWORTH, 1993  The effect of deleterious mutations on neutral molecular variation. Genetics 134:1289-1303[Abstract].

CHARLESWORTH, B., C. H. LANGLEY, and P. SNIEGOWSKI, 1997  Transposable element distributions in Drosophila.. Genetics 147:1993-1995[Medline].

COEN, E. S. and R. CARPENTER, 1986  Transposable elements in Antirrhinum majus: generators of genetic diversity. Trends. Genet. 2:292-296.

DELLAPORTA, S. L., J. WOOD, and J. B. HICKS, 1983  A plant DNA minipreparation: version II. Plant Mol. Biol. Rep. 1:19-21.

DOONER, H. K. and A. BELACHEW, 1991  Chromosome breakage by pairs of closely linked transposable elements of the Ac/Ds family in maize. Genetics 129:855-862[Abstract].

FRANK, M. J., D. LIU, Y.-F. TSAY, C. USTACH, and N. M. CRAWFORD, 1997  Tag1 is an autonomous transposable element that shows somatic excision in both Arabidopsis and Tobacco. Plant Cell 9:1745-1756[Abstract].

FRANK, M. J., D. PREUSS, A. MACK, T. C. KUHLMANN, and N. M. CRAWFORD, 1998  The Arabidopsis thaliana transposable element Tag1 is widely distributed among Arabidopsis ecotypes. Mol. Gen. Genet. 257:478-484[Medline].

GOLDING, G. B., C. F. AQUADRO, and C. H. LANGLEY, 1986  Sequence evolution within populations under multiple types of mutation. Proc. Natl. Acad. Sci. USA 83:427-431[Abstract/Free Full Text].

HENK, A. D., R. F. WARREN, and R. W. INNES, 1999  A new Ac-like transposon of Arabidopsis is associated with a deletion of the RPS5 disease resistance gene. Genetics 151:1581-1589[Abstract/Free Full Text].

HEY, J., 1989  The transposable portion of the genome of Drosophila algonquin is very different from that in D. melanogaster.. Mol. Biol. Evol. 6:66-79[Abstract].

KÄRKKÄINEN, K., H. KUITTINEN, R. VAN TREUREN, C. VOGL, and S. OIKARINEN et al., 1999  Genetics basis of inbreeding depression in Arabis petraea. Evolution 53:1354-1365.

KIDWELL, M. G. and D. R. LISCH, 2000  Transposable elements and host genome evolution. Trends Ecol. Evol. 15:95-99[Medline].

KORSWAGEN, H. C., R. M. DURBIN, M. T. SMITS, and R. H. A. PLASTERK, 1996  TransposonTc1-derived, sequence tagged sites in Caenorhabditis elegans as markers for gene mapping. Proc. Natl. Acad. Sci. USA 93:14680-14685[Abstract/Free Full Text].

LE, Q. H., S. WRIGHT, Z. YU, and T. BUREAU, 2000  Transposon diversity in Arabidopsis thaliana.. Proc. Natl. Acad. Sci. USA 97:7376-7381[Abstract/Free Full Text].

LEUTWILER, L. S., B. R. HOUGH-EVANS, and E. M. MEYEROWITZ, 1984  Mol. Gen. Genet. 194:15-23.

MAYNARD SMITH, J. and J. HAIGH, 1974  The hitch-hiking effect of a favorable gene. Genet. Res. 231:1114-1116.

MONTGOMERY, E. A. and C. H. LANGLEY, 1983  Transposable elements in Mendelian populations. II. Distribution of three copia-like elements in a natural population. Genetics 104:473-483[Abstract/Free Full Text].

MONTGOMERY, E. A., S.-M. HUANG, C. H. LANGLEY, and B. H. JUDD, 1991  Chromosome rearrangement by ectopic recombination in Drosophila melanogaster: genome structure and evolution. Genetics 129:1085-1098[Abstract].

MORGAN, M. T., 2001  Transposable element number in mixed mating populations. Genet. Res. in press.

MURATA, S. N., N. TAKASAKI, M. SAITOH, H. TACHIDA, and N. OKADA, 1996  Details of retrotranspositional genome dynamics that provide a rationale for a genetic division: the distinct branching of all the pacific salmon and trout (Oncorhynchus) from the atlantic salmon and trout (Salmo). Genetics 142:915-926[Abstract].

SANMIGUEL, P., A. TIKHONOV, J. YOUNG-KWAN, N. MOTCHOULSKAIA, and D. ZAKHAROV et al., 1996  Nested retrotransposons in the intergenic regions of the maize genome. Science 274:765-768[Abstract/Free Full Text].

SANMIGUEL, P., B. S. GAUT, A. TIKHONOV, Y. NAKAJIMA, and J. L. BENNETZEN, 1998  The paleontology of intergene retrotransposons of maize. Nat. Genet. 20:43-45[Medline].

SAVOLAINEN, O., C. H. LANGLEY, B. P. LAZZARO, and H. FREVILLE, 2000  Contrasting patterns of nucleotide polymorphism at the alcohol dehydrogenase locus in the outcrossing Arabidopsis lyrata and the selfing Arabidopsis thaliana.. Mol. Biol. Evol. 17:645-655[Abstract/Free Full Text].

SAWYER, S. A., D. E. DYKHUIZEN, and D. L. HARTL, 1987  Confidence interval for the number of selectively neutral amino acid polymorphisms. Proc. Natl. Acad. Sci. USA 84:6225-6228[Abstract/Free Full Text].

SHALEV, G. and A. A. LEVY, 1997  The maize transposable element Acinduces recombination between the donor site and an homologous ectopic sequence. Genetics 146:1143-1151[Abstract].

