NeSL-1, an Ancient Lineage of Site-Specific Non-LTR Retrotransposons From Caenorhabditis elegans

Corresponding author: Thomas H. Eickbush, Department of Biology, Hutchison Hall, University of Rochester, Rochester, NY 14627., eick{at}uhura.cc.rochester.edu (E-mail)

Communicating editor: J. A. BIRCHLER

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Phylogenetic analyses of non-LTR retrotransposons suggest that all elements can be divided into 11 lineages. The 3 oldest lineages show target site specificity for unique locations in the genome and encode an endonuclease with an active site similar to certain restriction enzymes. The more "modern" non-LTR lineages possess an apurinic endonuclease-like domain and generally lack site specificity. The genome sequence of Caenorhabditis elegans reveals the presence of a non-LTR retrotransposon that resembles the older elements, in that it contains a single open reading frame with a carboxyl-terminal restriction-like endonuclease domain. Located near the N-terminal end of the ORF is a cysteine protease domain not found in any other non-LTR element. The N2 strain of C. elegans appears to contain only one full-length and several 5' truncated copies of this element. The elements specifically insert in the Spliced leader-1 genes; hence the element has been named NeSL-1 (Nematode Spliced Leader-1). Phylogenetic analysis confirms that NeSL-1 branches very early in the non-LTR lineage and that it represents a 12th lineage of non-LTR elements. The target specificity of NeSL-1 for the spliced leader exons and the similarity of its structure to that of R2 elements leads to a simple model for its expression and retrotransposition.

TRANSPOSABLE elements have had a profound effect in the shaping of eukaryotic genomes. The deleterious nature of transposable elements is well documented as they constitute a significant fraction of spontaneously arising phenotypic mutations (MACKAY 1989 ; KAZAZIAN and MORAN 1998 ). In addition, certain essential features of eukaryotic genomes appear to owe their origins to transposable elements. Examples of these may include the nuclear spliceosomal machinery (SHARP 1991 ), telomerase (NAKAMURA et al. 1997 ), and immunoglobulin gene rearrangement (AGRAWAL et al. 1998 ). There are even examples of the apparent domestication of a transposable element. The latter include the HeT-A and TART retrotransposons, which have taken over the role of telomerase in Drosophila melanogaster (LEVIS et al. 1993 ; PARDUE et al. 1996 ).

Transposable elements employ different strategies for their evolutionary persistence. DNA-mediated elements like the P and mariner elements rely on cross-species transfers for their propagation, as they are otherwise destined for extinction by mutational inactivation (CLARK et al. 1994 ; HARTL et al. 1997 ; ROBERTSON 1997 ). Among the transposable elements that transpose via an RNA intermediate, both horizontal transfer and short-term vertical inheritance have been documented for long terminal repeat (LTR) bearing retrotransposons (SPRINGER and BRITTEN 1993 ; SONG et al. 1994 ; GONZALEZ and LESSIOS 1999 ). Surprisingly in the case of non-LTR retrotransposable elements (elements that lack long terminal repeats), a recent study found no evidence for horizontal transfers in the data available to date (MALIK et al. 1999 ). Strict vertical inheritance poses an interesting challenge for non-LTR retrotransposons as any chance elimination would be permanent. In spite of this precarious position, the abundance of non-LTR element lineages in most phyla suggests the ability to avoid elimination. For example, extensive studies of the R1 and R2 elements that insert into the 28S ribosomal genes of arthropod species have revealed only one elimination event: the loss of R2 in a lineage of Drosophila that is itself nearly extinct (EICKBUSH and EICKBUSH 1995 ; LATHE et al. 1995 ; LATHE and EICKBUSH 1997 ; BURKE et al. 1998 ).

Because of its reliance on vertical descent, the choice of target sites in the host genome becomes one of paramount importance for the non-LTR element. Consistent with this assumption, a number of non-LTR lineages have specialized to insert into what appear to be particularly advantageous locations in the host genome. On the basis of a phylogenetic analysis of the reverse transcriptase (RT) domain of all known non-LTR elements, it was inferred that these site-specific non-LTR elements were ancestral and encoded an endonuclease domain similar to that of certain restriction enzymes (MALIK et al. 1999 ; YANG et al. 1999 ). The monophyletic acquisition of a domain similar to the apurinic endonucleases (AP-endonucleases; MARTIN et al. 1995 ; FENG et al. 1996 ) was concomitant with the subsequent loss of site specificity in most but not all non-LTR element lineages (MALIK et al. 1999 ). Here we present an analysis of a new site-specific non-LTR retrotransposon found in the Caenorhabditis elegans genome. The structure and phylogenetic location of this element, named NeSL-1, are precisely as predicted on the basis of the proposed evolutionary scheme of non-LTR retrotransposons.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Sequence and phylogenetic analysis:
Sequence analysis was carried out using the MacVector package of programs (IBI Technologies). The open reading frame (ORF) was obtained from the conceptual translation of DNA sequence (GenBank accession no. Z82058). Additional NeSL-1 copies were obtained using BLASTN (ALTSCHUL et al. 1997 ) searches in the C. elegans genome using E values of 0.05 or lower. Comparisons of the different NeSL-1 clones to each other, as well to other non-LTR elements, were carried out using the pairwise and multiple alignment features of the CLUSTALX package of programs (THOMPSON et al. 1997 ). Identification of the cysteine protease domains was conducted with PSI-BLAST searches (ALTSCHUL et al. 1997 ). Phylogenetic analyses were conducted using neighbor-joining (SAITOU and NEI 1987 ) with both the default (identity) matrix in PAUP* and the PAM250 matrix of PHYLIP (FELSENSTEIN 1993 ), as well as maximum parsimony methods (heuristic options with swapping carried out using tree-bisection-reconnection routines with up to five trees held at each step). Bootstrap analysis was implemented using the PAUP* (beta2 version) package (SWOFFORD 1999 ).

