pmc logo imageJournal ListSearchpmc logo image
Logo of nihpaNIHPA bannerabout author manuscriptssubmit a manuscript
Journal List > NIHPA Author Manuscripts
Abstract
>Full Text
PDF (1.1M)

PubMed articles by:
Gladyshev, E.
Meselson, M.
Arkhipova, I.
Gene.Author manuscript; available in PMC 2007 April 1.
Published in final edited form as:
Gene. 2007 April 1; 390(1-2): 136–145.
Published online 2006 October 5. doi: 10.1016/j.gene.2006.09.025.
PMCID: PMC1839950
NIHMSID: NIHMS19032
Copyright notice and Disclaimer
A deep-branching clade of retrovirus-like retrotransposons in bdelloid rotifers
Eugene A. Gladyshev,1 Matthew Meselson,1,2 and Irina R. Arkhipova1,2*
1 Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138, USA
2 Josephine Bay Paul Center for Comparative Molecular Biology and Evolution, Marine Biological Laboratory, Woods Hole, MA 02543, USA
Address for correspondence: *Dr. Irina Arkhipova, Department of Molecular and Cellular Biology, Harvard University, Cambridge, MA 02138, USA., Tel. (617) 495-7899, Fax: (617) 496-2444, E-mail: arkhipov/at/fas.harvard.edu
Small right arrow pointing to: The publisher's final edited version of this article is available at Gene.
Small right arrow pointing to: See other articles in PMC that cite the published article.
Publisher's Disclaimer
>Abstract
Abstract
Rotifers of class Bdelloidea, a group of aquatic invertebrates in which males and meiosis have never been documented, are also unusual in their lack of multicopy LINE-like and gypsy-like retrotransposons, groups inhabiting the genomes of nearly all other metazoans. Bdelloids do contain numerous DNA transposons, both intact and decayed, and domesticated Penelope-like retroelements Athena, concentrated at telomeric regions. Here we describe two LTR retrotransposons, each found at low copy number in a different bdelloid species, which define a clade different from previously known clades of LTR retrotransposons. Like bdelloid DNA transposons and Athena, these elements have been found preferentially in telomeric regions. Unlike bdelloid DNA transposons, many of which are decayed, the newly described elements, named Vesta and Juno, inhabiting the genomes of Philodina roseola and Adineta vaga, respectively, appear to be intact and to represent recent insertions, possibly from an exogenous source. We describe the retrovirus-like structure of the new elements, containing gag, pol, and env-like open reading frames, and discuss their possible origins, transmission, and behavior in bdelloid genomes.
Keywords: LTR retrotransposons, asexual reproduction, reverse transcriptase, integrase, gag, pol, env
>1. Introduction
1. Introduction
Rotifers of the class Bdelloidea, comprising some 370 described species, are microscopic freshwater invertebrates of worldwide distribution. The class is of particular interest because its apparent asexuality challenges the widely held view that sexual reproduction is essential for long-term evolutionary success and that its abandonment leads to early extinction (Bell, 1982; Mark Welch and Meselson, 2000; Normark et al., 2003). Although there are many hypotheses, there is no general agreement as to what accounts for the nearly universal occurrence of sexual reproduction in animals and plants and the relatively early extinction of asexual lineages (Kondrashov, 1993; Barton and Charlesworth, 1998; Otto and Lenormand, 2002). According to one of the hypotheses, a central factor in the maintenance of sexual reproduction is its role in limiting the increase of parasitic mobile genetic elements, which, if unchecked, would drive their hosts to extinction (M.M., in Arkhipova et al., 1995; Arkhipova and Meselson, 2000, 2005a).
In order to investigate this hypothesis and the possible connections between long-term asexuality and transposable elements (TE), we initiated studies of TE content in several bdelloid species. We first performed a PCR screen with highly degenerate primers for the most widespread TE superfamilies, namely LINE-like and gypsy-like retrotransposons and mariner/Tc-like DNA transposons (Arkhipova and Meselson, 2000). No LINE-like or gypsy-like reverse transcriptases were detected in any of the 5 bdelloid species tested, although LINEs were easily detected in 39 and gypsies in 35 other species representing 23 animal phyla, while mariner/Tc-like transposases were found in bdelloids and exhibited their characteristically patchy distribution among other species. In a subsequent survey of DNA transposons, the pattern of mariner evolution was found to be consistent with repeated horizontal transfers from an exogenous source, followed by limited intragenomic proliferation and eventual decay (Arkhipova and Meselson, 2005b). These and other types of DNA transposons (hAT, piggyBac, helitron, foldback), often inserted into each other, were found to be largely concentrated near chromosome ends, associated with Athena retroelements, but notably lacking in the more proximal, gene-rich regions thus far sequenced. Athena retroelements cannot be assigned to either LTR or non-LTR retrotransposons, but instead belong to Penelope-like elements, a recently described class of eukaryotic retroelements forming a sister clade to telomerase reverse transcriptases (Arkhipova et al., 2003). The movement of these intron-containing retroelements is apparently confined to chromosome ends, where they may act in a backup pathway for telomere elongation (E.G. and I.A., submitted).
