Transposable elements in a marginal plant population: temporal fluctuations provide new insights into genome evolution of wild diploid wheat

RESEARCH Open Access Transposable elements in a marginal plant population: temporal fluctuations provide new insights into genome evolution of wild diploid wheat Alexander Belyayev 1* , Ruslan Kalendar 2 , Leonid Brodsky 1 , Eviatar Nevo 1 , Alan H Schulman 2,3* , Olga Raskina 1 Abstract Background: How new forms arise in nature has engaged evolutionary biologists since Darwin’s seminal treatise on the origin of species. Transposable elements (TEs) may be among the most important internal sources for intraspecific variability. Thus, we aimed to explore the temporal dynamics of several TEs in individual genotypes from a small, marginal population of Aegilops speltoides. A diploid cross-pollinated grass species, it is a wild relative of the various wheat species known for their large genome sizes contributed by an extraordinary number of TEs, particularly long terminal repeat (LTR) retrotransposons. The population is characterized by high heteromorphy and possesses a wide spectrum of chromosomal abnormalities including supernumerary chromosomes, heterozygosity for translocations, and variability in the chromosomal position or number of 45S and 5S ribosomal DNA (rDNA) sites. We propose that variability on the morphological and chromosomal levels may be linked to variability at the molecular level and particularly in TE proliferation. Results: Significant temporal fluctuation in the copy number of TEs was detected when processes that take place in small, marginal populations were simulated. It is known that under critical external conditions, outcrossing plants very often transit to self-pollination. Thus, three morphologically different genotypes with chromosomal aberrations were taken from a wild population of Ae. speltoides, and the dynamics of the TE complex traced through three rounds of selfing. It was discovered that: (i) various families of TEs vary tremendously in copy number between individuals from the same population and the selfed progenies; (ii) the fluctuations in copy number are TE-family specific; (iii) there is a great difference in TE copy number expansion or contraction between gametophytes and sporophytes; and (iv) a small percentage of TEs that increase in copy number can actually insert at novel locations and could serve as a bona fide mutagen. Conclusions: We hypothesize that TE dynamics could promote or intensify morphological and karyotypical changes, some of which may be potentially important for the process of microevolution, and allow species with plastic genomes to survive as new forms or even species in times of rapid climatic change. Background Populations are generally viewed as the elementary evolutionary unit [1,2]. A population exists as an integration of individuals in time and space that can vary over a set of successive generations. Under the influence of spontaneous mutations, each population becomes heterogeneous in its genetic structure over time. Thus, a population comprises a mix of different genotypes even if its individuals are more or less phenotypically similar. Under intensive pressure from a particular (usually environmental) factor, a common phenomenon in marginal populations, a shift in the genotypic structure of the population can occur. Such a shift may be regarded as the elementary event of microevolution. Transposable elements (TEs) may be among the most important internal sources for genotypic population * Correspondence: [email protected]aifa.ac.il; [email protected] 1 Institute of Evolution, University of Haifa, Mount Carmel, Haifa, Israel 2 MTT/BI Plant Genomics Laboratory, Institute of Biotechnology, Viikki Biocenter, University of Helsinki, Helsinki, Finland Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 © 2010 Belyayev et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. change as a result of their ability to create mutations, alter gene expression, and promote chromosomal aberrations [3-5]. These ancient, ubiquitous, and dynamic components of eukaryotic genomes comprise up to 80% of the large genomes of cereals [6,7]. Data accumulated over the last 30 years suggest that TEs can have major effects on genome organization and function [4,6,8,9]. Nevertheless, it is still unclear whether and to what degree the TEs contribute to evolutionarily significant shifts in the genotypic structure of populations. Population genetic theory assumes that the dynamics of transposable elements in natural populations reflect a balance between the tendency of these elements to increase through transposition