Genomic blot protocols:
The N2 stock was obtained from the Caenorhabditis Genetics Center, University of Minnesota. Genomic DNA was isolated as outlined by the Ambros Laboratory at Dartmouth College (http://www.dartmouth.edu/artsci/bio/ambros/protocols.html). The worms were grown on agarose plates with OP50 bacteria, rinsed off the plates in M9 buffer, and pelleted in a clinical centrifuge. The worms were then floated on cold 60% sucrose and pelleted in TEN buffer (20 mM Tris, pH 7.5, 50 mM EDTA, 100 mM NaCl). This pellet was resuspended in TEN buffer along with 0.5% SDS, 0.1 mg/ml Proteinase K, and 1 µl of ß-mercaptoethanol and incubated at 55° with frequent resuspension. After 3 hr, the solution was extracted with phenol/chloroform (isoamyl alcohol) and the genomic DNA spooled in cold ethanol. The genomic DNA was digested with restriction enzymes EcoRI and HindIII, electrophoresed on 1% agarose gels, blotted onto nitrocellulose, and probed with labeled PCR products specific to either the 5' or 3' ends of the NeSL-1 element. The 5' end probe (nucleotides 1–922) was generated by PCR with primers 5'-CTTCTCATTGCACAATCCACA-3' and 5'-GCTCACTTTCTATCGTGTT-3', and the 3' probe (nucleotides 6631–7006) was generated with the primers 5'-CGCTGGAACGAAAAGAACGG-3' and 5'-GGCAAGAACCCGAATTATC-3'.

PCR amplification, cloning, and sequencing protocol:
To obtain potentially different 5' junctions of the NeSL-1 element with its target site, PCR was conducted using one primer complementary to the most upstream cysteine-histidine motif, 5'-CTTCTCATTGCACAATCCACA-3' within the element, in conjunction with a primer in the target site, 5'-AACGTGATATGGTCGTAAGC-3'. The product was cloned (BURKE et al. 1995 ) and individual clones were sequenced.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Structure of NeSL-1 elements:
Screens of reverse transcriptase-like sequences in the C. elegans genome have revealed numerous copies of non-LTR elements belonging to either the RTE lineage (MALIK and EICKBUSH 1998 ) or the CR1 lineage (MARIN et al. 1998 ). However, two RT sequences (nucleotide accession nos. Z82058 and AL031628) did not appear to be closely related to either of these groups of non-LTR elements. The longer of these two elements (Z82058) encodes a single ORF of 1693 amino acids. We call this element NeSL-1. The first methionine codon encountered in this ORF is the seventh codon, suggesting a possible start site for protein translation; however, translation does not necessarily begin at the first methionine codon of a non-LTR element ORF (see, for example, MALIK and EICKBUSH 1998 ; GEORGE and EICKBUSH 1999 ). The shorter element (AL031628) encodes a similar ORF (<5% amino acid divergence) beginning at codon 440 and continuing uninterrupted to the end of the ORF.

A comparison of the known features of the NeSl-1 ORF with that of several other non-LTR elements is shown in Fig 1A. R2 (BURKE et al. 1999 ), R4 (BURKE et al. 1995 ), and Dong (XIONG and EICKBUSH 1993 ) elements insert specifically in the ribosomal DNA units of arthropods or nematodes, while the CRE1 (GABRIEL et al. 1990 ), CRE2 (TENG et al. 1995 ), SLACS (AKSOY et al. 1990 ), and CZAR (VILLANUEVA et al. 1991 ) elements insert in the miniexon arrays of trypanosomes. The single ORF of the NeSL element includes a centrally located RT domain with excellent agreement (data not shown) to the 11 highly conserved motifs identified in the RT domain of all non-LTR elements (MALIK et al. 1999 ). The large region of the NeSL ORF upstream of the RT domain contains two closely spaced CCHH type cysteine-histidine motifs near the aminoterminal end that are similar to prototypic TFIIIA DNA-binding motifs (BERG and SHI 1996 ). While the N-terminal domains of the elements in Fig 1A are highly variable in length, as shown in Fig 1B, the CCHH motifs of NeSL are similar to the putative nucleic acid binding motifs found upstream of the RT domains in CRE1, CRE2, CZAR, SLACS, and R2 elements. R4 and Dong elements, on the other hand, contain no obvious nucleic acid binding motifs (XIONG and EICKBUSH 1993 ; BURKE et al. 1995 ).

View larger version (52K): In this window In a new window Download PPT slide   Figure 1. Domain structure of the NeSL-1 open reading frame. (A) ORF structure of the site-specific non-LTR elements that contain an endonuclease domain downstream of their RT domain. The CCHC and PD..D motifs associated with the endonuclease are shaded. The cysteine protease domain in NeSL-1 is abbreviated PR. Putative nucleic acid binding cysteine-histidine motifs upstream of the RT domain are numbered and their sequences are shown in B. In addition to either one or three cysteine-histidine motifs, R2 elements contain a putative c-myb DNA-binding motif (BURKE et al. 1999 ). CZAR and SLACS elements contain two ORFs that are in the same frame but separated by one or two termination codons (thin lines). The sequenced CRE1 element appears to be a 5' truncated copy as the ORF extends to the end of the element (GABRIEL et al. 1990 ). (B) Summary of the N-terminal cysteine-histidine motifs found in the non-LTR elements shown in A. Motifs are numbered starting from the RT domain as in A. (C) Comparison of the putative endonuclease domain of NeSL-1 to that of other site-specific non-LTR elements. The domain definition includes the C-terminal CCHC motif, as well as conserved RH(N/D)..(RK)PD..D..K..Y residues shown in reverse type, abbreviated PD..D.

The separation of the two CCHH motifs from the RT domain of the NeSL-1 element is significantly longer than that found in the other elements shown in Fig 1A. We have therefore used this sequence from NeSL-1 to search for sequence homology in the various protein databases (see MATERIALS AND METHODS). As shown in Fig 2, significant matches were found to a series of cysteine proteases. The best characterized of these are the Smt-3 and Smt-4 proteases, which remove ubiquitin-like motifs bound to proteins (LI and HOCHSTRASSER 1999 ). All four of the critical residues that are believed to be associated with the protease's active site (solid arrows) are conserved in the NeSL-1 element. This is the first report of a protease domain in a non-LTR retrotransposable element. The protease domain is presented in the 5' truncated copy of NeSL-1 (AL031128) confirming that this domain accompanies NeSL in its retrotransposition.

View larger version (93K): In this window In a new window Download PPT slide   Figure 2. Sequence alignment of the NeSL-1 putative cysteine protease domain to that of other cysteine proteases. The sequence comparison is similar to that in LI and HOCHSTRASSER 1999 . Solid arrows, critical histidine, aspartate, and cysteine residues at the catalytic site of the protein and conserved glutamine residues predicted to form the oxyanion hole of the active site. Open arrows, additional invariant residues. Accession nos. are as follows: Schizosaccharomyces pombe: Smt3p (CAA17063.1), Smt4p (CAB11507), and unknown (BAA21400); D. melanogaster: Smt3p-like (AF145608.1); Saccharomyces cerevisiae: Smt3p (S63462), Smt4p (886766); Homo sapiens: a, Smt3p-like (CAB43384.1), b, Smt4p-like (BAA34517.1); Arabidopsis thaliana: Smt3p-like (3047118); C. elegans: a (Q09275), b (1280153), c (4226115), NeSL-1 (CAB04870).