During the sequencing of a telomere-associated Athena-containing cosmid from the bdelloid Philodina roseola (Arkhipova and Meselson, 2005b), we encountered an LTR retroelement that appeared to be more closely related to chromoviruses (Kordis, 2005) than to any other known group of LTR retrotransposons. We now find that, together with a related but distinct element from Adineta vaga, a bdelloid species from a family that separated from the Philodinidae tens of millions of years ago, these new elements, designated Vesta and Juno, respectively, define a deep-branching clade of LTR retrotransposons, appearing as a sister clade to chromoviruses. Within each species, the copies of these elements are only slightly divergent and several copies have been localized to regions near telomeres. Present at very low copy number and apparently intact, Vesta and Juno may have entered the respective bdelloid genomes only recently, their horizontal entry being facilitated by a putative env region from an unknown source.
>2. Materials and Methods
2. Materials and Methods
2.1. DNA manipulations
The genomic cosmid library of P. roseola embryo DNA (ca. 5x coverage) prepared by partial Sau3AI digestion (Mark Welch et al., 2004) was screened with a 32P-labeled Vesta probe (Fig. 1a) and a PCR-generated Athena probe (Arkhipova et al., 2003). The A. vaga genomic fosmid library (ca. 6x coverage) prepared from randomly sheared embryo DNA (J. Hur, Ph.D. dissertation, Harvard University) was screened with a 32P-labeled (TGTGGG)4 probe and a Juno probe (Fig. 1b). Library screening and isolation of cosmid and fosmid clones were performed according to instructions from manufacturers of library production kits (Stratagene and Epicentre Technologies). Hybridizing cosmids and fosmids were analyzed by restriction enzyme digestion and by direct sequencing with standard T3, T7 and pCC1Fos primers and with primers used to generate the probes. Fosmid/cosmid inserts selected for complete sequencing were sheared by sonication, subcloned into pBluescript II SK-(Stratagene), and sequenced with T3 and T7 primers using BigDye Terminator v3.1 (Applied Biosystems) and ABI 3730 XL capillary sequencers at the high-throughput W.M. Keck Ecological and Evolutionary Genetics Facility at the MBL. Four additional Juno copies were sequenced by primer walking with a set of 12 custom oligonucleotides.
Figure 1Figure 1
Structural organization of (A) Vesta and (B) Juno retrotransposons. Shaded triangles, LTRs; open boxes, ORFs. Functional domains within each ORF are discussed in the text. The predicted introns connect two putative exons in ORF3, although may not necessarily (more ...)
2.2. Bioinformatics
Sequences were assembled with Phred/Phrap/Consed (www.phrap.org), and aligned with the AlignX program, based on the ClustalW algorithm, from the VectorNTI suite 7 (InforMax). Alignments are available from the corresponding author upon request. Ks and Ka/Ks ratios were calculated by the DIVERGE program of the GCG Wisconsin package (Accelrys), which uses the method of Pamilo and Bianchi (1993). Phylogenetic analysis was performed with MEGA 3.1 (Kumar et al., 2004) by neighbor-joining (pairwise deletion; Poisson correction model; uniform among-site rate variation), and with MrBayes 3.1.1 (Ronquist and Huelsenbeck, 2003) using the mixed amino acid substitution model, invariable plus gamma among-site rate variation, 106 generations, each 100th tree sampled, and the first 200 trees discarded as burn-in. Sequences obtained in this study were deposited in GenBank under accession numbers DQ985390-DQ985395.
>3. Results
3. Results
3.1. Structural organization
The two LTR retrotransposons characterized in this study share a highly similar overall organization, with three open reading frames (ORFs) in the same order in both elements (Fig. 1). The P. roseola element, Vesta, was found on an Athena-containing telomeric cosmid, while the A. vaga element, Juno, was initially found on a fosmid selected from a genomic library by hybridization with A. vaga telomeric repeats. Additional Juno copies were isolated by probing the A. vaga genomic library with an internal fragment (Fig. 1). The cloned copy of Vesta is 7981 bp in length, while Juno is slightly longer (consensus 8136 bp), mainly because of a longer ORF3. Both elements generate a 5-bp non-specific target-site duplication (TSD) upon insertion. A comparative description of the functionally significant regions of the two elements is given below.