and their removal through natural selection acting against individuals with a high element copy number [10,11]. Several studies from different biological systems indicate that the patterns of a single TE family can vary intraspecifically [12-15] and temporally [16]. Examination of the intraspecific dynamics of TEs, especially in marginal populations where microevolutionary processes are intensified [2], could considerably contribute to the understanding of key biological events such as speciation. One of the most effective ways to understand the dynamics of the TEs over time and space may be an ecological approach, whereby several generations of plants from natural populations are investigated. Here, we have chosen a small, marginal and threatened population of Aegilops speltoides (2n = 2× = 14) as a model for exploration of the dynamics of several TE families belonging to Classes I and II (respectively moving via RNA and DNA intermediates). The species is a wild relative of the various wheat species known for their large genome sizes [17,18] contributed to by an extraordinary number of TEs, particularly long terminal repeat (LTR) retrotransposons. Ae. speltoides was proposed to be the closest to the wild diploid progenitor of the G genomes and B genomes of polyploid wheat [19-21]. The study population is located on the western banks of the Kishon River (Haifa Bay area, Israel). The population is characterized by high heteromorphy and possesses a wide spectrum of chromosomal abnormalities, including supernumerary chromosomes, heterozygosity for translocations, and variability in the chromosomal position or number of 45S and 5S ribosomal DNA (rDNA) sites [22,23]. We propose that variability in morphological and chromosomal levels may point to variability at the molecular level and particularly in TE proliferation. Thus we aimed to detect and evaluate differences in TE copy numbers between individuals from the Kishon population and to trace inheritance of these deviations in successive generations. The broader goal of the study was to examine whether TE dynamics could be associated with morphological or karyotypical changes, some of which are potentially important for the evolutionary process. Results Experimental design Ae. speltoides is predominantly a cross-pollinated but self-compatible species [24]. Three original plants were selected from a small (approximately 100 m 2 ), marginal, degrade population of Ae. speltoides (see Methods). Each selected original genotype represents three groups of previously investigated plants (five to seven individual original spikes in each group), which have been clustered due to their morphological and cytogenetic similarity, including spike morphology, B chromosome existence, appearance of additional 5S rDNA chromosomal clusters, and specific chromosomal rearrangements [22,23]. The progeny from each genotype were obtained in three rounds of selfing. We simulated the situation in nature where, in marginal populations under critical external conditions, outcrossing plants and particularly Ae. speltoides very often transit to self-pollination [24-26]. The copy numbers ofseveralTEfamilieswere determined by quantitative polymerase chain reaction (qPCR) for each original genotype and its offspring. qPCR data were verified by dot blot analysis. The transpositional activities of TEs were inferred by inter-retrotransposon amplified polymorphism (IRAP) retrotransposon display [27]. Generational changes in the copy number of TE families were compared with those of the non-mobile, highly repetitive tandem repeat Spelt 52 and the 5S ribosomal RNA (rRNA) genes (rDNA). Dynamics of TE in successive generations The copy numbers of TE families that we investigated are presented in Additional file 1 and in graphic form in Figure 1a. The changes in TE copy number in a set of successive generations was measured by two different methods: qPCR and dot blot. These produced consistent results. The first salient point is that generative tissues, in most cases, possess higher numbers of TEs than do vegetative tissues. This observation is true for most TEs except those related to the Athila family (Sabrina and Wham). For example, the number of WIS2 retrotransposons in the spikes of G13 S1 exceeded that in leaves by 59%, or approximately 18,000 copies. The copy numbers of both the tribe-specific Spelt 52 tandem repeats and the 5S rRNA genes were also generally higher in spikes than in leaves (Additional file 1, Figure 1a). The second remarkable feature is that the copy numbers of all investigated TEs varied significantly over three successive generations, and that each genotype