Downstream of the RT domain the elements shown in Fig 1A are uniform. This region contains (Fig 1C) a series of conserved residues that are present in all non-LTR elements that lack the AP-endonuclease domain found in most non-LTR elements (MARTIN et al. 1995 ; FENG et al. 1996 ). Two critical components of this sequence conservation are a cysteine-histidine motif of the structure CCHC and a PD..D motif. This latter motif consists of two aspartate residues separated by a 12- or 14-amino-acid sequence predicted to form a ß-turn (YANG et al. 1999 ). This motif has been shown to be present in certain restriction enzymes (SELENT et al. 1992 ; WAUGH and SAUER 1993 ) and lies in close proximity to the scissile phosphoester bonds of the protein:DNA restriction site complexes (WINKLER et al. 1993 ; WAH et al. 1997 ). In the case of the R2 element, site-directed mutagenesis has confirmed that the first aspartate is part of the endonuclease of the element (YANG et al. 1999 ).

Given the variations in structure at the N-terminal end of the elements shown in Fig 1A and the fact that only a single long element was found, it is not known whether the NeSL-1 element in cosmid Z82058 represents a "full-length" element. However, the presence of a large 5' untranslated region (described below) and the similarity of its ORF to those in R2, R4, and Dong elements to which NeSL-1 is most related (phylogenetic analysis below) suggests that the copy in clone Z82058 probably represents a full-length NeSL-1 element.

NeSL-1 is site specific for the spliced leader genes of C. elegans:
All previously described non-LTR elements with an endonuclease motif at the C-terminal end of their encoded protein insert into specific sites in the genome. The nucleotide sequences flanking both of the initially identified copies of NeSL-1 were found to be homologous to a ~1-kb repeat in C. elegans containing the spliced leader 1 exon (SL1; KRAUSE and HIRSH 1987 ). This 1-kb repeat also contains a divergently transcribed 5S ribosomal RNA gene and is organized into large tandem arrays in the genome (NELSON and HONDA 1985 ). Surprisingly, while the 5' truncated NeSL-1 copy in clone AL031628 was located in a tandem array of 1-kb SL1/5S repeats, the putative full-length NeSL-1 sequence in clone Z82058 was located in an isolated (solo) repeat (Fig 3A). Comparison of the SL1 repeats with and without the NeSL-1 insertions suggested that both elements had inserted 2 bp downstream of the 22-bp spliced leader exon, sometimes referred to as a miniexon (Fig 3B). Identification of this target site allowed the delineation of the probable full-length NeSL-1 element in cosmid Z82058. The element is 7028 bp in length with the 1693-amino-acid ORF flanked by a 376-bp 5' untranslated region and an unusually long 1570-bp 3' untranslated region. The NeSL-1 element in cosmid AL031628 is more complicated as it was part of a complex of three tandemly arranged NeSL-1 copies (Fig 3A). The 3' most copy of this array is truncated at its 5' end by 1755 bp but is otherwise normal. The two NeSL-1 copies located 5' to this element are truncated at both ends. Given our current understanding of non-LTR retrotransposition we suggest that the current organization of these tandem copies probably arose by recombination between NeSL-1 elements in different SL1 repeats, rather than by the insertion of three NeSL-1 copies into the same SL1 repeat. These three elements are located at one end of a tandem array of at least 12 SL repeats. The fourth SL1 repeat in this array also contains a NeSL-1 element. While this copy is a highly 5' truncated version, it is located in the identical position of the spliced leader repeat. NeSL-1 insertions result in a 5-bp target site deletion, nucleotides 3–7 downstream of the exon (Fig 3B). The 3' end of NeSL-1 is a precise GC-rich sequence unlike the variable length A-rich tails seen in many other non-LTR elements.

View larger version (33K): In this window In a new window Download PPT slide   Figure 3. The location of NeSL-1 within the SL1 repeats of C. elegans. (A) Diagram of the NeSL-1 elements within SL1 repeats. The ~1-kb SL1 repeat is defined as starting at the BamHI site in each repeat (short vertical lines). The SL1 gene within the repeat is shown as a solid arrow and the oppositely transcribed 5S gene as an open arrow. NeSL-1 sequences are indicated by stippled boxes, with the first and last nucleotide of the element indicated based on the NeSL-1 element in Z82058. Diagonally shaded boxes, additional (unrelated) insertions within the SL1 repeat. Solid boxes, flanking non-SL1 repeat sequences. Only a single SL1 repeat is in clone Z82058. The tandem array of SL1 repeats is defined by the cosmid sequence AL031628 at its left end and cosmid sequence Z82085 at its right end with a potential 104 bp overlap between these two sequences. It is likely that additional SL1 repeats are present in this tandem array, as over 100 SL1 repeats have been estimated in C. elegans (NELSON and HONDA 1985 ). (B) Diagram of one 5S and SL1 repeat showing the NeSL-1 insertion site. The ~100-nt SL1 gene is divided into the 22-nt exon sequences that are spliced onto pre-mRNAs (vertical shading) and the remaining discarded sequences (solid shading). The nucleotide sequence surrounding the uninserted NeSL-1 target site and the junction sequences of the NeSL-1 copies are shown. A 5-bp deletion is associated with NeSL-1 insertions.

It should be noted that the SL repeats shown in Fig 3A are the only SL units of C. elegans to be sequenced to date. The two sequenced cosmids shown in Fig 3A clearly represent the two ends of an array of the 1-kb SL1/5S RNA repeats. However, the putative overlap between these two cosmids is only 104 bp and, because most SL repeats are nearly identical in sequence, it is thus likely that a large number of additional repeats are actually present in the array. Cosmids containing large tandem arrays of identical sequences are subject to internal deletions and thus are unstable.

To determine if additional copies of NeSL existed in the genome of C. elegans outside the SL1 repeats we used the complete 7028-bp NeSL sequence and identified five additional sequences (Z92976, Z29115, Z29094, AF000266, and AF106590) over 125 bp in length that contained significant (>75% nucleotide identity) homology to NeSL-1. All five NeSL-like sequences were not inserted into spliced leader repeats; however, one copy did contain the 22-bp leader exon sequence at its 5' end. Presumably, this 22-bp sequence was present on the RNA template used for reverse transcription (see DISCUSSION). All five copies of NeSL located outside the SL repeat contained an intact 3' end of the NeSL-1 element, suggesting that they arose by a non-LTR retrotransposition target primed reverse transcription mechanism (LUAN et al. 1993 ). However, these non-SL1 copies contained significant sequence divergence (often over 20%) as well as frequent deletions, suggesting that they represent very old insertion events. It should be noted that this large number of nonspecific insertions relative to specific insertions may result from their greater stability within the genome. NeSL-1 copies located outside the normal SL1 target site are not subject to the same recombinational forces associated within the concerted evolution of the tandemly repeated SL1 gene. As a consequence these nonspecific insertions can actually remain within the genome longer than those copies inserted within the normal SL1 target site. We have noted a similar tendency with R1 and R2 elements that occasionally insert outside the rDNA units of insects (XIONG et al. 1988 ; JAKUBCZAK et al. 1992 ).