3.1.1. LTRs and noncoding regions
The LTRs of Vesta and Juno are very compact, 343 and 283 bp in length, respectively (Fig. 1; Table 1). Both begin with the canonical TG dinucleotide and have short terminal inverted repeats at their boundaries, TGTTA...TAACA in Vesta and TgTTA...TAAtA in Juno. While such short LTRs are not likely to accommodate enhancer sequences, the elements of a basal eukaryotic promoter can easily be discerned: transcription probably starts within a canonical initiator (Inr) element, TCAGT (Smale and Baltimore, 1989; Arkhipova, 1995), with Juno also containing a TATATA motif 25–30 bp upstream of Inr. Both LTRs also carry a polyadenylation signal, AATAAA, as required for production of a putative transcript with a short terminal redundancy (R) and formation of the characteristic U3-R-U5 structure of retrotransposon LTRs (Arkhipova et al., 1986). The TGG of the putative primer-binding site, presumed to be complementary to the 3' CCA of an unidentified host tRNA, is separated from the left LTR by a dinucleotide (as in retroviruses) in Juno, and by 3 bp in Vesta (Table 1). Although polypurine tracts (PPT) adjacent to the right LTRs are not very pronounced (Table 1), functional studies of Ty1 PPT in yeast demonstrate that a PPT with 6 purines out of 9 nucleotides is sufficient for plus-strand priming (Heyman et al., 1995). The 5' and 3' UTRs are also very compact (Tables 1, 2). The compact nature and low sequence complexity of the LTR and UTR regions indicates that Vesta and Juno likely possess a basal promoter but have not acquired any extended regulatory regions from the host, as often happens during host-TE co-evolution (e.g. Wilson et al., 1998; Spana et al., 1988). Finally, the non-coding region between Juno ORF2 and ORF3 exhibits hyper-variability in comparison with the rest of the element (Table 2).
Table 1Table 1
Characteristics of LTRs and adjacent regions in Vesta and Juno.
Table 2Table 2
Comparison of the Vesta and Juno coding regions.
3.1.2. Coding regions
3.1.2.1. Gag
In most retrovirus-like elements ORF1, corresponding to the gag region, codes for proteins that bind to viral template RNA to form the nucleoprotein core particle (Coffin et al., 1997). Both Vesta and Juno ORF1s contain regions homologous to the major capsid (CA) (Retrotrans_gag, pfam03732) and nucleocapsid (NC) proteins (zf-CCHC, pfam00098) (Fig. 2). The Vesta NC protein contains one Zn knuckle motif with the characteristic retrovirus-like spacing CX2CX4HX4C, expected to play a role in nucleic acid binding, and the Juno NC protein contains two such motifs (Fig. 2, top). The gag region of retroviruses/retrotransposons is normally connected to the pol region via a translational frameshift (Coffin et al., 1997). Both Juno and Vesta exhibit this characteristic –1 frameshift, and apparently use the same mechanism to regulate pol expression.
Figure 2Figure 2
Multiple sequence alignment of Vesta and Juno with known representatives of Metaviridae and Retroviridae. Shown are the most conserved regions in the gag-like ORF and in the PR, RT, RH, and IN domains of pol. Highly conserved residues are denoted by asterisks. (more ...)
3.1.2.2. Pol
ORF2, the pol region, includes domains coding for protease, reverse transcriptase, RNase H, and integrase (PR-RT-RH-IN), in that order, as is characteristic of the Ty3/gypsy-like LTR retrotransposons, the Metaviridae (Boeke et al., 2004). A typical retroviral aspartyl protease domain (RVP, pfam00077), found in both elements, has the potential for processing of a gag-pol polyprotein. The reverse transcriptase (RT) domain (RVT, pfam00078), although bearing sufficient similarity to other Metaviridae RTs, differs from them in one of its most conserved motifs, RT4(B): in the nearly universal PFG (or PYG), the aromatic residue in the middle is replaced in Vesta by a methionine (which is found in this position only in Penelope-like elements and hepadnaviruses), and in Juno by a glutamine (typical of retroviruses). The deviant sequences of the most highly conserved RT motifs apparently precluded earlier detection of these retrotransposons in our nested PCR-based assays. The RNase H domain, which is needed for removal of the template RNA within the cytoplasmic RNP particle and for generation of the plus-strand primer, is present in all LTR retrotransposons, and its presence in Vesta and Juno indicates, together with the presence of the gag domain, that these elements undergo reverse transcription in a cytoplasmic RNP particle. Finally, the IN domain (RVE, pfam00665) displays all the characteristics of a retroviral/retrotransposon integrase, including a HH-CC Zn finger and the DDE catalytic domain. The C-terminal GPF/Y motif (Malik and Eickbush, 1999) is clearly evident in Juno, but exhibits signs of degeneration in the cloned copy of Vesta (Fig. 2, bottom). This domain is an optional component of the integrase enzyme and is not present in early-branching LTR retrotransposons (Ty1/copia, BEL, DIRS), but was acquired by retroviruses and late-branching Metaviridae, with subsequent loss in many retroviruses and in the gypsy superfamily, also known as insect errantiviruses (Boeke et al., 1999).