possessed individual TE dynamics over this period. The Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 2 of 16 TE copy numbers decreased or increased significantly, with a rise in one generation followed by a drop in the next and vice versa in an oscillatory fashion. In generative tissues, these oscillations displayed much higher amplitudes. Particular TEs decreased up to 23% from their initial quantity, as in the case of Daniela retrotransposons in G9 spikes where the copy number of this element dropped from approximately 62,000 copies in S1 to approximately 15,000 copies in S2 (Additional file 1, Figure 1a). We could not trace further TE dynamics in this genotype since the S2 generation was sterile. The largest absolute rise in TE number was documented also for Daniela in G13 spikes where the copy number rose from approximately 18,000 copies in S2 to approximately 52,000 copies in S3. The largest relative rise in TE abundance in one generation was seen for the Sabrina element, where in G14 its number in spikes increased by 672% in S3 as compared to S2. A similar jump in this genotype between S2 and S3 was observed also for WIS2,Wilma, and Wham. Genotypes 9 and 13 were distinct by the amplitudes in their numbers of copies for particular TE families over the generations analyzed, but the tendencies were the same. Common trends included a significant increase of the highly abundant retrotransposons, WIS2,Daniela, and Fatima, in the generative tissues of the S1 Figure 1 Copy numbers and morphology.(a) Dynamics of transposable element (TE) copy numbers in three self-pollinated generations of three genotypes from the Kishon population of Aegilops speltoides (shown by lines). TE copy numbers of the TS-84 population were used as controls. TE copy numbers for available sibs in selfing generations are shown by separate dots. (b) Changes in spike morphology in three selfpollinated generations of three genotypes from the Kishon population of Ae. speltoides. Spike morphology of plants from the TS-84 population was used as the control. Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 3 of 16 generation, followed by a similar decrease in these elements in the S2 generation (Additional file 1, Figure 1a). En/Spm-like transposons, Cassandra retrotransposons, and 5S rRNA genes exhibited similar dynamics in these genotypes. Wilma retrotransposons did not change in the S1 generation but then increased in number in S2. The copy number of Sukkula elements did not change in the S1 generation, but then dropped in S2. Sabrina and Wham elements remained approximately on the same level as in the original plants. The TE dynamics in vegetative tissues were approximately the same as in generative tissues but with less amplitude. For technical reasons, we do not have measurements of the copy number of Spelt 52 in the spikes of the original plants of G9, but the rest of the measurements showed a tendency for this tandem repeat to decrease in abundance in generative and vegetative tissues over the generations examined. Genotype 14 differed in TE copy number temporal dynamics from the two previously described genotypes. Cytogenetic data (see below) make it possible to propose that this genotype was already self-pollinated for at least one generation. WIS2, Wilma, Sabrina, Wham, and Sukkula elements demonstrated significant, successive decreases in copy numbers in S1 and S2 that were then followed by increases in S3. Daniela, En/Spm,andCassandra elements increased in copy number in S1 and successively decreased in S2 and S3. Fatima retrotransposons exhibited copy number dynamics in G14 similar to those seen in G9 and G13: increased abundance in S1, decreasing in S2, and again increasing in S3. In G14, too, the TE dynamics in vegetative tissues were approximately the same as for generative tissues, but with less amplitude. The number of Spelt 52 repeats, both in spikes and leaves, decreased in S1, increased in S2, and decreased in S3. In spite of a significant increase of copy number in S2, the overall tendency was for reduction. The 5S rRNA genes successively increased in copy number in S1 and S2, and then decreased in S3. To explore segregation in the progenies of a single genotype, TE copy numbers in sibs from G9 and G13 genotypes were determined (Table S1 in Additional file 2). Significant variations over three generations of two genotypes were observed. The amplitude of TE abundance in sibs could not be explained by simple chromosome segregation (see below) and, therefore, points to mobile element activity. A TE display method such as IRAP [27] would be expected to show polymorphisms consonant with large