MBE Advance Access originally published online on November 13, 2007
Molecular Biology and Evolution 2008 25(1):229-237; doi:10.1093/molbev/msm250
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Articles |
Rapid R2 Retrotransposition Leads to the Loss of Previously Inserted Copies via Large Deletions of the rDNA Locus
Department of Biology, University of Rochester
E-mail: eick{at}mail.rochester.edu.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
R2 non–long terminal repeat retrotransposable elements insert specifically into the 28S rRNA genes of a wide range of animals. These elements maintain long-term stable relationships with the host genome. By scoring the variation present at the 5' ends of individual R2 copies, lines of Drosophila simulans have been identified with high rates of R2 retrotransposition. Comparing the R2 elements present in the parents with that of their progeny after 1 or 30 generations in this report revealed that retrotransposition rates were higher through the female germ line compared with the male germ line. In addition, most events in females occur late in germ line development. Surprisingly, the gain of new R2 insertions by retrotranspositions was counterbalanced by deletions of preexisting R2 insertions. These deletions occurred by the loss of large segments of the rDNA units that contained on average an estimated 15 R2 elements. When monitored over single generations, the rate of loss of preexisting elements was higher than the rate of new insertions. However, the chromosomes with the largest deletions appear to be eliminated from the population because the rates of R2 insertions and deletions after 30 generations were approximately equal. These findings suggest that high rates of R2 retrotransposition do not necessarily lead to dramatic increases in the level of R2 insertions in the rDNA locus but can lead to a more rapid turnover of rDNA units.
Key Words: retrotransposable element • rates of retrotransposition • D. simulans • rDNA locus
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Transposable elements comprise significant proportions of most eukaryotic genomes. These elements have long been suggested to be parasites or junk DNA, although numerous instances of the beneficial effects of specific insertions have been identified in numerous organisms, including Drosophila (Lin et al. 1998
; Aminetzach et al. 2005). The level of transposable elements in a genome is believed to be a balance between the efficiencies of the various element transposition machineries, attempts by the host to control their transposition, and the elimination of elements by selection and ectopic recombination (Charlesworth et al. 1994). Studies of naturally occurring transposition in Drosophila laboratory stocks have revealed element insertion rates less than 10–4 transposition events per copy per generation, suggesting that transposable elements are usually under rigid host control (Biemont et al. 1994; Nuzhdin and Mackay 1995; Maside et al. 2001; Perez-Gonzalez and Eickbush 2002). However, occasional stocks with elevated transposition rates have been identified that are correlated with dramatic increases in the number of elements within the genome (Kim et al. 1990; Pasyukova and Nuzhdin 1993; Desset et al. 1999).
R2 retrotransposable elements insert exclusively into the 28S rRNA genes of eukaryotes (Eickbush 2002). The host 28S genes are tandemly repeated hundreds of times within one or more ribosomal RNA (rDNA) loci. Both the total number of rDNA units within the host and the fraction of these units inserted with R2 can vary several fold within populations (Lyckegaard and Clark 1991; Jakubczak et al. 1992; Zhang and Eickbush 2005). R2 elements are abundant in most lineages of arthropods (Jakubczak et al. 1991; Burke et al. 1998) as well as many other animal phyla (Kojima and Fujiwara 2005; Kojima et al. 2006). Phylogenetic studies of R2 elements have found no evidence of transfers between species, suggesting that they are stably maintained within a lineage by vertical transmission (Eickbush and Eickbush 1995; Burke et al. 1998; Malik et al. 1999; Gentile et al. 2001; Kojima et al. 2006). This long-term stability of R2 elements is remarkable given the rapid turnover of rDNA units by recombination within the rDNA locus (Averbeck and Eickbush 2005) and the selective pressure against R2 insertions interrupting the ability of rDNA units to synthesize 28S RNA (Eickbush and Eickbush 2003).
We have previously reported that isofemale lines of Drosophila simulans originally derived from a single natural population showed a wide range of R2 retrotransposition levels (Zhang and Eickbush 2005). Lines with high and low rates of R2 retrotransposition appeared stable for many years; yet, the levels of R2 elements in the active lines were only about twice that observed in the inactive lines. The total sizes of their rDNA loci (i.e., number of units) were also similar between the active and inactive lines (Zhang and Eickbush 2005). In this report, we further study the dynamics of R2 within a locus by monitoring the direct effects of R2 retrotransposition on the rDNA locus in single generations as well as the accumulating affects of R2 retrotransposition after 30 generations. We find that the rapid rates of R2 retrotransposition in the female germ line lead to the rapid loss of R2 elements already present in the rDNA locus. The net effect of high R2 activity in the rDNA locus is therefore a rapid turnover of the rDNA units, but the size and composition of the rDNA locus remains relatively constant.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Fly Stocks and Crosses
The D. simulans lines (Zhang and Eickbush 2005
) were originally collected by M. Turelli from Paradise, CA. The 2 most active lines identified in the previous study were used here (lines 58 and 89). The attached–X chromosome stock (C(1)RM,yw) was a gift from H. A. Orr.