3.1.2.3. Env
One of the most interesting questions posed by the three-ORF structure of Vesta and Juno is whether they represent invertebrate retroviruses, i.e. whether their ORF3s correspond to the retroviral env region and confer the capacity for entering the cell in a manner resembling retroviral infection. The env region of retroviruses normally codes for two proteins, a surface protein (SU) binding a host receptor and a transmembrane protein (TM), which are cleaved from the precursor by a host endopeptidase (Coffin et al., 1997). Six lineages of LTR retrotransposons are known to have independently acquired such an ORF3, and in three cases a viral source could be identified (Malik et al., 2000). Although we could not identify a source for the env regions of Vesta or Juno using the same methods, this is not surprising, since sequence information regarding viruses of invertebrates from aquatic habitats is extremely limited (see below).
Of the expected amino acid sequence motifs characteristic of retroviral env proteins (Coffin et al., 1997; Misseri et al. 2003), we could identify transmembrane (TM) regions (three in Vesta, one or two in Juno), several N-linked glycosylation sites, a few cysteines that could participate in formation of disulfide bridges, and a putative furin-like protease cleavage site (Fig. 3). Interestingly, TM regions are detected only if ORF3 is extended beyond the stop codon/frameshift, either via splicing of a predicted intron, or via bypassing a frameshift occurring in a T-rich region, e.g. by ribosomal frameshifting. Similar TM-containing ORF3 extensions in a different reading frame were observed in several endogenous plant retroviruses (Wright and Voytas, 2002). It is conceivable that expression of the functional env protein was taking place in the pre-bdelloid hosts, but ceased after entering bdelloid genomes.
Figure 3Figure 3
Characterization of the putative env-like proteins encoded by (A) Vesta and (B) Juno. The putative N-linked glycosylation sites (NxS/T) are shaded; the potential furin-like protease cleavage sites (RxxR) and the canonical fusion tripeptide (FxG) motifs (more ...)
3.2. Copy numbers
Several approaches were used to estimate copy numbers. For Vesta, we performed Southern blot hybridization using a probe spanning the gag-pol junction (Fig. 1). Up to four hybridizing bands were seen in restriction digests of P. roseola genomic DNA (Fig. 4A), while hybridization at low stringency to genomic DNA from A. vaga and from another bdelloid species, Habrotrocha constricta, gave no signal (not shown). An independent estimate of 4–5 copies was obtained by probing the P. roseola genomic cosmid library with the same Vesta fragment and comparison of the number of hybridizing cosmids with the number of cosmids hybridizing under the same conditions to the hsp82 gene, known to be present in P. roseola in four copies, with pairwise divergence between two members of a lineage 4–5% and between members of different lineages up to 47% (Mark Welch et al., 2004). Finally, fluorescent in situ hybridization with the Vesta-containing cosmid (which also contains Athena elements, mariner/Tc-like DNA transposons, and a helitron, but no recognizable structural genes in the 40 kb insert) to P. roseola embryo nuclei gave 4–6 hybridization signals per nucleus, localized near telomeres (E.G. and I.A., submitted).
Figure 4Figure 4
Southern blot hybridization estimates of copy numbers of Vesta (A) and Juno (B) in genomic DNA. Hybridization probes are indicated in Fig. 1. (A) P. roseola genomic DNA digested with (1) EcoR1, (2) HindIII and (3) PvuII. EcoR1 has a single recognition (more ...)
For Juno, copy number was estimated by probing the A. vaga genomic fosmid library with a Juno probe spanning the highly conserved RT motifs 4 through 7 of the pol region (Fig. 1). All hybridizing fosmids were isolated, and the sequences of Juno copies and 0.6–0.7 kb of adjacent flanking regions were determined by primer walking. The 26 hybridizing fosmids obtained from this exhaustive screen were found to correspond to only 5 distinct copies, each with different flanking sequences. The screen was done under conditions in which approximately the same number of fosmids hybridize to a probe for the hsp82 gene, of which, as in P. roseola, there are four copies (J. Hur, personal communication). Copy number estimates from Southern blotting experiments with the same Juno probe agree with the estimates obtained from library screening, yielding four strongly hybridizing and one weakly hybridizing band (Fig. 4B).