NeSL-1 represents another ancient lineage of non-LTR retrotransposons:
A second prediction made by the identification of the restriction-like endonuclease domain in NeSL-1 (Fig 1C) is that this element should phylogenetically place close to the ancestral non-LTR elements that also bear this endonuclease domain (YANG et al. 1999 ). Because the only protein domain common to all non-LTR elements is the RT domain, we aligned the complete RT domain from NeSL-1 to that of other non-LTR elements (data not shown). The phylogenetic relationship of NeSL-1 to that of the 11 previously identified lineages of non-LTR elements is shown in Fig 4. While our analysis of the phylogenetic position of NeSL was based on a comprehensive analysis of all available non-LTR elements (MALIK et al. 1999 ), to simplify the figure only a limited number of taxa from each lineage is shown in Fig 4. The phylogenetic analysis clearly groups NeSL-1 close to other non-LTR lineages that contain the restriction-like endonuclease domain (bold lines). NeSL-1 also appears to form its own lineage that is phylogenetically distinct from the R2 and R4 lineages. For example, assuming a vertical inheritance model (MALIK et al. 1999 ), the divergence between the R4 and Dong elements in the R4 lineage is proposed to coincide with (or predate) that between Arthropoda and Nematoda. The divergence between the NeSL-1 element from C. elegans and any other non-LTR element is thus older than the split between these two phyla. This analysis establishes the NeSL-1 element from C. elegans as the first identified member of an ancient lineage of site-specific non-LTR retrotransposons.

View larger version (30K): In this window In a new window Download PPT slide   Figure 4. Phylogenetic position of NeSL-1 within the family of non-LTR retrotransposable elements. The amino acid sequences of the RT domains of NeSL-1 and all other non-LTR elements were aligned, and phylogenetic analyses were conducted using Group II intron-encoded RT domains as the outgroup as previously described (MALIK et al. 1999 ). Presented here is a neighbor-joining tree using the identity matrix in PAUP* with bootstrap values indicated in bold and only a sampling of the most highly divergent elements in each clade. Branches with <50% bootstrap support were collapsed. A neighbor-joining tree using the PAM250 matrix of PHYLIP gave the identical topology and nearly identical bootstrap values (not shown). Results of a maximum parsimony bootstrap analysis (see MATERIALS AND METHODS) are presented in italics, whenever >50%. Lineages bearing the PD..D restriction-like endonuclease domain are shown in bold. NeSL-1 clearly groups in with these lineages but is not a member of a previously defined clade of elements. A divergence scale is indicated. A complete phylogeny of all non-LTR elements can be found in MALIK et al. 1999 .

NeSL-1 copy number:
At the time of writing, 85% of the C. elegans genome has been sequenced (C. ELEGANS SEQUENCING CONSORTIUM 1998). While it has been estimated that there are ~110 copies of the SL1 repeat (NELSON and HONDA 1985 ) only 12 complete and several partial repeats have been sequenced to date. As stated above, it is likely that additional SL1 repeats are located within the middle of the array shown in Fig 3A.

To determine the true copy number of NeSL-1 in the C. elegans strain sequenced (N2), we conducted Southern blots (Fig 5) using as probes the 5' and 3' ends of the NeSL-1 element (see MATERIALS AND METHODS). To our surprise, while the 3' probe generated many hybridizing bands that could be largely corroborated with the genomic sequence available (open circles), the 5' probe hybridized to only one band, representing the copy that we had already analyzed. This band appears to represent a single NeSL-1 copy because the restriction enzymes used cleaved at unique locations outside the solo SL1 repeat of the sequenced copy. To further confirm this Southern analysis, we conducted PCR amplification of N2 genomic DNA to determine whether we could identify 5' junctions of different NeSL-1 copies (see MATERIALS AND METHODS). All 20 PCR products sequenced contained diagnostic nucleotide substitutions found upstream of the target sequences of the solo SL1 repeat in Z82058, supporting the conclusion that there was only one putative full-length NeSL-1 element in the genome of C. elegans. Because most of the genomic bands hybridizing to a 3' NeSL-1 probe (Fig 5) could be accounted for in the copies sequenced to date, we also predict only a limited number of additional 5' truncated copies are present in the many SL tandem arrays not sequenced.

View larger version (82K): In this window In a new window Download PPT slide   Figure 5. Southern blots of the N2 strain of C. elegans probed with NeSL-1 sequences. Genomic DNA was digested with either EcoRI or HindIII. The size and coordinates of the 5' UTR and 3' UTR probes are described in MATERIALS AND METHODS. The 3' UTR probe revealed multiple bands including those predicted for the sequenced full-length element (solid circle) and partial copies (open circles), as well as some that represent as-yet-unsequenced copies or restriction site polymorphisms (?). The 5' probe revealed only a single band (even in overexposures) consistent with that predicted for the single NeSL-1 element in sequence Z82058.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In this report we have identified a non-LTR retrotransposable element that specifically inserts into the spliced leader (SL1) exons of C. elegans. The NeSL-1 element is most related in structure and phylogeny to a group of non-LTR elements that insert into unique locations within the genome. These site-specific elements contain a C-terminal endonuclease domain rather than the N-terminal AP-like endonuclease domain found in the non-site-specific elements. NeSL-1 is not a member of any of the previously identified 11 clades of non-LTR elements, indicating that they represent a distinct lineage dating back at least 600 million years. The only unusual aspect of the NeSL-1 structure is the presence of a cysteine protease domain upstream of the RT domain in its single ORF. The best-characterized members of the cysteine proteases related to the NeSL-1 domain are the Smt-3 and -4 proteases of yeast, enzymes involved in removal of ubiquitin-like linkages from proteins (LI and HOCHSTRASSER 1999 ).