The 5'-Truncation Profiles and Scoring New R2 Insertions and Deletions
In D. simulans, rDNA units only exist on the X chromosome thus male flies have a single rDNA locus. To score female germ line events, single females were crossed with single males and the R2 5'-truncation profiles of the sons were compared with their mother's. To score male germ line events, single males were crossed with multiple attached–X (X XY) females and the R2 5'-truncation profiles of the sons were compared with their father's. The 5'-truncated R2 elements were amplified by polymerase chain reaction (PCR) with a forward primer complementary to the 28S sequences 80 bp upstream of the R2 insertion site and reverse primers complementary to regions 0.8 kb, 1.0 kb, 1.3 kb, 1.6 kb, 2.0 kb, 2.4 kb, 2.8 kb, 3.2, and 3.6 kb from the 5' end of the R2 element. The primer sequences were reported in Zhang and Eickbush (2005). PCR products were separated on 8.75% polyacrylamide gels. These gels enabled the resolution of bands that differed by more than 5 bp in length. To be scored as a new R2 insertion or the deletion of a parental R2 copy, variant PCR bands seen in the sons had to be confirmed by at least 2 adjacent primers in combination with the upstream primer.
To score retrotransposition and deletion events after 30 generations, sublines of line 58 were established from single-pair matings and subsequently maintained by mass mating. Randomly selected males from the 30th generation were compared with the founding pair of the sublines. The 5'-truncation profiles of each of the 2 chromosomes in the female of the founding pair were determined by monitoring the profiles of her sons.
Rates of R2 Insertion and Deletion
The rate of R2 insertion per chromosome per generation was calculated as the total number of new R2 5' truncations observed in the sons divided by the number of X chromosomes (sons) scored and the number of generations. The rate of R2 elimination per chromosome per generation was calculated as the total number of ancestral R2 insertions that were found missing divided by the number of X chromosomes (sons) scored and the number of generations.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
R2 elements are members of the non–long terminal repeat (non-LTR) class of retrotransposons (also known as Long Interspersed Nucleotide Elements [LINEs]). As is characteristic of this class of elements, all R2 elements within an organism have nearly identical 3'-end (junction) sequences, whereas their 5' junctions often contain deletions (truncations) from a few base pairs to nearly the entire length of the element (see fig. 1). Most of these 5'-truncated elements are present at 1 copy per rDNA locus, suggesting that few copies are duplicated by recombination within the rDNA locus (Perez-Gonzalez and Eickbush 2001
, 2002; Averbeck and Eickbush 2005; Zhang and Eickbush 2005). We have previously described a PCR approach to score the entire collection of 5'-truncated R2 elements from individual flies (i.e., their 5'-truncation profiles) (Perez-Gonzalez and Eickbush 2001; Zhang and Eickbush 2005). In these assays, a primer that anneals to the upstream 28S gene sequence is paired with a series of primers that anneal to equally spaced regions along the R2 element (fig. 1). New R2 insertions as well as the deletion of previous insertions can be scored by comparing the 5'-truncation profiles of the offspring to that of their parents. Drosophila simulans is especially convenient for such studies because the rDNA units are exclusively located on the X chromosome (Lohe and Roberts 1990). Thus, D. simulans males contain a single rDNA locus with its associated R2 elements.
|
In the D. simulans lines used for this study, 58 and 89, about one-third of the R2 elements contain 5' truncations (Zhang and Eickbush 2005
). The remaining two-thirds of the R2 elements in these lines have lengths at or near that of a full-length R2 element. Although the insertion of "full-length" elements can sometimes be scored (see fig. 4 in Zhang and Eickbush 2005), most full-length insertions are missed because they give rise to PCR products of the same length as preexisting R2 copies. Therefore, scoring changes in the patterns of 5'-truncated R2 elements represent the most reliable method to compare insertion and deletion rates.
|
Rates of R2 Retrotransposition and Deletion through the Male and Female Germ Lines
To score R2 changes through the female germ line, the R2 5'-truncation profile of the mother was compared with the profiles of her sons. To score R2 activity in the male germ line, individual males were crossed to multiple attached–X (X XY) females. All males from these crosses had obtained their X chromosome from their father; therefore, the 5'-truncation profiles of the sons were compared with that of their father.