3.3. Divergence
Each of the three ORFs in each of the five Juno copies appears intact, even in the most divergent copies (Juno1 and Juno4). Total nucleotide divergence between Juno copies in all comparisons is less than 2%, if the hypervariable region between ORF2 and ORF3 is excluded. For each ORF, values of synonymous divergence (Ks) and the ratio of non-synonymous to synonymous divergence (Ka/Ks) are given in Table 3. The excess of synonymous divergence for ORF1 and ORF2 is evidence of purifying selection, while for ORF3 Ka/Ks is close to unity. Comparison between LTRs of different Juno copies reveals the identity of LTRs in Juno2 and Juno3, which differ from the LTR of Juno5 by one substitution and two 1-bp slippages in oligo(T) stretches; Juno4 LTR differs from them by two more T slippages, and the most divergent LTRs, those of Juno1, differ from LTRs of Juno2-5 by a substitution, 3 slippages involving 2, 4, and 7 bp, and a 7-bp insertion. Differences apparently caused by replication slippage also prevail in the hypervariable region between ORF2 and ORF3, and within the C-terminal part of ORF3, including the presumptive intron. Five Vesta-containing cosmids selected from the P. roseola genomic library, when sequenced with several internal primers, did not reveal any differences from the fully sequenced copy, although we do not know how many different copies these cosmids represent.
Table 3Table 3
Divergence between five genomic Juno copies (Juno1-5) in the coding regions.
Divergence between LTRs of a given element has been taken as a measure of the time since the element last underwent transposition, since both LTRs are copied from the same regions of the RNA template during this process (R and U5 regions of each LTR are copied from the 5' RNA end, and U3 – from the 3' end). On this view, the identity of left and right LTRs (except for Juno4) implies that Vesta and Juno transposed recently. Taking the Drosophila synonymous substitution rate of 0.016 substitutions per site per Myr (Li, 1997), such identity between LTRs of ca 300 bp suggests that the time since the most recent transposition of any element may be no more than a few hundred thousand years. In contrast, if it is assumed that all five copies of Juno derive from a single invading copy, its arrival would have occurred several million years ago, as estimated from Ks values between the most divergent copies (Juno1,4) and the three more similar copies. Regardless of the particular value taken for the rate of synonymous substitution, the disparity between the lack of LTR-LTR divergence in individual Junos and the substantial values of Ks between copies requires explanation. The simplest possibility is that the most divergent and the least divergent Juno copies underwent at least two independent entries from an unidentified donor or donors. Less plausibly, the LTRs of individual elements may have been homogenized by conversion, observed in LTR retrotransposons only rarely (e.g. Johnson and Coffin, 1999). If so, such conversion would have had to be confined to LTRs belonging to the same element and not extending into adjacent regions on either side.
3.4. Phylogenetic placement
Of special interest is the place occupied by Vesta and Juno in the phylogeny of LTR retrotransposons. Initial phylogenetic analysis grouped the Vesta RT-RH and IN domains with chromoviruses (not shown), although examination of the IN region showed the presence of the GPF/Y domain and the lack of a chromodomain. We then extended the alignment to include the entire pol region (PR-RT-RH-IN). The tree was rooted with vertebrate retroviruses, which are a sister lineage to Metaviridae (Malik and Eickbush, 1999, 2001). The retrotransposon RNase H domain was aligned with the tether (connection) domain of retroviruses, previously shown to represent a remnant of an ancestral retrotransposon-like RNase H domain (Malik and Eickbush, 2001). The retroviral sequences included in the alignment also contain the GPY/F domain, which was subsequently lost in most retroviruses. Thus, the sequences employed in the alignment include all of the functional domains present in the pol region.
Although the two bdelloid LTR retrotransposons identified in this study are quite different from each other (25% and 50% aa identity of the gag and pol regions, respectively), they were found to form a distinct clade, with a bootstrap support value of 95% in neighbor-joining analysis, and 100% clade credibility value in Bayesian analysis (Fig. 5). The DGLT-A LTR retrotransposon from the social amoeba Dictyostelium discoideum (Glockner et al., 2001) appears to be most closely related to Vesta and Juno, although it cannot be placed in the same clade with comparable confidence (the bootstrap support in a neighbor-joining phylogeny is 46%). DGLT-A contains a single ORF, and none of the neighboring clades carry ORF3, suggesting its independent acquisition by Vesta and Juno. The bdelloid elements appear to form a sister clade to chromoviruses (Marin and Llorens, 2000; Kordis, 2005), a large and diverse clade of LTR retrotransposons with wide phylogenetic distribution, for which the major synapomorphy is the acquisition of the chromodomain C-terminally to the GPF/Y domain of the integrase.
Figure 5Figure 5
Phylogenetic placement of Vesta and Juno. The phylogram shows the results of the neighbor-joining analysis, with the clade support values from 1000 bootstrap replications indicated below the branches, and with the clade credibility values obtained from (more ...)