Two possible roles for the NeSL-1 protease can be postulated. The usual role of a protease encoded by a retrotransposable element is the processing of a polyprotein encoded by the element into its individual catalytic domains. Based on our current predictions of the domain structure and catalytic activity of the NeSL-1 ORF (see Fig 1), the protease would likely cleave the RT domain of the protein from the DNA-binding and/or endonuclease domains. This model would suggest that the other non-LTR elements with ORFs similar to that of NeSL-1 either do not need this processing or can utilize an endogenous cellular protease for such a function. The NeSL-1 protease domain would thus be similar to the RNase H domain found in some non-LTR elements but absent in close phylogenetic neighbors (MALIK et al. 1999 ). A more speculative role for the NeSL-1 protease domain is that it is not required for NeSL-1 activity but is rather required for one or more cellular functions of the host. Such a role would suggest that NeSL-1 had assumed a beneficial role for the host, in a manner similar to the assumption of telomerase function by the TART and HeT-A elements in D. melanogaster (LEVIS et al. 1993 ; PARDUE et al. 1996 ). Eukaryotes contain a number of different cysteine proteases [e.g., three such sequences were readily identified in the C. elegans genome (see Fig 2)]. Support for this model could be obtained by monitoring the effects of eliminating NeSL-1 activity or associating a specific cysteine protease found in related organisms to the NeSL-1 element in C. elegans.

Expression model for NeSL-1:
SL1 genes of C. elegans are transcribed from their own promoters and the ~100-nt RNA is modified with a 5' trimethyl-guanosine cap (LIOU and BLUMENTHAL 1990 ). This SL1 RNA is responsible for the processing of polycistronic messages into monocistronic units by trans-splicing the 22 nt at its 5' terminus onto the most 5' cistron of the polycistronic mRNA, thereby allowing cap-dependent translation (Fig 6A). SL1 genes are found in all nematodes examined with the 22-bp exon sequence itself exhibiting no sequence variation in all of Nematoda (BEKTESH et al. 1988 ; NILSEN et al. 1989 ; XIE et al. 1994 ). Additional spliced leader genes, including SL2, that have been found in C. elegans and other nematodes (ROSS et al. 1995 ; EVANS et al. 1997 ) are responsible for the processing of downstream cistrons in the polycistronic message. It has been estimated that up to 70% of the C. elegans mRNAs may require trans-splicing (ZORIO et al. 1994 ).

View larger version (67K): In this window In a new window Download PPT slide   Figure 6. Expression model for NeSL-1. (A) Trans-splicing in C. elegans. SL1 genes are transcribed separately from a normal gene, Gene X. Post-transcriptionally, the 22-bp spliced leader (thicker area) is trans-spliced to the 5' end of Gene X pre-mRNA, carrying with it the trimethyl-guanosine cap (star). After trans-splicing the mature mRNA is available for translation. (B) Expression of NeSL-1 elements. NeSL-1 is predicted to be cotranscribed with the SL1 gene and to acquire a trimethyl guanosine cap and is thus ready for translation. (C) Conservation of the SL1 gene in Nematoda. The SL1 genes (transcribed regions) of diverse nematodes are aligned. SL1 genes are identical in sequence in the region corresponding to the spliced leader itself (boxed) and an 11-nucleotide region located ~50 bp downstream (solid bars). The guanosine residue used in trans-splicing is enlarged, while the arrow shows the insertion site of NeSL-1 in C. elegans. Insertion of NeSL-1 is accompanied by a 5-bp deletion (boxed nucleotides AAACA) starting 2 bp downstream of the insertion site (see also Fig 4).

Based on this role of the SL exon, a simple model for the expression of NeSL-1 can be proposed. NeSL-1 elements do not need their own promoters because they can be cotranscribed with the SL1 gene and thus their RNA is immediately available for translation (Fig 6B). The integration of NeSL-1 into the target site is also readily modeled based on the similarity in domain structure between NeSL-1 and R2 elements. R2 elements insert by a mechanism termed target-primed reverse transcription (TPRT). Integration involves sequence-specific recognition and cleavage of the target site followed by polymerization of the reverse transcript (cDNA) directly onto the target site using the 3' hydroxyl group exposed by the cleavage as primer. Reverse transcription of the RNA starts at the 3' end of the element's RNA transcript with prematurely aborted reverse transcripts or degradation of the RNA template resulting in 5' truncated copies (LUAN et al. 1993 ). As shown in Fig 6C, the sequence of the 22-bp leader exon (and the following 3 bases) is identical in all nematodes. Therefore, as in the case of R2 (XIONG and EICKBUSH 1988 ), we suggest that DNA recognition of the target site by the NeSL-1 endonuclease is to the highly conserved sequences upstream of the insertion site. The initial cleavage (nick) by the NeSl-1 endonuclease would be on the noncoding strand of the SL1 gene at a location 7 bp downstream of the end of the leader exon while cleavage of the coding (RNA synonymous) strand of the target DNA would occur 5 bp upstream of the first nick (i.e., 2 bp downstream of the leader exon). Second-strand cleavage upstream of first-strand cleavage results in a 5-bp deletion (boxed residues in Fig 6C), similar to the 2-bp deletion generated in R2 element insertions. Recently we have suggested that attachment of the 5' end of the R2 reverse transcript to the upstream target site involves a recombination or template switch between homologous sequences of the target 28S gene and these same sequences on the newly made cDNA (EICKBUSH et al. 2000 ). An identical model postulating recombination between the 22 nt of the exon sequences at the 5' end of a SL1/NeSL-1 cotranscript and the target site could be argued. It is interesting to note that this model predicts that NeSL elements inserting outside the SL repeats might occasionally bring with them the 22-bp exon sequence. This was in fact found with one of the non-SL1 repeat insertions of NeSL-1 sequences (AF106590).

Spliced leader genes are thus an ideal "expression cassette" for a gene and are especially attractive for a transposable element to ensure its own expression. It is not surprising, in this context, that two different lineages of non-LTR elements have adopted spliced leader genes as their homes, one in trypanosomes (CRE1, CRE2, SLACS, and CZAR elements) and one in nematodes (NeSL-1). The trypanosome spliced leader-specific non-LTR elements also insert near the end of the leader exon (which in this case is 39 nt) and thus can also be cotranscribed and translated by virtue of the leader sequence. Integration of these trypanosome elements by TPRT differs from that of NeSL-1 only in the locations of the DNA cleavages of the two strands of the target site. Cleavage of the first (noncoding) strand in all four trypanosome elements can be predicted to occur following the 11th base of the exon. Cleavage of the second (coding) strand differs between the elements but is always at least 22 bp downstream of the first strand. Cleavage downstream in the TPRT model means that the sequences between the two cleavage sites become a target site duplication, and thus the entire leader exon is located at the 5' end of the newly inserted element.