Figure 2 shows examples of the 3 types of R2 changes we detected in the female germ line. In each panel, lane 1 corresponds to a localized region of a polyacrylamide gel containing the PCR products of the mother, whereas lanes 2–8 correspond to the PCR products obtained from 7 randomly selected sons. Because the R2 profiles in the 58 and 89 lines are highly variable (Zhang and Eickbush 2005), the 2 X chromosomes of the mother may not be identical. In these situations, the mother's profile was represented by 2 chromosomes (a and b), whereas her sons have either the a or b chromosome. In panel A, 1 son (lane 3) was found to have an additional PCR product not seen in the mother. Such new PCR bands identified in the sons were interpreted as retrotransposition events occurring in the germ line of the mother, rather than somatic events in the son, because of their similar intensity to the common bands shared by mother and son, and our inability to directly score somatic events using DNA from isolated tissues of an animal (data not shown). Panel B is an example of an R2 deletion that occurred in the germ line of the mother. The top most PCR band found in the mother and 6 of 7 sons is missing in 1 son (lane 2). Finally, panel C shows an example where more than one son had a new R2 insertion of similar length (lanes 2 and 8). The presence of new R2 insertions of similar length in more than one son was rare. Because our resolution between different length R2 insertions was 5 bp, and independent retrotransposition events can give rise to 5' truncations of identical length, even these rare insertions shared by 2 sons may represent independent insertions. Only in 2 instances were the same R2 insertion observed in larger numbers of sons.
|
As summarized in table 1, R2 insertions and deletions were investigated in the germ line of 8-line 58 females (families 58.1–58.8), 6-line 89 females (families 89.1–89.6), and 10-line 89 males (families 89.7–89.16). The number of sons scored from each family varied from 4 to 50 with a median number of 16. A total of 39 R2 retrotransposition events giving rise to distinct 5' truncations were detected. Most of the new R2 insertions (36 of 39) were detected in individual sons. The 3 instances of a common insertion length in multiple sons all occurred through the female germ line with 2 insertions shared by 2 sons (families 58.6 and 89.6) and 1 insertion shared by 6 sons (family 58.3).
|
Among the 420 sons analyzed in these family studies, 19 contained deletions of from 2 to 12 R2 elements (table 1). If one assumes that all elements deleted from a chromosome occurred as a single event, then there were 2 such events. Five events were scored in single sons (families 58.2, 58.8, 89.2 and 2 in 89.6), 2 events were shared by 2 sons (families 89.4 and 89.6), and 1 deletion event was shared by 10 sons (family 58.6). All observed deletions occurred in the female germ line.
In these family studies, the mean number of new R2 element insertions per chromosome per generation can be calculated as the total number of new R2 copies present in the sons divided by the number of sons scored. The estimated rates of accumulation of new R2 elements in the female germ lines of 58 and 89 are, therefore, approximately 0.12 and 0.15 insertions/chromosome/generation, respectively. The rate of accumulation of R2 elements through the 89 male germ line was calculated as 0.05 elements/chromosome/generation. The lower rate of R2 retrotransposition through the male germ line compared with the female germ line appears significant (P < 0.05, chi-square test). The rates of R2 loss per chromosome per generation can be calculated as the number of parental R2 elements deleted in the sons divided by the number of sons scored. The rates of R2 loss in the female germ line of the 2 strains were estimated as 0.44 (line 58) and 0.22 (line 89) insertions/chromosome/generation. The 2-fold higher rate of deletion in line 58 was due to a single event shared by 10 sons (family 58.6).
Timing of R2 Activity in the Female Germ Line
The finding of similar length R2 insertions in more than one son suggested that some events occur during germ line stem cell proliferation. To more accurately estimate the timing of retrotransposition events, the 5'-truncation profiles of all 214 sons from 1-line 89 female were generated (family 89.17, bottom of table 1). The large number of insertion and deletion events scored in this family are summarized in figure 3. At the top of each panel in figure 3 is a diagram of the R2 elements of the mother with vertical lines indicating the end points of each 5'-truncated R2 element present on the X chromosomes (a and b) of the mother. The differences in the R2 5'-truncation profiles of the sons are shown on the horizontal lines below these R2 diagrams with vertical lines, indicating new insertions and ovals indicating the loss of R2 elements. One hundred sons inherited the a chromosome from the mother, and 114 sons inherited the b chromosome.
|
Thirty-two distinct new 5'-truncated R2 elements were detected in a total of 39 sons. As in the smaller family studies, most new insertions (27) were observed in only 1 son. Four of the insertions appeared to have shared by 2 sons and 1 insertion was shared by 13 sons (shared insertions are marked in fig. 3 with triangles). Three of the 4 insertions shared by 2 sons occurred in the same 2 sons (chromosomes marked at left with open squares), suggesting that they did not occur independently in each son. The 1 insertion shared by 13 sons also appears to have resulted from a common event (chromosomes marked with a filled circle). Thus, there is only 1 possible example (chromosomes marked with open circles) where independent events may have given rise to similar length 5' truncations.