changes in TE copy number. Here, IRAP analyses were conducted on DNA from spikes at the microsporogenesis stage, and showed a high level of polymorphism from generation to generation (Figure 2). G13 showed the most unique bands, with retrotransposons WIS2 and Daniela producing the greatest number. Various forms of recombination may, of course, bring TEs sufficiently close that a new IRAP band would appear in the absence of a new integration event. For any given IRAP polymorphism, its origin as an integration event can be explicitly proven only by identifying an empty site, from a plant line missing the IRAP product, for one of the two TEs that together served as the template. To find evidence for individual transformation events, we cloned and sequenced 20 unique IRAP bands (Figure 3). At least half of the new insertion sites were in repetitive DNA, and therefore unsuited for identifying the original empty sites. Of the 20, (Table S2 in Additional file 2), 4 of them (a Daniela,aWis2,and2Sukkula insertions) were in non-repetitive sites that could be identified in a sequence database. The advent of new, unique, amplified IRAP bands that resulted from new TE insertions confirms the existence of transpositionally active TEs in the three genotypes of Ae. speltoides. Multivariate analyses The data distribution in principle component analysis (PCA) (Figure 4a) demonstrates the clear separation of leaves and spikes for each generation, normalized pairwise for each TE and non-TE repeat in control plants. The separation is along the first principal component (PC1), which covers 62% of initial data variability. The PC1 (Figure 4b) contrasts a group of TEs (Sukkula, Fatima, Cassandra, En/Spm, and Daniela) and two nonTE sequences (Spelt 52 and 5S rDNA) versus another group of TEs (Wilma, Sabrina, Wham,andWIS2). Therefore, the patterns of relative generational copy number variation of these two groups of TEs regarding their numbers in control plants are different in spikes and leaves. The first group of TEs and both non-TE sequences show larger fluctuations in leaves over successive generations, and the TEs of the second group show larger changes in spikes. The second principal component (PC2) represents the variability that is independent of the leaf-spike contrast (Figure 4b). When this component is further dissected by type of repeat (Figure 4c), the tissue-independent variance shows opposite trends for TE and non-TE repeats. This dichotomous trend holds across generations, genotypes, and tissues. We further investigated the nature of this trend for all nine TE families. Both raw copy numbers and those normalized by the control data were thus normalized by the number of one of the non-TE sequences (5S rDNA) and by control plants (TS-84) (Figure 4d). For both types of data normalization, the centralized patterns of change in copy number for the TEs and for Spelt 52 over the three genotypes and across tissues and generations demonstrate that: (i) the copy number fluctuation of non-TE sequence Spelt Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 4 of 16 52 with respect to 5S rDNA is insignificant across tissues and generations (broken red line) while fluctuations of TE sequences are high; (ii) the intergenerational pattern of copy number changes for all TEs is highly correlated; (iii) there is little relative TE copy number variation in leaves though it is very strong in spikes. It shows a minimum in the S2 generation, and a maximum in the S3 generation. In sum, the separation of TE and non-TE sequences in PC2 appears related to their different patterns of copy number dynamics across generations in spikes. Deviations of real copy numbers in leaves and spikes from the stochastic model The original plants of cross-pollinated Ae. speltoides are heterozygous for repetitive DNA sequences, chromosomal patterns, and chromosomal rearrangements (Figures 5 and 6). In order to separate the role of the pre-existing heterozygosity on the TE copy number in Figure 2 Inter-retrotransposon amplified polymorphism (IRAP) analyses for several transposable elements (TEs) in the progeny of three genotypes. Unique bands that appear in S2 and are inherited in S3 are shown by red arrows. An example of heterozygosity displayed in the IRAP pattern is shown in the blue square. An example of band loss in S2 and S3 is shown in the green square. An example of band appearance in S2 and S3 is shown in the yellow square. Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 5 of 16 the original plants from other mechanisms that can change TE amounts in successive generations, such as transposition, excision, or recombinational elimination, we created a ‘naïve’model in which the segregation of parental chromosomes in selfed progenies was simulated. The null hypothesis implies that: (i) cross-pollinated parental plants were initially heterozygous for the content of TEs in the homologues; (ii) only the random segregation of homologues in meiosis during male and female gametogenesis leads to a change in TE copy number in inbred offspring. For modeling we took only the G13 genotype as the genotype with a stable chromosome number, and only the dynamics of Class I elements were simulated. We carried out 100 realizations for stochastic time course simulations of copy numbers in spikes and leaves across ten generations. Thus, the copy numbers in leaves and spikes were calculated in 100 simulated siblings of each generation of plants, and these 100 simulated siblings from each of the first three plant generations form the generation-specific basic distributions of copy numbers in leaves and spikes. The deviations of copy numbers in real siblings from the basic simulated distributions, separately in leaves and spikes, test the null hypothesis. The significance of deviations in each generation was calculated by a t-test and the non-parametric MannWhitney U test (Table S7 in Additional file 2). The significance results of two tests are in good mutual concordance. Taking this into consideration, the F-test statistics were used for significance estimation of the cumulative deviation in positive or negative directions of real data from simulated siblings across three generations for each type of TE separately. The F statistic is: FW Wmax min / where W max and W min are either W p or W n depending on which one is larger. The statistic W p is defined as the sum of squares either of real t-test values, if t-test is Figure 3 Analysis of unique inter-retrotransposon amplified polymorphism (IRAP) bands.Inthisexample,IRAPwasconductedwith primer 2109, matching the Daniela retrotransposon long terminal repeat (LTR) (Table S2 in Additional file 2). The band was cut and sequenced. Two new primers were designed to match the sequence and the uniqueness of the band was checked on the set of DNA samples. This new insertion is in repetitive DNA (Table S2 in Additional file 2). Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 6 of 16 positive, or alternative expected positive t-tests (a small value for Q25 left quantile of t-test value distribution), if the real t-test is negative: W t Pra t Pra t Pra p test test test 1 2 2 2 3 2 (,) (, ) (, ) Similarly: W t Nra t Nra t Nra n test test test 1 2 2 2 3 2 (,) (, ) (, ) Statistics W p and W n are chi-square distributed with degrees of freedom equaling three each. Thus, the F-test has (3,3) degrees of freedom. The test shows how corroborative t-test directions and values are across three plant generations for the particular TE type. The + and -logPvalues of this F-test statistic across TE types are showninFigure4e.ThesignofthelogPvalue reflects the voting of t-tests either up (W max =W p ;W min =W n ) or down (W max =W n ;W min =W p )fromsimulations. Thus, long down bars indicate significant cumulative deviation of real data down from simulations across three plant generations. The graph (Figure 4e) shows that copy numbers of all TE types besides Daniela in leaves of G13 were significantly downregulated in relation to the simulations, but this is not true for spikes. The explanation could be that a high TE copy number in the male gametes of the original plant infers a generally high simulated level of copy numbers in spikes and leaves. However, a plant could survive only if the TE copy number is relatively low in diploid cells, permitting, however, a high copy number in gametes. Thus, plants with high copy numbers, both in gametes and diploid cells, will be eliminated from the population, providing the observed difference between real copy numbers in gametes and in diploid cells for survivors. Morphological and karyotypical changes in successive generations Ae. speltoides is a dimorphic species appearing in natural populations as two morphotypes or subspecies;Ae. speltoides ssp. ligustica and Ae. speltoides ssp. aucheri. The aucheri type is characterized by cylindrical spikes with widely interspersed spikelets. The apical spikelet is awned, and the rachis is tough. The ligustica type is characterized by denser, two-rowed ears in which the lateral spikelets are also awned and the rachis is brittle (Figure 1b). Significant morphological and karyotypic abnormalities in successive generations were observed, particularly: reduction of spike awns, appearance of intermediate ligustica-aucheri phenotypes, abnormal quantity of spikelets and florets, and spike size reduction (Figure 1b). Figure 4 Principle component analysis (PCA).