Is NeSL-1 on its way out?
Growth of the hermaphroditic organism C. elegans in the laboratory has subjected it to strong bottlenecking. Such inbreeding imposes a strong challenge for a transposable element, and it has been suggested that transposable elements are lost during long-term laboratory rearing. This may well be occurring in C. elegans. Recent studies on the Sam and Frodo non-LTR elements (CR1 lineage) in the N2 C. elegans genome have uncovered many distinct lineages that appear to be present in one or two copies with most lineages having no active members (MARIN et al. 1998 ). In a similar manner the RTE2 lineage of C. elegans appears to have no active elements in the N2 strain (MALIK and EICKBUSH 1998 ).

Based on these previous findings, it is not too surprising that we have found only one putative intact copy of NeSL-1 in the N2 strain of C. elegans. Finding this copy in a SL1 repeat separate from the major tandem array of repeats is also not a surprise. The concerted evolution of the SL1 repeats within the tandem array will eliminate insertion sequences. The putative full-length NeSL-1 element located outside the SL1 locus, on the other hand, is protected from elimination and could thus be serving as a "master copy" for the insertion of new copies into the SL1 locus. The solo SL1 repeat itself is quite old as its sequence is over 10% divergent from the standard SL1 repeats and contains numerous minor and one major (115 bp) insertions and deletions. The full-length NeSL-1 copy in this solo repeat, on the other hand, contains only 0.7% nucleotide divergence with the highly truncated NeSL-1 element located within the fourth unit of the SL1 tandem array. Thus it is possible that the full-length element within the solo repeat could have recently given rise to an insertion within the SL1 array. Another possibility is that a full-length element within the array recently gave rise to both copies, but was itself subsequently eliminated by recombination. Sequence divergence between the full-length NeSL element and the three tandemly arranged NeSL-1 copies at the end of the SL1 array is >5%, suggesting that these copies have been accumulating mutations over an extended period of time.

"Homes" for the ancestral non-LTR elements:
The identification of the NeSL-1 element serves to highlight the choice made by many lineages of non-LTR elements to specialize for a target site that will cause little damage to the host. One could propose that in two separate cases, nematodes and trypanosomes, non-LTR elements have become specialized for spliced leader genes. The alternative model would postulate that the original non-LTR elements were all site specific for spliced leaders and all lineages except CRE and NESL-1 have switched their specificity elsewhere. It is not known whether the nematode SL1 exon itself is descended from trypanosomes. Trans-splicing of a leader exon has also been identified in flatworms (RAJKOVIC et al. 1990 ; DAVIS et al. 1994 ) and in Euglena (TESSIER et al. 1991 ). Further characterization of these systems and the identification of yet other non-LTR elements associated with this target site may help resolve this issue. In a similar manner, it is not clear whether the R2 and R4 lineages, both similar in age to that of NeSL-1 (see Fig 5) but which insert at different locations in the large rRNA gene, represent one specialization event and a change in sequence specificity or two separate specialization events. No doubt other multigene families could be potential homes for non-LTR elements, including the 5S ribosomal and transfer RNA genes. The many genomic sequencing projects currently being undertaken will undoubtedly reveal the diversity of residences adopted by this class of elements and eventually their pathways of evolution.

	FOOTNOTES

¹ Present address: Fred Hutchinson Cancer Research Center, 1100 Fairview Ave. N, Seattle, WA 98109.

	ACKNOWLEDGMENTS

We gratefully acknowledge the assistance of T. L. Stiernagle and colleagues. We especially thank Fred Hagen for his advice on culturing C. elegans, Brooke Schuster for help in the analysis of the C. elegans genomic sequences, and Bill Burke and Danna Eickbush for comments on the manuscript. Nematode strains used in this work were provided by the Caenorhabditis Genetics Center, which is funded by the National Institutes of Health National Center for Research Resources (NCRR). This work was supported by a National Science Foundation grant (MCB-9601198) to T.H.E.

Manuscript received August 16, 1999; Accepted for publication September 29, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AGRAWAL, A., Q. M. EASTMAN, and D. G. SCHATZ, 1998  Transposition mediated by RAG1 and RAG2 and its implications for the evolution of the immune system. Nature 394:744-751[Medline].

AKSOY, S., S. WILLIAMS, S. CHANG, and F. F. RICHARDS, 1990  SLACS retrotransposon from Trypanosoma brucei gambiense is similar to mammalian LINEs. Nucleic Acids Res. 18:785-792[Abstract/Free Full Text].

ALTSCHUL, S. F., T. L. MADDEN, A. A. SCHÄFFER, J. ZHANG, and Z. ZHANG et al., 1997  Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25:3389-3402[Abstract/Free Full Text].

BEKTESH, S., K. VAN DOREN, and D. HIRSH, 1988  Presence of the Caenorhabditis elegans spliced leader on different mRNAs and in different genera of nematodes. Genes Dev. 2:1277-1283[Abstract/Free Full Text].

BERG, J. M. and Y. SHI, 1996  The galvanization of biology: a growing appreciation for the roles of zinc. Science 271:1081-1085[Abstract].

BURKE, W. D., F. MÜLLER, and T. H. EICKBUSH, 1995  R4, a non-LTR retrotransposon specific to the large subunit rRNA genes of nematodes. Nucleic Acids Res. 23:4628-4634[Abstract/Free Full Text].

BURKE, W. D., H. S. MALIK, W. C. LATHE, III, and T. H. EICKBUSH, 1998  Are retrotransposons long-term hitchhikers? Nature 392:141-142[Medline].

BURKE, W. D., H. S. MALIK, J. P. JONES, and T. H. EICKBUSH, 1999  Conserved structure and mechanism of integration of the R2 retrotransposable element in all arthropods. Mol. Biol. Evol. 16:502-511[Abstract].

Genome sequence of the nematode C. elegans: a platform for investigating biology. (1998) Science 282:2012-2018[Abstract/Free Full Text].

CLARK, J. B., W. P. MADDISON, and M. G. KIDWELL, 1994  Phylogenetic analysis supports horizontal transfer of P transposable elements. Mol. Biol. Evol. 11:40-50[Abstract].

DAVIS, R. E., H. SINGH, C. BOTKA, C. HARDWICK, and M. ASHRAF EL MEANAWY et al., 1994  RNA trans-splicing in Fasciola hepatica. Identification of a spliced leader (SL) RNA and SL sequences on mRNAs. J. Biol. Chem. 269:20026-20030[Abstract/Free Full Text].

EICKBUSH, D. G. and T. H. EICKBUSH, 1995  Vertical transmission of the retrotransposable elements R1 and R2 during the evolution of the Drosophila melanogaster species subgroup. Genetics 139:671-684[Abstract].