3.5. Genomic environment
The sequenced Vesta-containing cosmid clearly derives from a subtelomeric location, as seen in its localization by FISH and also in the presence of interspersed telomeric repeats and Athena retroelements, as well as various DNA transposons (Arkhipova et al., 2003; E.G. and I.A., submitted). One of the Juno copies (Juno4) is also located on a sequenced telomeric fosmid, as judged by the presence of a long stretch of A. vaga telomeric repeats at one end of the fosmid, followed by two Athena retroelements. This copy, the only one to carry a 1-bp difference between left and right LTRs, is slightly 3' truncated (at nt 167 of the 3' LTR, just after the TATATA motif) and joined to a 3'-truncated Athena, indicating either a deletion of the intervening region, or gene conversion between the 3'-truncated Juno4 and another telomere. The fosmid containing Juno5, initially selected by hybridization to the A. vaga telomeric repeat probe, contains a hAT-like transposase and could also be located in a subtelomeric region. In addition, a gag fragment was found on a P. roseola telomere adjacent to telomeric repeats in our telomere cloning experiments (E.G. and I.A., submitted).
Given the low copy number, such high incidence of occurrence in telomeric regions appears non-random. Any insertion site specificity at the level of nucleotide sequence, however, may be ruled out, since the TSDs and the immediate 0.6–0.7 kb flanking regions of the remaining Juno copies exhibit no sequence similarity, although the possibility of regional insertion specificity, broadly targeted towards specific chromatin domains, remains open. The sequenced flanking regions of the remaining Junos reveal no matches to known genes, with the exception of the most divergent copy, Juno1 (96–97% identity to other four copies). It is inserted into an apparent ORF with homology to ligand-gated ion channel receptors, a multigene family containing 30–60 members in most eukaryotic species (Chiu et al., 2002; Iwama and Gojobori, 2002). Overall, as contrasted with the high density of structural genes seen in more proximal regions of bdelloid genomes (D. Mark Welch, J. Mark Welch and M. Meselson, unpublished), their lack in sequenced regions surrounding Vesta and Juno argues in favor of their enrichment in subtelomeric heterochromatin, an environment highly enriched in other transposable elements as well, including Athena retroelements and DNA transposons (Arkhipova and Meselson 2005b; E.G. and I.A., submitted).
>4. Discussion
4. Discussion
In this study, we identified and characterized two LTR retrotransposons in bdelloid rotifers of two species, P. roseola and A. vaga, belonging to different families that separated tens of millions of years ago. Both retrotransposons exhibit typical structural features of retroviruses, including the presence of the third ORF corresponding to an env-like region. The env-like genes, with their potential to mediate horizontal entry into host cells, have been acquired by LTR retrotransposons at least six times, in three cases from a viral source (Malik et al., 2000). The well-studied insect errantiviruses from the gypsy clade were shown to have acquired their env region from baculoviruses, and the env regions of the nematode Cer and Tas retroviruses were acquired from a phlebovirus and a herpesvirus, respectively. So far, only one virus from aquatic invertebrates has been sequenced, a large DNA virus that can pass between rotifer and shrimp hosts (Yan et al., 2004). Although its ORFs do not reveal similarities with bdelloid env-like genes, additional viruses inhabiting aquatic invertebrates will hopefully be characterized in the future (e.g. Comps et al., 1991).
The simplest explanation of our findings, consistent with the presence of an env-like region in both elements, is that Vesta and Juno entered their respective bdelloid hosts as retroviruses from unidentified donors sometime after P. roseola and A. vaga separated. The identity of the left and right LTRs in Vesta and in four out of five Junos indicates that all of them inserted relatively recently, consistent with our failure to find decayed copies. In contrast, the synonymous site divergence between different copies of Juno, ranging up to 5% or more, suggests that they have been diverging for a much longer time, apparently in their donor host(s), before entering A. vaga. Consistent with the Ka/Ks values, purifying selection would have operated on ORFs 1 and 2 during retrotransposition in their previous host, in the production of infective viruses, and during whatever limited transposition may have occurred within A. vaga. Being analogous to antigenic sites, the env coding regions may have been under both purifying and positive selection as in retroviruses, resulting in Ka/Ks values near unity.
The low copy numbers of Vesta and Juno (Fig. 3) and their solo LTRs (data not shown) may have several explanations, including relatively recent arrival in their present hosts. But our failure so far to discover other LTR retrotransposons suggests that additional factors may be at work, tending to deplete bdelloid genomes of retrotransposons generally. One possibility is suggested by the recent discovery that bdelloid rotifers are highly resistant to ionizing radiation (E.G. and M.M., unpublished) and the implication that such resistance is an adaptation to repair high levels of DNA breakage experienced during the desiccation and rehydration to which bdelloids are frequently exposed in their ephemerally aquatic habitats. Such repair may favor ectopic crossing-over of repeated elements such as transposons, with selection then acting against inviable translocations. Although Vesta and Juno, like other retroviruses and LTR retrotransposons, may have insertion site preferences determined by tethering to specific chromosomal proteins (Bushman, 2003), e.g. similar to yeast Ty5 (Xie et al., 2001), ectopic crossing-over could also account for the preferential telomeric localization of these elements and of bdelloid DNA transposons (Arkhipova and Meselson, 2005b), as rearrangements confined to regions lacking essential genes proximal to the crossover breakpoints would presumably not be detrimental.