SURZYCKI, S. A. and W. R. BELKNAP, 1999  Characterization of repetitive DNA elements in Arabidopsis. J. Mol. Evol. 48:684-691[Medline].

TAJIMA, F., 1989  Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. Genetics 123:585-593[Abstract/Free Full Text].

TAKASAKI, N., T. YAMAKI, M. HAMADA, L. PARK, and N. OKADA, 1997  The salmon SmaI family of short interspersed repetitive elements (SINEs): interspecific and intraspecific variation of the insertions of SINEs in the genomes of chum and pink salmon. Genetics 146:369-380[Abstract].

TSAY, Y.-F., M. J. FRANK, T. PAGE, C. DEAN, and N. M. CRAWFORD, 1993  Identification of a mobile endogenous transposon in Arabidopsis thaliana.. Science 260:342-344[Abstract/Free Full Text].

VAN DEN BROECK, D., T. MAES, M. SAUER, J. ZETHO, and P. DE KEUKELEIRE et al., 1998  Transposon display identifies individual transposable elements in high copy number lines. Plant J. 13:121-129[Medline].

VOS, P., R. HOGERS, M. BLEEKER, M. REIJANS, and T. VAN DE LEE et al., 1995  AFLP: a new technique for DNA fingerprinting. Nucleic Acids Res. 23:4407-4414[Abstract/Free Full Text].

WAUGH, R., K. MCLEAN, A. J. FLAVELL, S. R. PEARCE, and A. KUMAR et al., 1997  Genetic distribution of BARE-1 retrotransposable elements in the barley genome revealed by sequence-specific amplification polymorphisms. Mol. Gen. Genet. 253:687-694[Medline].

WRIGHT, S. I. and D. J. SCHOEN, 1999  Transposon dynamics and the breeding system. Genetica 1(3):139-148.

ZHANG, J. and T. PETERSON, 1999  Genome rearrangements by nonlinear transposons in maize. Genetics 153:1403-1410[Abstract/Free Full Text].

This article has been cited by other articles:

				A. Kawabe, A. Forrest, S. I. Wright, and D. Charlesworth High DNA Sequence Diversity in Pericentromeric Genes of the Plant Arabidopsis lyrata Genetics, June 1, 2008; 179(2): 985 - 995. [Abstract] [Full Text] [PDF]

				K. M. Hazzouri, A. Mohajer, S. I. Dejak, S. P. Otto, and S. I. Wright Contrasting Patterns of Transposable-Element Insertion Polymorphism and Nucleotide Diversity in Autotetraploid and Allotetraploid Arabidopsis Species Genetics, May 1, 2008; 179(1): 581 - 592. [Abstract] [Full Text] [PDF]

				B. S. Gaut and J. Ross-Ibarra Selection on Major Components of Angiosperm Genomes Science, April 25, 2008; 320(5875): 484 - 486. [Abstract] [Full Text] [PDF]

				J. D. Hollister and B. S. Gaut Population and Evolutionary Dynamics of Helitron Transposable Elements in Arabidopsis thaliana Mol. Biol. Evol., November 1, 2007; 24(11): 2515 - 2524. [Abstract] [Full Text] [PDF]

				R. A. Subramanian, P. Arensburger, P. W. Atkinson, and D. A. O'Brochta Transposable Element Dynamics of the hAT Element Herves in the Human Malaria Vector Anopheles gambiae s.s. Genetics, August 1, 2007; 176(4): 2477 - 2487. [Abstract] [Full Text] [PDF]

				D. R. Hoen, K. C. Park, N. Elrouby, Z. Yu, N. Mohabir, R. K. Cowan, and T. E. Bureau Transposon-Mediated Expansion and Diversification of a Family of ULP-like Genes Mol. Biol. Evol., June 1, 2006; 23(6): 1254 - 1268. [Abstract] [Full Text] [PDF]

				J. L. MacKenzie, F. E. Saade, Q. H. Le, T. E. Bureau, and D. J. Schoen Genomic Mutation in Lines of Arabidopsis thaliana Exposed to Ultraviolet-B Radiation Genetics, October 1, 2005; 171(2): 715 - 723. [Abstract] [Full Text] [PDF]

				D. C. ALBACH and J. GREILHUBER Genome Size Variation and Evolution in Veronica Ann. Bot., December 1, 2004; 94(6): 897 - 911. [Abstract] [Full Text] [PDF]

				D. E. Neafsey, J. P. Blumenstiel, and D. L. Hartl Different Regulatory Mechanisms Underlie Similar Transposable Element Profiles in Pufferfish and Fruitflies Mol. Biol. Evol., December 1, 2004; 21(12): 2310 - 2318. [Abstract] [Full Text] [PDF]

				C. Rizzon, E. Martin, G. Marais, L. Duret, L. Segalat, and C. Biemont Patterns of Selection Against Transposons Inferred From the Distribution of Tc1, Tc3 and Tc5 Insertions in the mut-7 Line of the Nematode Caenorhabditis elegans Genetics, November 1, 2003; 165(3): 1127 - 1135. [Abstract] [Full Text] [PDF]

				S. I. Wright, N. Agrawal, and T. E. Bureau Effects of Recombination Rate and Gene Density on Transposable Element Distributions in Arabidopsis thaliana Genome Res., August 1, 2003; 13(8): 1897 - 1903. [Abstract] [Full Text] [PDF]

C. Bartolome, X. Maside, and B. Charlesworth On the Abundance and Distribution of Transposable Elements in the Genome of Drosophila melanogaster Mol. Biol. Evol., June 1, 2002; 19(6): 926 - 937. [Abstract] [Full Text] [PDF] <