EICKBUSH, D. G., D. D. LUAN, and T. H. EICKBUSH, 2000  Integration of R2 sequences from Bombyx mori into the 28S rDNA loci of Drosophila melanogaster.. Mol. Cell. Biol. in press.

EVANS, D., D. ZORIO, M. MACMORRIS, C. E. WINTER, and K. LEA et al., 1997  Operons and SL2 trans-splicing exist in nematodes outside the genus Caenorhabditis. Proc. Natl. Acad. Sci. USA 94:9751-9756[Abstract/Free Full Text].

FELSENSTEIN, J., 1993 PHYLIP (phylogeny inference package). Version 3.55. Distributed by the author, Department of Genetics, University of Washington, Seattle.

FENG, Q., J. V. MORAN, H. H. KAZAIAN, JR., and J. D. BOEKE, 1996  Human L1 retrotransposon encodes a conserved endonuclease required for retrotransposition. Cell 87:905-916[Medline].

GABRIEL, A., T. J. YEN, D. C. SCHWARTZ, C. L. SMITH, and J. D. BOEKE et al., 1990  A rapidly rearranging retrotransposon within the miniexon gene locus of Crithidia fasciculata.. Mol. Cell. Biol. 10:615-624[Abstract/Free Full Text].

GEORGE, J. A. and T. H. EICKBUSH, 1999  Conserved features at the 5' end of Drosophila R2 retrotransposable elements: implications for transcription and translation. Insect Mol. Biol. 8:3-10[Medline].

GONZALEZ, P. and H. A. LESSIOS, 1999  Evolution of sea urchin retroviral-like (SURL) elements: evidence from 40 echinoid species. Mol. Biol. Evol. 16:938-952[Abstract].

HARTL, D. L., A. R. LOHE, and E. R. LOZOVSKYA, 1997  Modern thoughts on an ancyent marinere: function, evolution, regulation. Annu. Rev. Genet. 31:337-358[Medline].

JAKUBCZAK, J. L., M. K. ZENNI, R. C. WOODRUFF, and T. H. EICKBUSH, 1992  Turnover of R1 (Type I) and R2 (Type II) retrotransposable elements in the ribosomal DNA of Drosophila melanogaster.. Genetics 131:129-142[Abstract].

KAZAZIAN, H. H. and J. V. MORAN, 1998  The impact of L1 retrotransposition on the human genome. Nat. Genet. 19:19-24[Medline].

KRAUSE, M. and D. HIRSH, 1987  A trans-spliced leader sequence on actin mRNA in C. elegans.. Cell 49:753-761[Medline].

LATHE, W. C., III and T. H. EICKBUSH, 1997  A single lineage of R2 retrotransposable elements is an active, evolutionarily stable component of the Drosophila rDNA locus. Mol. Biol. Evol. 14:1232-1241[Abstract].

LATHE, W. C., III, W. D. BURKE, D. G. EICKBUSH, and T. H. EICKBUSH, 1995  Evolutionary stability of the R1 retrotransposable element in the genus Drosophila. Mol. Biol. Evol. 12:1094-1105[Abstract].

LEVIS, R. W., R. GANESAN, K. HOUTCHENS, L. A. TOLAR, and F. M. SHEEN, 1993  Transposons in place of telomeric repeats at a Drosophila telomere. Cell 75:1083-1093[Medline].

LI, S. J. and M. HOCHSTRASSER, 1999  A new protease required for cell-cycle progression in yeast. Nature 398:246-251[Medline].

LIOU, R. F. and T. BLUMENTHAL, 1990  Trans-spliced Caenorhabditis elegans mRNAs retain trimethylguanosine caps. Mol. Cell. Biol. 10:1764-1768[Abstract/Free Full Text].

LUAN, D. D., M. H. KORMAN, J. L. JAKUBCZAK, and T. H. EICKBUSH, 1993  Reverse transcription of R2Bm RNA is primed by a nick at the chromosomal target site: a mechanism for non-LTR retrotransposition. Cell 72:595-605[Medline].

MACKAY, T., 1989  Transposable elements and fitness in Drosophila melanogaster.. Genome 31:284-295[Medline].

MALIK, H. S. and T. H. EICKBUSH, 1998  The RTE class of non-LTR retrotransposons is widely distributed in animals and is the origin of many SINEs. Mol. Biol. Evol. 15:1123-1134[Abstract].

MALIK, H. S., W. D. BURKE, and T. H. EICKBUSH, 1999  The age and evolution of non-LTR retrotransposons. Mol. Biol. Evol. 16:793-805[Abstract].

MARIN, I., P. PLATA-RENGIFO, M. LABRADOR, and A. FONTDEVILA, 1998  Evolutionary relationships among the members of an ancient class on non-LTR retrotransposons found in the nematode Caenorhabditis elegans.. Mol. Biol. Evol. 15:1390-1402[Free Full Text].

MARTIN, F., C. MARANON, M. OLIVARES, C. ALONSO, and M. C. LOPEZ, 1995  Characterization of a non-long terminal repeat retrotransposon cDNA (L1Tc) from Trypanosoma cruzi: homology of the first ORF with the Ape family of DNA repair enzymes. J. Mol. Biol. 247:49-59[Medline].

NAKAMURA, T. M., G. B. MORIN, K. B. CHAPMAN, S. L. WEINRICH, and W. H. ANDREWS et al., 1997  Telomerase catalytic subunit homologs from fission yeast and human. Science 277:955-959[Abstract/Free Full Text].

NELSON, D. W. and B. M. HONDA, 1985  Genes coding for 5S ribosomal RNA of the nematode Caenorhabditis elegans.. Gene 38:245-251[Medline].

NILSEN, T. W., J. SHAMBAUGH, J. DENKER, G. CHUBB, and C. FASER et al., 1989  Characterization and expression of a spliced leader RNA in the parasitic nematode Ascaris lumbricoides var. suum. Mol. Cell. Biol. 9:3543-3547[Abstract/Free Full Text].

PARDUE, M. L., O. N. DANILEVSKAYA, K. LOWENHAUPT, F. SLOT, and K. L. TRAVERSE, 1996  Drosophila telomeres: new views on chromosome evolution. Trends Genet. 12:48-52[Medline].

RAJKOVIC, A., R. E. DAVIS, J. N. SIMONSEN, and F. M. ROTTMAN, 1990  A spliced leader is present on a subset of mRNAs from the human parasite Schistosoma mansoni.. Proc. Natl. Acad. Sci. USA 87:8879-8883[Abstract/Free Full Text].