 
Acknowledgments
We thank the US National Science Foundation (MCB-0614142) and the National Institutes of Health (5R01GM072708-02) for financial support.
 
Abbreviations
aaamino acid(s)
bpbase pair(s)
INintegrase
kbkilobase(s)
LTRlong terminal repeat
Myrmillion years
ORFopen reading frame
RTreverse transcriptase
TEtransposable element
TMtransmembrane
UTRuntranslated region

 
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
>References
References
  • Arkhipova, IR; Mazo, AM; Cherkasova, VA; Gorelova, TV; Schuppe, NG; Ilyin, YV. The steps of reverse transcription of Drosophila mobile dispersed genetic elements and U3-R-U5 structure of their LTRs. Cell. 1986;44:555–563. [PubMed]
  • Arkhipova, IR; Lyubomirskaya, NV; Ilyin, YV. Drosophila retrotransposons. Austin, TX: RG Landes Co; 1995.
  • Arkhipova, IR. Promoter elements in Drosophila melanogaster revealed by sequence analysis. Genetics. 1995;139:1359–1369. [PubMed]
  • Arkhipova, I; Meselson, M. Transposable elements in sexual and ancient asexual taxa. Proc Natl Acad Sci USA. 2000;97:14473–14477. [PubMed]
  • Arkhipova, I; Meselson, M. Deleterious transposable elements and the extinction of asexuals. Bioessays. 2005a;27:76–85. [PubMed]
  • Arkhipova, IR; Meselson, M. Diverse DNA transposons in rotifers of the class Bdelloidea. Proc Natl Acad Sci USA. 2005b;102:11781–11786. [PubMed]
  • Arkhipova, IR; Pyatkov, KI; Meselson, M; Evgen'ev, MB. Retroelements containing introns in diverse invertebrate taxa. Nat Genet. 2003;33:123–124. [PubMed]
  • Bae, YA; Moon, SY; Kong, Y; Cho, SY; Rhyu, MG. CsRn1, a novel active retrotransposon in a parasitic trematode, Clonorchis sinensis, discloses a new phylogenetic clade of Ty3/gypsy-like LTR retrotransposons. Mol Biol Evol. 2001;18:1474–1483. [PubMed]
  • Barton, NH; Charlesworth, B. Why sex and recombination? Science. 1998;281:1986–1990. [PubMed]
  • Bell, G. The Masterpiece of Nature: The Evolution and Genetics of Sexuality. Berkeley: University of California Press; 1982.
  • Boeke, JD; Eickbush, TH; Sandmeyer, SB; Voytas, DF. Virus Taxonomy: VIIth Report of the International Committee on Taxonomy of Viruses. Murphy FA. , editor. New York: Springer-Verlag; 1999.
  • Boeke, JD; Eickbush, T; Sandmeyer, SB; Voytas, DF. Metaviridae. In: Fauquet CM. , editor. Virus Taxonomy: VIIIth Report of the International Committee on Taxonomy of Viruses. New York: Academic Press; 2004.
  • Bushman, F. Targeting survival: Integration site selection by retroviruses and LTR retrotransposons. Cell. 2003;115:135–138. [PubMed]
  • Chiu, JC; Brenner, ED; DeSalle, R; Nitabach, MN; Holmes, TC; Coruzzi, GM. Phylogenetic and expression analysis of the glutamate-receptor-like gene family in Arabidopsis thaliana. Mol Biol Evol. 2002;19:1066–1082. [PubMed]
  • Coffin, JM; Hughes, SH; Varmus, HE. Retroviruses. Cold Spring Harbor Laboratory Press, Cold Spring Harbor; New York: 1997.
  • Comps, M; Mari, J; Poisson, F; Bonami, JR. Biophysical and biochemical properties of an unusual birnavirus pathogenic for rotifers. J Gen Virol. 1991;72:1229–1236. [PubMed]
  • Friesen, PD; Nissen, MS. Gene organization and transcription of TED, a lepidopteran retrotransposon integrated within the baculovirus genome. Mol Cell Biol. 1990;10:3067–3077. [PubMed]
  • Glockner, G; Szafranski, K; Winckler, T; Dingermann, T; Quail, MA; Cox, E; Eichinger, L; Noegel, AA; Rosenthal, A. The complex repeats of Dictyostelium discoideum. Genome Res. 2001;11:585–594. [PubMed]
  • Goodwin, TJ; Poulter, RT. A group of deuterostome Ty3/gypsy-like retrotransposons with Ty1/copia-like pol-domain orders. Mol Genet Genomics. 2002;267:481–491. [PubMed]
  • Heyman, T; Agoutin, B; Friant, S; Wilhelm, FX; Wilhelm, ML. Plus-strand DNA synthesis of the yeast retrotransposon Ty1 is initiated at two sites, PPT1 next to the 3' LTR and PPT2 within the pol gene. PPT1 is sufficient for Ty1 transposition. J Mol Biol. 1995;253:291–303. [PubMed]
  • Iwama, H; Gojobori, T. Identification of neurotransmitter receptor genes under significantly relaxed selective constraint by orthologous gene comparisons between humans and rodents. Mol. Biol. Evol. 2002;19:1891–1901.