(a) PCA analysis: spikes versus leaves for transposable element (TE) copy number changes over the generations, normalized by the TS-84 control. (b) PCA analysis: two groups of TE, spike active and leaf active. (c) Second principal component (PC2)-based order of TE activation. (d) Normalized and centralized patterns of the three genotypes for Spelt 52 and TEs across tissues and generations. (e) Deviations of real copy numbers in leaves and spikes from the stochastic model. Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 7 of 16 For karyotypic analysis we used several chromosomal markers in fluorescent in situ hybridization (FISH) experiments: 5S rDNA and 45S rDNA probes, cereal centromere-specific sequence 1 (CCS-1) [28], speciesspecific Spelt 1 [29] and tribe-specific Spelt 52 [30] tandem repeats. These probes were used for chromosome identification and localization of chromosomal rearrangements. The genomic formulas for the investigated plants were as follows: G9 (2n = 14 + 3B), G9 S1 to S2 (2n = 14 + 5B), G13, S1 to S3 (2n = 14), G14 (2n = 14 + 2B), G14 S1 (2n = 14 + 3B), G14 S2 (2n = 14 + 4B), and G14 S3 (2n = 14 + 3B) (Figures 5 and 6). Dynamics of Spelt 1, Spelt 52, and 5S rDNA clusters in successive generations revealed by FISH Cytogenetical analysis revealed dramatic differences between genome organizations of plants from the Kishon population in comparison with TS-84 plants representing the center of the Ae. speltoides distribution. First, Kishon plants lost a majority of species-specific tandem repeat Spelt 1, which appears usually as a component of terminal adenine and thymine (AT) -rich heterochromatic blocks in both arms of almost all chromosomes in TS-84 (Figure 6d). Only one or two terminal clusters of Spelt 1 per diploid genome were revealed in all studied genotypes, which means that up to 97% of the clusters were eliminated (Figure 5b, c, e and Figure 6c, Table 1). We observed a similar picture for Spelt 52 distal clusters: 57% to 82% were eliminated in genotypes from the Kishon population (Figures 5 and 6, Table 1). Thus, the amount of Spelt 1 is 13 times lower and Spelt 52 1.5 times lower than in the control genotype TS-84 from the center of the species Figure 5 Fluorescence in situ hybridization (FISH) and differential staining with 4’,6-diamidino-2-phenylindole (DAPI) on somatic and meiotic chromosomes of Aegilops speltoides (part 1).(a) FISH with 5S rDNA, 45S rDNA and staining with DAPI on somatic chromosomes of the original G9 plant (left). Chromosomes 1, 6 (arrows), and B chromosomes carry additional 5S rDNA sites (right). (b) FISH with Spelt 52, Spelt 1 (arrows on B chromosomes), 5S rDNA and 45S rDNA on the somatic chromosomes of the G9 S1 plant (left); FISH with 5S rDNA (right). Chromosomes 1 and 6 carry additional 5S rDNA sites (arrowed). (c) From left to right: FISH with 5S rDNA and 45S rDNA, and DAPI on the meiotic chromosomes of the G9 S2 plant; 5S rDNA probe alone, chromosomes 1, 6 (arrows), and B chromosomes carry additional 5S rDNA sites; FISH with Spelt 52 and Spelt 1 on the same chromosomes; FISH with CCS-1 and 45S rDNA (a pericentric inversion is arrowed); the scheme of the main chromosomal rearrangements (see legend). (d) DAPI (left) and FISH (middle) with 5S rDNA and 45S rDNA on the meiotic chromosomes of the G13 S2 plant. A scheme of the main chromosomal rearrangements (right). (e) FISH with Spelt 52 and Spelt 1 on the meiotic chromosomes of the G13 S3 plant (left). Small Spelt 52 cluster (arrow) marks paracentric inversion in the long arm of the chromosome 5. FISH with 5S rDNA and 45S rDNA with DAPI staining (middle). Both termini of chromosome 5 are involved in heterologous synapses (white arrows); heterozygous deletion on the chromosome 6 is shown by yellow arrow. Chromosomes 1 and 6 carry additional 5S rDNA clusters (arrows in right). The DNA probes and staining: (a-e) 5S rDNA, Spelt 52 and cereal centromere-specific sequence 1 (CCS-1) in red; 45S rDNA and Spelt 1 in green; differential staining with DAPI in blue; (b) 5S rDNA (yellow) and 45S rDNA (blue) in pseudocolors. Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 8 of 16 distribution. These data show that in the marginal Kishon population a high ratio of recombination occurs in terminal and distal chromosomal regions. Evidently, terminal clusters of Spelt 1 are the primary targets for deletions, followed by distally located Spelt 52 clusters. To the extent that chromosome termini have a similar structure (AT-rich heterochromatic bands enriched with clusters of Spelt 1 and/or Spelt 52 tandem repeats and TEs), they could be involved not only in homologous but also in heterologous recombination followed by chromosomal rearrangements. In the Kishon population we observed an irreversible elimination of terminal, species-specific Spelt 1 tandem repeats and a significant reduction of the distal, tribe-specific tandem repeat Spelt 52 as a result of high rates of homologous and heterologous recombination in this small inbred population. FISH experiments revealed that a significant increase or decrease in copy number might happen without change in the cluster quantity. For example, we traced a significant reduction in Spelt 52 copy number from the G9 original plant to G9 S2 (Figure 1a), while cluster numbers increase from 9 to 11 in S1 and drop again to 9inS2(Table1).InspikesofG14S29,200copiesof Spelt 52 are shared between 9 clusters while in S3 only 3,800 copies are assigned to 10 chromosomal clusters (Table 1 and Additional file 1). The dynamics of Spelt 52 clusters in G14 are as follows: the G14 original line, 10 clusters; G14 S1 to S8; G14 S2 to S9; and G14 S3 to S10 clusters per diploid genome. However, the chromosomal distribution of Spelt 52 clusters significantly differs between these genotypes. Thus, both homologues of chromosome 6 carry Spelt 52 clusters in the long arms of the original (Figure 6a) and S1 plants, while no clusters were detected (that is, all clusters have been deleted) as early as in S1 (not shown), S2, and S3 (Figure 6a and 6b). One homologue of chromosome 5 in the G14 original plant carries two clusters in the heterozygote in the long arm, where the second cluster is a result of translocation (Figure 6a). In S1 and S2 only one cluster still exists in the heterozygote, which provides evidence for deletion of a second cluster as a Figure 6 Fluorescence in situ hybridization (FISH) and differential staining with 4’,6-diamidino-2-phenylindole (DAPI) on somatic and meiotic chromosomes of Aegilops speltoides (part 2).(a) From left to right: DAPI and FISH with the CCS-1 and 45S rDNA on the meiotic chromosomes of the original G14 plant; FISH with Spelt 52 and Spelt 1. The clusters of Spelt 1 that mark a homozygous paracentric inversion on chromosome 4, and a Spelt 1 cluster on the B chromosome are arrowed. FISH with 5S rDNA and 45S rDNA on the same chromosomes. A scheme of the main chromosomal rearrangements. (b) FISH with Spelt 52 and Spelt 1 on the meiotic chromosomes of the G14 S2 plant (left). The clusters of Spelt 1 that mark a homozygous paracentric inversion on chromosome 4 and cluster of Spelt 52 that marks a heterozygous inversion on the chromosome 5 are arrowed. FISH with 5S rDNA and 45S rDNA (middle). B chromosomes carry 5S rDNA clusters in both arms. The scheme of the main chromosomal rearrangements (right). (c) FISH with Spelt 52 and Spelt 1 on the meiotic chromosomes of the G14 S3 plant (left). Homozygous paracentric inversion on chromosome 4 is arrowed. FISH with 5S rDNA and 45S rDNA on the same chromosomes (right). (d) Somatic chromosomes of TS 84, staining with DAPI (left). FISH with Spelt 52 and Spelt 1 on the same chromosomes (right). The DNA probes and staining: (a-d) 5S rDNA, Spelt 52 and cereal centromere-specific sequence 1 (CCS-1) in red; 45S rDNA and Spelt 1 in green; differential staining with DAPI in blue; (a) 5S rDNA (yellow) and 45S rDNA (blue) in pseudocolors. Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 9 of 16 25. Darwin CD: On the origin of species by means of natural selection, or preservation of favoured races in the struggle for life. London, UK: John Murray 1859. 26. Stebbins GL: Adaptive radiation of reproductive characteristics in angiosperms. 1. Pollination mechanisms. Ann Rev Ecol Syst 1970, 1:307-326. 27. Kalendar R, Schulman AH: IRAP and REMAP for retrotransposon-based genotyping and fingerprinting. Nat Protoc 2006, 1:2478-2484. 28. Aragon-Alcaide L, Miller T, Schwarzacher T, Reader S, Moore G: A cereal centromere sequence. Chromosoma 1996, 105:261-268. 29. Salina EA, Pestsova EG, Vershinin AV: Spelt1 - new family of cereal tandem repeats. Russian J Genet 1997, 33:352-357. 30. Anamthawat-Jonsson K, Heslop-Harrison JS: Isolation and characterization of genome-specific DNA sequences in Triticeae species. Mol Gen Genet 1993, 240:151-158. 31. Raskina O, Barber JC, Nevo E, Belyayev A: Repetitive DNA and chromosomal rearrangements: speciation-related events in plant genomes. Cytogenet Gen Res 2008, 120:351-357. 32. Belyayev A, Raskina O, Nevo E: Chromosomal distribution of reverse transcriptase containing retroelements in two Triticeae species. Chromosome Res 2001, 9:129-136. 