It should be noted that the 2 examples of siblings with shared insertions had additional differences in their R2 5'-truncation profiles, indicating subsequent events. First, in the case of the 13 sons sharing the same R2 insertion on chromosome a (filled circles), 2 sons had additional distinct R2 insertions. These additional insertions presumably occurred in one or more cell divisions after the shared insertion. Second, the 2 sons with the same set of 3 new insertions (open squares) differed in the number of R2 elements deleted, suggesting that at least one deletion event occurred in a cell division after the common insertions. In total, there were 48 new R2 elements in the 214 sons for a mean insertion rate of 0.22 elements/chromosome/generation (table 1).
In the case of R2 deletions, 10 of the 214 sons contained chromosomes that had undergone deletion events involving from 1 to 15 elements (ovals on the vertical lines). Most of the deletions occurred on the b chromosome where 60% of the 5'-truncated copies originally present on this chromosome (23 of 38) were deleted in at least one son. There was a total loss of 55 R2 elements in all 214 sons for a mean deletion rate of 0.26 elements/chromosome/generation, similar to the rate seen for the smaller 89 families in table 1.
The large 89.7 family study revealed that although the timing of new insertions was straightforward, the timing and size of the deletion events could not always be determined. For example, although 3 sons (chromosome b, asterisks) had unique patterns of deleted R2 elements suggesting single-deletion events late in development of the germ cells, a total of 5 sons shared the same set of 4 deleted R2 elements (filled boxes). Two of these sons had the a chromosome of the mother, whereas 3 sons had the b chromosome, indicating that similar deletions had occurred on each chromosome. Because this area of the rDNA locus appears frequently deleted, it is possible that this deletion may have occurred independently multiple times on both the a and b chromosome. Finally, the 2 sons sharing the same set of 3 new R2 insertions (open squares) also had 3 and 7 R2 deletions, with 3 of the deletions shared. Thus, there could have been an initial deletion of 3 elements in the common lineage of these sons followed by a second deletion involving 4 elements later in the lineage of 1 son. Alternatively, separate but overlapping deletions involving 3 and 7 elements could have occurred after the common R2 insertions.
The Patterns of Insertions and Deletions
Figure 4 plots the number of new R2 insertions and deletions per chromosome for all 511 sons monitored through the female germ line (table 1). Of the 74 sons with new insertions, 60 sons had 1 insertion, 11 sons had 2 new insertions, and 2 sons had 3 new insertions. With a mean rate of 0.17 insertions/chromosome (88 new insertions/511 sons, table 1), these data suggest a model in which all chromosomes inherited through the female germ line had a similar probability of acquiring new insertions, rather than models in which subsets of germ cells are released from constraint and many new R2 insertions accumulate.
In contrast to the low numbers of new R2 insertions on any chromosome, chromosomes that had undergone the deletion of parental R2 elements lost from 1 to 15 5'-truncated elements with an average of 5.2 elements (fig. 4). Because only about one-third of the R2 elements in these lines have 5' truncations that can be scored by our PCR approach, the average number of R2 elements lost per deletion event is approximately 15. Thus, the loss of R2 elements from a chromosome involves the deletions of large numbers of rDNA units.
Accumulating Effects of R2 Activity over 30 Generations
Sublines of the 58 line were next established in an attempt to monitor the accumulating effect of high levels of R2 activity. Each subline was started with a single random pair and subsequently maintained by mass mating for 30 generations (sublines 58A–58H). At the 30th generation, 3 males from each subline were selected and their R2 5'-truncation profiles were compared with that of the original founding pair of the subline. The comparisons for each subline are shown in figure 5. In each panel, thin vertical lines on the 3 horizontal lines indicate R2 elements that were present on the chromosomes of the founding pair, thick vertical lines indicate new R2 insertions, and ovals indicate deletions of R2 elements originally present in the founding pair.
|
Seven of the 8 sublines showed evidence of retrotransposition events. To determine if the single subline without new insertions (subline 58H) had in fact lost R2 retrotransposition activity, the 5'-truncation profiles of 18 additional 30th-generation males from this subline were generated (data not shown). Four R2 insertions were detected in these additional males, suggesting that R2 activity was reduced but not eliminated in this subline. Seven of the 8 sublines also underwent deletions events. These deletions involved a large percentage of the original R2 elements present in line 58. Of the 30 5'-truncated elements that were at high frequency in line 58 during the start of the experiment (present in a progenitor of at least 6 of the 8 sublines), 19 elements or 63% had undergone a deletion. Only 6 of these elements underwent deletions in more than one subline, suggesting that the deletions were widely scattered over the rDNA locus.