  • Johnson, WE; Coffin, JM. Constructing primate phylogenies from ancient retrovirus sequences. Proc Natl Acad Sci USA. 1999;96:10254–102560. [PubMed]
  • Kondrashov, AS. Classification of hypotheses on the advantage of amphimixis. J Hered. 1993;84:372–387. [PubMed]
  • Kordis, D. A genomic perspective on the chromodomain-containing retrotransposons: Chromoviruses. Gene. 2005;347:161–173. [PubMed]
  • Kumar, S; Tamura, K; Nei, M. MEGA3: Integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics. 2004;5:150–163. [PubMed]
  • Li, WH. Molecular Evolution. Sinauer; Sunderland, MA: 1997.
  • Malik, HS; Eickbush, TH. Modular evolution of the integrase domain in the Ty3/Gypsy class of LTR retrotransposons. J Virol. 1999;73:5186–5190. [PubMed]
  • Malik, HS; Eickbush, TH. Phylogenetic analysis of ribonuclease H domains suggests a late, chimeric origin of LTR retrotransposable elements and retroviruses. Genome Res. 2001;11:1187–1197. [PubMed]
  • Malik, HS; Henikoff, S; Eickbush, TH. Poised for contagion: evolutionary origins of the infectious abilities of invertebrate retroviruses. Genome Res. 2000;10:1307–1318. [PubMed]
  • Marin, I; Llorens, C. Ty3/Gypsy retrotransposons: description of new Arabidopsis thaliana elements and evolutionary perspectives derived from comparative genomic data. Mol. Biol. Evol. 2000;17:1040–1049.
  • Mark Welch, D; Meselson, M. Evidence for the evolution of bdelloid rotifers without sexual reproduction or genetic exchange. Science. 2000;288:1211–1215. [PubMed]
  • Mark Welch, JL; Mark Welch, DB; Meselson, M. Cytogenetic evidence for asexual evolution of bdelloid rotifers. Proc Natl Acad Sci USA. 2004;101:1618–1621. [PubMed]
  • Misseri, Y; Labesse, G; Bucheton, A; Terzian, C. Comparative sequence analysis and predictions for the envelope glycoproteins of insect endogenous retroviruses. Trends Microbiol. 2003;11:253–256. [PubMed]
  • Normark, BB; Judson, O; Moran, N. Genomic signatures of ancient asexual lineages. Biol. J. Linn. Soc. 2003;79:69–84.
  • Otto, SP; Lenormand, T. Resolving the paradox of sex and recombination. Nature Rev Genet. 2002;3:252–260. [PubMed]
  • Pamilo, P; Bianchi, NO. Evolution of the Zfx and Zfy genes: rates and interdependence between the genes. Mol Biol Evol. 1993;10:271–281. [PubMed]
  • Ronquist, F; Huelsenbeck, JP. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinformatics. 2003;19:1572–1574. [PubMed]
  • Smale, ST; Baltimore, D. The "initiator" as a transcription control element. Cell. 1989;57:103–113. [PubMed]
  • Spana, C; Harrison, DA; Corces, VG. The Drosophila melanogaster suppressor of Hairy-wing protein binds to specific sequences of the gypsy retrotransposon. Genes Dev. 1988;2:1414–1423. [PubMed]
  • Wilson, S; Matyunina, LV; McDonald, JF. An enhancer region within the copia untranslated leader contains binding sites for Drosophila regulatory proteins. Gene. 1998;209:239–246. [PubMed]
  • Wright, DA; Voytas, DF. Athila4 of Arabidopsis and Calypso of soybean define a lineage of endogenous plant retroviruses. Genome Res. 2002;12:122–131. [PubMed]
  • Xie, W; Gai, X; Zhu, Y; Zappulla, DC; Sternglanz, R; Voytas, DF. Targeting of the yeast Ty5 retrotransposon to silent chromatin is mediated by interactions between integrase and Sir4p. Mol Cell Biol. 2001;21:6606–6614. [PubMed]
  • Xiong, Y; Eickbush, TH. Origin and evolution of retroelements based upon their reverse transcriptase sequences. EMBO J. 1990;9:3353–3362. [PubMed]
  • Yan, DC; Dong, SL; Huang, J; Yu, XM; Feng, MY; Liu, XY. White spot syndrome virus (WSSV) detected by PCR in rotifers and rotifer resting eggs from shrimp pond sediments. Dis Aquat Organ. 2004;59:69–73. [PubMed]