ROBERTSON, H. M., 1997  Multiple Mariner transposons in flatworms and hydras are related to those of insects. J. Hered. 88:195-201[Abstract/Free Full Text].

ROSS, L. H., J. H. FREEDMAN, and C. S. RUBIN, 1995  Structure and expression of novel spliced leader RNA genes in Caenorhabditis elegans.. J. Biol. Chem. 270:22066-22075[Abstract/Free Full Text].

SAITOU, N. and M. NEI, 1987  The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4:406-425[Abstract].

SELENT, U., T. RÜTER, E. KOHLER, M. LIEDTKE, and V. THIELKING et al., 1992  A site-directed mutagenesis study to identify amino acid residues involved in the catalytic function of the restriction endonuclease EcoRV. Biochemistry 31:4808-4815[Medline].

SHARP, P. A., 1991  Five easy pieces. Science 254:663[Free Full Text].

SONG, S. U., T. GERASIMOVA, M. KURKULOS, J. D. BOEKE, and V. G. CORCES, 1994  An env-like protein encoded by a Drosophila retroelement: evidence that gypsy is an infectious retrovirus. Genes Dev. 8:2046-2057[Abstract/Free Full Text].

SPRINGER, M. S. and R. J. BRITTEN, 1993  Phylogenetic relationships of reverse transcriptase and RNase H sequences and aspects of genome structure in the gypsy group of retrotransposons. Mol. Biol. Evol. 10:1370-1379[Abstract].

SWOFFORD, D. L., 1999 PAUP 4.0. Laboratory of Molecular Systematics. Smithsonian Institution, Washington, DC.

TENG, S.-H., S. X. WANG, and A. GABRIEL, 1995  A new non-LTR retrotransposon provides evidence for multiple distinct site-specific elements in Crithidia fasciculata miniexon arrays. Nucleic Acids Res. 23:2929-2936[Abstract/Free Full Text].

TESSIER, L. H., M. KELLER, R. L. CHAN, R. FOURNIER, and J. H. WEIL et al., 1991  Short leader sequences may be transferred from small RNAs to pre-mature mRNAs by trans-splicing in Euglena. EMBO J. 10:2621-2625[Medline].

THOMPSON, J. D., T. J. GIBSON, F. PLEWNIAK, F. JEANMOUGIN, and D. G. HIGGINS, 1997  The CLUSTAL-X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 25:4876-4882[Abstract/Free Full Text].

VILLANUEVA, M. S., S. P. WILLIAMS, C. B. BEARD, F. F. RICHARDS, and S. AKSOY, 1991  A new member of a family of site-specific retrotransposons is present in the spliced leader RNA genes of Trypanosoma cruzi.. Mol. Cell. Biol. 11:6139-6148[Abstract/Free Full Text].

WAH, D. A., J. A. HIRSCH, L. F. DORNER, I. SCHILDKRAUT, and A. K. AGGARWAL, 1997  Structure of the multimodular endonuclease FokI bound to DNA. Nature 388:97-100[Medline].

WAUGH, D. S. and R. T. SAUER, 1993  Single amino acid substitutions uncouple the DNA binding and strand scission activities of FokI endonuclease. Proc. Natl. Acad. Sci. USA 90:9596-9600[Abstract/Free Full Text].

WINKLER, F. K., D. W. BANNER, C. OEFNER, D. TSERNOGLOU, and R. S. BROWN et al., 1993  The crystal structure of EcoRV endonuclease and of its complexes with cognate and non-cognate DNA fragments. EMBO J. 12:1781-1795[Medline].

XIE, H., O. BAIN, and S. A. WILLIAMS, 1994  Molecular phylogenetic studies on filarial parasites based on 5S ribosomal spacer sequences. Parasite 1:141-151[Medline].

XIONG, Y. and T. H. EICKBUSH, 1988  Functional expression of a sequence-specific endonuclease encoded by the retrotransposon R2Bm. Cell 55:235-246[Medline].

XIONG, Y. and T. H. EICKBUSH, 1993  Dong, a non-long terminal repeat (non-LTR) retrotransposable element from Bombyx mori.. Nucleic Acids Res. 21:1318[Free Full Text].

XIONG, Y., W. D. BURKE, J. L. JAKUBCZAK, and T. H. EICKBUSH, 1988  Ribosomal DNA insertion elements R1Bm and R2Bm can transpose in a sequence specific manner to locations outside the 28S genes. Nucleic Acids Res. 16:10561-10573[Abstract/Free Full Text].

YANG, J., H. S. MALIK, and T. H. EICKBUSH, 1999  Identification of the endonuclease domain encoded by R2 and other site-specific, non-long terminal repeat retrotransposable elements. Proc. Natl. Acad. Sci. USA 96:7847-7852[Abstract/Free Full Text].

ZORIO, D. A., N. N. CHENG, T. BLUMENTHAL, and J. SPIETH, 1994  Operons as a common form of chromosomal organization in C. elegans.. Nature 372:270-272[Medline].

This article has been cited by other articles:

				J. L. Crainey, C. F. Garvey, and C. A. Malcolm The Origin and Evolution of Mosquito APE Retroposons Mol. Biol. Evol., November 1, 2005; 22(11): 2190 - 2197. [Abstract] [Full Text] [PDF]

				K. K. Kojima and H. Fujiwara An extraordinary retrotransposon family encoding dual endonucleases Genome Res., August 1, 2005; 15(8): 1106 - 1117. [Abstract] [Full Text] [PDF]

				K. I. Pyatkov, I. R. Arkhipova, N. V. Malkova, D. J. Finnegan, and M. B. Evgen'ev Reverse transcriptase and endonuclease activities encoded by Penelope-like retroelements PNAS, October 12, 2004; 101(41): 14719 - 14724. [Abstract] [Full Text] [PDF]

				K. K. Kojima and H. Fujiwara Cross-Genome Screening of Novel Sequence-Specific Non-LTR Retrotransposons: Various Multicopy RNA Genes and Microsatellites Are Selected as Targets Mol. Biol. Evol., February 1, 2004; 21(2): 207 - 217. [Abstract] [Full Text] [PDF]

				J. Biedler and Z. Tu Non-LTR Retrotransposons in the African Malaria Mosquito, Anopheles gambiae: Unprecedented Diversity and Evidence of Recent Activity Mol. Biol. Evol., November 1, 2003; 20(11): 1811 - 1825. [Abstract] [Full Text] [PDF]

W. D. Burke, D. Singh, and T. H. Eickbush R5 Retrotransposons Insert into a Family of Infrequently Transcribed 28S rRNA Genes of Planaria Mol. Biol. Evol.,