The combined number of R2 changes on each chromosome (insertions and deletions) after 30 generations averaged 5.8 (range 1–16). Lines differed in whether they had undergone an increase or decrease in the number of R2 5'-truncated elements with sublines 58B, 58C, 58F, and 58G undergoing net increases of at least one R2 element/chromosome and sublines 58D and 58H undergoing net decreases of at least one R2 element/chromosome.
Across all 8 sublines, we scored a total of 46 distinct R2 insertions. Because the same insertions were found in multiple 30th-generation chromosomes, the total increase was 63 R2 elements for a rate of addition of 0.09 elements/chromosome/generation. This value is similar to the rate of 0.12 elements/chromosome/generation seen in the single-generation experiments (see table 1). There was a total loss of 71 R2 elements in the 8 sublines for a mean deletion rate of 0.10 elements/chromosome/generation. The lower deletion rate observed in these 30th-generation experiments compared with the single-generation experiment (table 1) suggests that those flies which undergo large deletions of their rDNA locus are frequently less fit and are eliminated from the population.
The large number of both insertions and deletions in many sublines suggested that the fraction of the rDNA units inserted with R2 in each subline would not have changed significantly from the original 58 line. Because changes in only about one-third of the R2 elements were scored by the PCR approach, we also monitored changes in the composition of the rDNA locus in each subline by genomic blotting (see Zhang and Eickbush 2005, for the approach used). The fraction of all rDNA units in the sublines that were inserted with R2 varied from 0.27 to 0.32 (data not shown), which is within the experimental error of our determinations of these fractions. Thus, no evidence was obtained to suggest that the sublines had undergone changes in fraction of their rDNA locus inserted with R2 elements. Indeed, without the ability to monitor specific R2-inserted units, the loci of line 58 would not appear to be particularly dynamic.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Developmental Pattern of R2 Retrotransposition
In this report, we have investigated the patterns of R2 insertions and deletions in lines of D. simulans with high levels of R2 retrotransposition activity. The rate of accumulation of R2 elements through the female germ lines was estimated to be 0.12 and 0.19 elements/chromosome/generation in lines 58 and 89, respectively (table 1, the estimate for line 89 included the data from panels B and D). Because we estimate that only one-third of the total R2 insertions were scored by our PCR approach (Zhang and Eickbush 2005
), the actual rates of R2 retrotransposition are likely to be 3-fold higher. Assuming that the male and female germ lines have R2 retrotransposition machinery with similar efficiencies in making full-length versus 5'-truncated copies, the accumulation of R2 elements through the male germ line of line 89 was estimated to be 3-fold lower than that through the female germ line.
The R2 insertion rates in the D. simulans lines are about 100 times higher than the rate detected in a long-term experiment involving R2 accumulation in Drosophila melanogaster (11 insertions involving 5'-truncated elements were detected in 19 sublines after 343 generations) (Perez-Gonzalez et al. 2003). Because in D. melanogaster rDNA units are present on both the X and Y chromosomes and equal numbers of R2 insertions were detected on the X and Y loci, R2 retrotranspositions in this species also appear to occur in both germ lines.
No consistent pattern of transposable element activity has been found in the germ lines of Drosophila. P elements and gypsy elements are active in the male and female germ lines (Ilyin et al. 1991; Zhang and Spradling 1993), copia elements are active only in males (Pasyukova et al. 1997), whereas I elements are active only in females (Bucheton 1990). Interestingly, R1 elements, which like R2 specifically insert in the 28S rRNA gene, appear to be active predominantly in the male germ line (Perez-Gonzalez et al. 2003).
The family studies in this report allowed an estimation of when R2 retrotranspositions occur during the development of the female germ line. Each ovary of a D. melanogaster female contains 16–20 ovarioles with each ovariole containing 2–3 germ line stem cells and each stem cell giving rise to 5–7 eggs (King 1970; Wieschaus and Szabad 1979). Thus, a healthy female can produce 400–700 eggs. Females from the active D. simulans stocks can also produce about this number of eggs; therefore, our assay of the 214 sons of female 89.17 (fig. 3) enabled us to sample from 30% to 50% of the total eggs produced by this female. Of the 32 different R2 retrotransposition events detected, 27 events gave rise to insertions in a single son and 4 to insertions shared by 2 sons. Therefore, most of the retrotransposition events in this female appeared to occur during the final divisions of the germ cells giving rise to the eggs. Similarly, in the case of 50 sons scored in families 89.5 and 89.6 (table 1), only 1 of the 15 retrotransposition events detected in these 2 families was shared by 2 sons, again consistent with the model that most retrotransposition events occur late in the development of germ cells. In total, only 2 retrotransposition events, the common R2 insertion in 13 sons in family 89.17 and the common insertion in 6 sons in family 58.3, appeared to have occurred earlier in development during a period when the germ cells were still proliferating.