33. Mendelson D, Zohary D: Behaviour and transmission of supernumerary chromosomes in Aegilops speltoides.Heredity 1972, 29:329-339. 34. Zarchi Y, Hillel J, Simchen G: Supernumerary chromosomes and chiasma distribution in Triticum speltoides.Heredity 1974, 33:73-180. 35. Puertas MJ: Nature and evolution of B chromosomes in plants: a noncoding but information-rich part of plant genomes. Cytogenet Gen Res 2002, 96:198-205. 36. Jones N, Houben A: B chromosomes in plants: escapees from the A chromosome genome?. Trends Plant Sci 2003, 8:417-423. 37. Devos KM, Brown JK, Bennetzen JL: Genome size reduction through illegitimate recombination counteracts genome expansion in Arabidopsis.Genome Res 2002, 12:1075-1079. 38. Sears E: Amphidiploids in the seven-chromosome Triticinae.Univ Missouri Agric Exper Sta Res Bull 1941, 336:1-46. 39. Danilevskaya O, Slot F, Pavlova M, Pardue ML: Structure of the Drosophila HeT-A transposon: a retrotransposon-like element forming telomeres. Chromosoma 1994, 103:215-224. 40. Castro JP, Carareto CM: Drosophila melanogaster P transposable elements: mechanisms of transposition and regulation. Genetica 2004, 121:107-118. 41. Turner JM, Mahadevaiah SK, Fernandez-Capetillo O, Nussenzweig A, Xu X, Deng CX, Burgoyne PS: Silencing of unsynapsed meiotic chromosomes in the mouse. Nat Genet 2005, 37:41-47. 42. Ansari KI, Walter S, Brennan JM, Lemmens M, Kessans S, McGahern A, Egan D, Doohan FM: Retrotransposon and gene activation in wheat in response to mycotoxigenic and non-mycotoxigenic-associated Fusarium stress. Theor Appl Genet 2007, 114:927-937. 43. Wessler SR: Turned on by stress. Plant retrotransposons. Curr Biol 1996, 6:959-961. 44. Kashkush K, Feldman M, Levy AA: Transcriptional activation of retrotransposons alters the expression of adjacent genes in wheat. Nat Genet 2003, 32:102-106. 45. Cheng C, Daigen M, Hirochika H: Epigenetic regulation of the rice retrotransposon Tos17.Mol Genet Genomics 2006, 276:378-390. 46. McClintock B: The significance of responses of the genome to challenge. Science 1984, 226:792-801. 47. Shirasu K, Schulman AH, Lahaye T, Schulze-Lefert P: A contiguous 66 kb barley DNA sequence provides evidence for reversible genome expansion. Genome Res 2000, 10:908-915. 48. Charlesworth D, Charlesworth B: Transposable elements in inbreeding and outbreeding populations. Genetics 1995, 140:415-417. 49. Lewis H: The mechanism of evolution in the genus Clarkia.Evolution 1953, 7:1-20. 50. Moeller DA, Geber MA: Ecological context of the evolution of selfpollination in Clarkia xantiana: population size, plant communities, and reproductive assurance. Evolution 2005, 59:786-799. 51. Kelly JK: Family level inbreeding depression and the evolution of plant mating systems. New Phytol 2005, 165:55-62. 52. Kristensen TN, Sorensen P, Pedersen KS, Kruhoffer M, Loeschcke V: Inbreeding by environmental interactions affect gene expression in Drosophila melanogaster.Genetics 2006, 173:1329-1336. 53. Grant-Downton RT, Dickinson HG: Plants, pairing and phenotypes - two’s company?. Trends Genet 2004, 20:188-195. 54. Rapp RA, Wendel JF: Epigenetics and plant evolution. New Phytol 2005, 168:81-91. 55. Liu ZL, Han FP, Tan M, Shan XH, Dong YZ, Wang XZ, Fedak G, Hao S, Liu B: Activation of a rice endogenous retrotransposon Tos17 in tissue culture is accompanied by cytosine demethylation and causes heritable alteration in methylation pattern of flanking genomic regions. Theor Appl Genet 2004, 109:200-209. 56. Husband BC: Chromosomal variation in plant evolution. Am J Bot 2004, 91:621-625. 57. Ohta T: Near-neutrality in evolution of genes and gene regulation. Proc Natl Acad Sci USA 2002, 99:16134-16137. 58. Lynch M, Conery JS: The origins of genome complexity. Science 2003, 302:1401-1404. 59. Flowers JM, Purugganan MD: The evolution of plant genomes: scaling up from a population perspective. Curr Opin Genet Dev 2008, 18:565-570. 60. REP database. http://wheat.pw.usda.gov/ggpages/ITMI/Repeats/index.shtml. 61. MULTILAN. http://npsa-pbil.ibcp.fr/cgi-bin/npsa_automat.pl?page=/NPSA/ npsa_multalinan.html. 62. FastPCR. http://www.biocenter.helsinki.fi/bi/programs/fastpcr.htm. 63. Belyayev A, Raskina O: Heterochromatin discrimination in Aegilops speltoides by simultaneous genomic in situ hybridization. Chromosome Res 1998, 6:559-565. doi:10.1186/1759-8753-1-6 Cite this article as: Belyayev et al.: Transposable elements in a marginal plant population: temporal fluctuations provide new insights into genome evolution of wild diploid wheat. Mobile DNA 2010 1:6. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Belyayev et al.Mobile DNA 2010, 1:6 http://www.mobilednajournal.com/content/1/1/6 Page 16 of 16