Dynamics of R2 Insertion and Deletion from the rDNA Locus
The high levels of R2 activity in the D. simulans 58 and 89 lines were first observed in 2002 and have been frequently monitored since that time. Given this stable high level of activity, an unexpected property of these lines was that they did not have excessively high levels of R2 insertion (Zhang and Eickbush 2005). Indeed, the fractions of the rDNA units inserted with R2 in the various sublines studied in this report (0.27–0.32) are somewhat lower than the 0.34 value reported several years ago (Zhang and Eickbush 2005). One model to explain why R2 insertions do not accumulate to higher levels in these lines is that these lines are already at their maximum allowable fitness limit and that chromosomes with even higher levels of insertion are selected against. Although such selective pressure no doubt contributes to the long-term level of R2 elements in any population, the findings in this report suggest another important property of R2 retrotransposition that contributes dramatically to the total level of R2 elements in a population.
Our studies of the active D. simulans lines revealed that R2 elements were being rapidly deleted from the rDNA locus. Indeed, the rates of R2 deletion through the female germ line as monitored in single generations were higher than the rates of R2 insertion (table 1). Because non-LTR retrotransposons are not known to be able to excise (reviewed in Ostertag and Kazazian 2001) and the deletion "events" affected multiple R2 copies, the R2 eliminations appear to result from the loss of large blocks of R2-inserted rDNA units rather than the excision of elements from these units.
Because the D. simulans lines without R2 activity have R2 5'-truncation profiles that are stable for many generations (Zhang and Eickbush 2005; Zhang and Eickbush, unpublished data), we suggest that R2 retrotransposition activity itself is responsible for the rapid loss of R2 from the rDNA. The timing of the R2 deletions was similar to that of the retrotransposition. Most deletions were found in individual sons, a few deletions were shared by 2 sons, and only 1 event was shared by enough sons to suggest its occurrence during germ line proliferation (table 1, family 58.6).
What recombinational mechanism induced by R2 activity could be responsible for the observed R2 deletions? Gene conversions are unlikely because they involve only localized regions of a chromosome and thus are more likely to eliminate single insertions rather than the multiple R2 copies we observed. Unequal crossovers could result in the elimination of many R2 elements in a single event. However, unequal crossovers while deleting rDNA units with R2 elements from 1 recombinant chromosome duplicate those units on the reciprocal chromosome. Based on the uniform intensity of the PCR products derived from the many 5'-truncated R2 elements in the active lines, most new and old elements are single copy, suggesting that they infrequently undergo duplications by unequal crossover. Thus, unequal crossovers seem unlikely to explain the observed deletions. Only loop deletion, in which recombination between 2 rDNA units on the same DNA strand results in an extrachromosomal circle that is lost from the cell, can explain the elimination of multiple R2 elements in a single event without duplicating those elements on another chromosome.
We suggest that R2-retrotransposition activity itself is creating large deletions in the rDNA locus. There are precedents for non-LTR retrotransposition events involving large deletions. For example, insertion of the L1 and Alu elements in mammals gives rise to large deletions of DNA (tens of kilobase) upstream of the insertion site, presumably as the result of the cell machinery attempting to repair a chromosomal break (Gilbert et al. 2002, 2005; Callinan et al. 2005). The rDNA locus may be particularly prone to such deletion events when 2 retrotransposition events are simultaneously attempted in 1 locus. The deletions caused directly or indirectly by R2 retrotransposition appear to be distributed throughout the locus as 60% of the 5'-truncated elements in the 89.17 family (fig. 3) and 63% of the 5' truncation in line 58 (fig. 5) were involved in deletion events. The deletions are large with a mean of five 5'-truncated elements (or 15 total R2 elements) deleted per event (fig. 4). The largest deletions eliminated 40% of the 5'-truncated elements from the rDNA locus.
Monitoring the dynamics of R2 insertion and deletion in line 58 after 30 generations revealed an insertion rate (0.09 elements/chromosome/generation) similar to the single-generation estimates for the combined males and females germ line rates. Thus, the chromosomes in these populations that have undergone retrotransposition events are not strongly selected against. On the other hand, the 30th-generation experiments revealed lower rates of R2 deletions (0.07 elements/chromosome/generation) than that observed in the single-generation experiments (0.22–0.44 elements/chromosome/generation). This lower deletion rate suggests that individuals with the largest deletions are frequently lost from the population. Assuming that R2-inserted rDNA units are interspersed with uninserted rDNA units, the chromosomes with the largest deletions are presumably selected against due to shortages of functional rDNA units. Thus, only chromosomes with smaller deletions or deletions that involve region of the rDNA locus enriched for R2 insertions will survive.
The deletion of units from the rDNA locus by R2 retrotransposition will cause a net loss of rDNA units from the population and thus in the long term may be compensated by recombination (e.g., unequal crossovers) and the selection of recombinant chromosomes with larger numbers of rDNA units. Alternatively, there is the possibility that in active lines the cell has evolved the means to specifically amplify uninserted rDNA units. This latter suggestion is intriguing and brings to mind previous studies in Drosophila of DNA magnification and rolling circle models of replication (reviewed in Hawley and Marcus 1989).
What causes the high rates of R2 retrotransposition in the active D. simulans lines remains unclear. Genetic studies of these stocks suggest that much of the control over R2 activity resides within the rDNA locus itself (Eickbush D, Ye J, Burke W, Zhang X, Eickbush T, unpublished data). Given the rapid changes occurring in the rDNA loci of these lines, yet their ability to retain active R2 elements suggests that many of the 30–60 full-length R2 elements present in these lines are "active" and that these active elements are distributed throughout the locus.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Chris Skeehan for generating the analysis of some of the single-family data. We thank Danna Eickbush and Deborah Stage for comments on the manuscript. The work was supported by National Institutes of Health grant GM42790.
|
Footnotes |
|---|
Norihiro Okada, Associate Editor
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Aminetzach YT, Macpherson JM, Petrov DA. Pesticide resistance via transposition-mediated adaptive gene truncation in Drosophila. Science (2005) 309:764–767.
Averbeck KT, Eickbush TH. Monitoring the mode and tempo of concerted evolution in the Drosophila melanogaster rDNA locus. Genetics (2005) 171:1837–1846.
Biemont C, Lemeunier F, Garcia MP, Brookfield JF, Gautier C, Aulard S, Pasyukova EG. Population dynamics of the copia, mdg1, mdg3, gypsy, and P transposable elements in a natural population of Drosophila melanogaster. Genet Res (1994) 63:197–212.[Web of Science][Medline]
Bucheton A. I transposable elements and I-R hybrid dysgenesis in Drosophila. Trends Genet (1990) 6:16–21.[CrossRef][Web of Science][Medline]
Burke WD, Malik HS, Lathe WC III, Eickbush TH. Are retrotransposons long term hitchhikers? Nature (1998) 392:141–142.[CrossRef][Medline]
Callinan PA, Wang J, Herke SW, Garber RK, Liang P, Batzer MA. Alu retrotransposition-mediated deletion. J Mol Biol (2005) 348:791–800.[CrossRef][Web of Science][Medline]
Charlesworth B, Sniegowski P, Stephan W. The evolutionary dynamics of repetitive DNA in eukaryotes. Nature (1994) 371:215–220.[CrossRef][Medline]
Desset S, Conte C, Dimitri P, Calco V, Dastugue B, Vaury C. Mobilization of retroelements, ZAM and Idefix, in a novel unstable line of Drosophila melanogaster. Mol Biol Evol (1999) 16:54–66.[Abstract]
Eickbush DG, Eickbush TH. Vertical transmission of retrotransposable elements R1 and R2 during the evolution of Drosophila melanogaster. Genetics (1995) 139:671–684.[Abstract]
Eickbush DG, Eickbush TH. Transcription of endogenous and exogenous R2 elements in the rRNA gene locus of Drosophila melanogaster. Mol Cell Biol (2003) 23:3825–3836.
Eickbush TH. R2 and related site-specific non-LTR retrotransposons. In: Mobile DNA II—Craig NL, Craigie R, Gellert M, Lambowitz AM, eds. (2002) Washington (DC): American Society of Microbiology Press. 813–835.
Gentile KL, Burke WD, Eickbush TH. Multiple lineages of R1 retrotransposable elements can coexist in the rDNA loci of Drosophila. Mol Biol Evol (2001) 18:235–245.
Gilbert N, Lutz S, Morrish TA, Moran JV. Multiple fates of L1 retrotransposition intermediates in cultured human cells. Mol Cell Biol (2005) 25:7780–7795.
Gilbert N, Lutz-Prigge S, Moran J. Genomic deletion created upon LINE-1 retrotransposition. Cell (2002) 110:315–325.[CrossRef][Web of Science][Medline]
Hawley RS, Marcus CH. Recombinational controls of rDNA redundancy in Drosophila. Annu Rev Genet (1989) 23:87–120.[CrossRef][Web of Science][Medline]
Ilyin YV, Lyubomirskaya NV, Kim AI. Retrotransposon Gypsy and genetic instability in Drosophila. Genetica (1991) 85:13–22.[CrossRef][Web of Science][Medline]
Jakubczak JL, Burke WD, Eickbush TH. Retrotransposable elements R1 and R2 interrupt the rRNA genes of most insects. Proc Natl Acad Sci USA (1991) 88:3295–3299.
Jakubczak JL, Zenni MK, Woodruff RC, Eickbush TH. Turnover of R1 (type I) and R2 (type II) retrotransposable elements in the ribosomal DNA of Drosophila melanogaster. Genetics (1992) 131:129–142.[Abstract]
Kim AI, Belyaeva ES, Aslanian MM. Autonomous transposition of gypsy mobile elements and genetic instability in Drosophila melanogaster. Mol Gen Genet (1990) 224:303–308.[CrossRef][Web of Science][Medline]
King RC. Ovarian development in Drosophila melanogaster. (1970) New York: Academic Press.
Kojima KK, Fujiwara H. Long-term inheritance of the 28S rDNA-specific retrotransposon R2. Mol Biol Evol (2005) 22:2157–2165.
Kojima KK, Kuma K, Toh H, Fujiwara H. Identification of rDNA-specific non-LTR retrotransposons in Cnidaria. Mol Biol Evol (2006) 23:1984–1993.
Lin YJ, Seroude L, Benzer S. Extended life-span and stress resistance in the Drosophila mutant methuselah. Science (1998) 282:943–946.
Lohe AR, Roberts PA. An unusual Y chromosome of Drosophila simulans carrying amplified rDNA spacer without rRNA genes. Genetics (1990) 125:399–406.[Abstract]
Lyckegaard EM, Clark AG. Evolution of ribosomal RNA gene copy number on the sex chromosomes of Drosophila melanogaster. Mol Biol Evol (1991) 8:458–474.[Abstract]
Malik HS, Burke WD, Eickbush TH. The age and evolution of non-LTR retrotransposons. Mol Biol Evol (1999) 16:793–805.[Abstract]
Maside X, Bartolome C, Assimacopoulos S, Charlesworth B. Rates of movement and distribution of transposable elements in Drosophila melanogaster: in situ hybridization vs. Southern blotting data. Genet Res (2001) 78:121–136.[Web of Science][Medline]
Nuzhdin SV, Mackay TF. The genomic rate of transposable element movement Drosophila melanogaster. Mol Biol Evol (1995) 12:180–181.[Web of Science][Medline]
Ostertag EM, Kazazian HH. Biology of mammalian L1 retrotransposons. Annu Rev Genet (2001) 35:501–538.[CrossRef][Web of Science][Medline]
Pasyukova EG, Nuzhdin SV. Doc and copia instability in an isogenic Drosophila melanogaster stock. Mol Gen Genet (1993) 240:302–306.[CrossRef][Web of Science][Medline]
Pasyukova EG, Nuzhdin SV, Li W, Flavell AJ. Germline transposition of the copia retrotransposon in Drosophila melanogaster is restricted to males by tissue-specific control of copia RNA levels. Mol Gen Genet (1997) 255:115–124.[CrossRef][Web of Science][Medline]
Perez-Gonzalez CE, Burke WD, Eickbush TH. R1 and R2 retrotransposition and deletion in the rDNA loci on the X and Y chromosomes of Drosophila melanogaster. Genetics (2003) 165:675–685.
Perez-Gonzalez CE, Eickbush TH. Dynamics of R1 and R2 elements in the rDNA locus of Drosophila simulans. Genetics (2001) 158:1557–1567.
Perez-Gonzalez CE, Eickbush TH. Rates of R1 and R2 retrotransposition and elimination from the rDNA locus of Drosophila melanogaster. Genetics (2002) 162:799–811.
Weischaus E, Szabad J. The development and function of the female germ line in Drosophila melanogaster: a cell lineage study. Dev Biol (1979) 68:29–46.[CrossRef][Web of Science][Medline]
Zhang P, Spradling AC. Efficient and dispersed local P element transposition from Drosophila females. Genetics (1993) 133:361–373.[Abstract]
Zhang X, Eickbush TH. Characterization of active R2 retrotransposition in the rDNA locus of Drosophila simulans. Genetics (2005) 170:195–205.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  X. Zhang, M. T. Eickbush, and T. H. Eickbush Role of Recombination in the Long-Term Retention of Transposable Elements in rRNA Gene Loci Genetics, November 1, 2008; 180(3): 1617 - 1626. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. G. Eickbush, J. Ye, X. Zhang, W. D. Burke, and T. H. Eickbush Epigenetic Regulation of Retrotransposons within the Nucleolus of Drosophila Mol. Cell. Biol., October 15, 2008; 28(20): 6452 - 6461. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||