Transplantation of target site specificity by swapping the endonuclease domains of two LINEs

Structure of the R2 non-LTR retrotransposon initiating target-primed reverse transcription

Non-long terminal repeat (non-LTR) retrotransposons, or long interspersed nuclear elements (LINEs), are an abundant class of eukaryotic transposons that insert into genomes by target-primed reverse transcription (TPRT). During TPRT, a target DNA sequence is nicked and primes reverse transcription of the retrotransposon RNA. Here, we report the cryo-electron microscopy structure of the Bombyx mori R2 non-LTR retrotransposon initiating TPRT at its ribosomal DNA target. The target DNA sequence is unwound at the insertion site and recognized by an upstream motif. An extension of the reverse transcriptase (RT) domain recognizes the retrotransposon RNA and guides the 3' end into the RT active site to template reverse transcription. We used Cas9 to retarget R2 in vitro to non-native sequences, suggesting future use as a reprogrammable RNA-based gene-insertion tool.

Essential factors involved in the precise targeting and insertion of telomere-specific non-LTR retrotransposon, SART1Bm

Telomere length maintenance is essential for most eukaryotes to ensure genome stability and integrity. A non-long terminal repeat (LTR) retrotransposon, SART1Bm, targets telomeric repeats (TTAGG)n of the silkworm Bombyx mori and is presumably involved in telomere length maintenance. However, how many telomeric repeats are required for its retrotransposition and how reverse transcription is initiated at the target site are not well understood. Here, using an ex vivo and trans-in vivo recombinant baculovirus retrotransposition system, we demonstrated that SART1Bm requires at least three (TTAGG) telomeric repeats and a longer poly(A) tail for its accurate retrotransposition. We found that SART1Bm retrotransposed only in the third (TTAGG) tract of three repeats and that the A residue of the (TTAGG) unit was essential for its retrotransposition. Interestingly, SART1Bm also retrotransposed into telomeric repeats of other species, such as human (TTAGGG)n repeats, albeit with low retrotransposition efficiency. We further showed that the reverse transcription of SART1Bm occurred inaccurately at the internal site of the 3' untranslated region (UTR) when using a short poly(A) tail but at the accurate site when using a longer poly(A) tail. These findings promote our understanding of the general mechanisms of site-specific retrotransposition and aid the development of a site-specific gene knock-in tool.

Targeted gene knockin in zebrafish using the 28S rDNA-specific non-LTR-retrotransposon R2Ol

Background

Although most of long interspersed elements (LINEs), one class of non-LTR-retrotransposons, are integrated into the host genome randomely, some elements are retrotransposed into the specific sequences of the genomic regions, such as rRNA gene (rDNA) clusters, telomeric repeats and other repetitive sequenes. Most of the sequence-specific LINEs have been reported mainly among invertebrate species and shown to retrotranspose into the specific sequences in vivo and in vitro systems. Recenlty, 28S rDNA-specific LINE R2 elements are shown to be distributed among widespread vertebrate species, but the sequence-specific retrotransposition of R2 has never been demonstrated in vertebrates.

Results

Here we cloned a full length unit of R2 from medaka fish Oryzias latipes, named R2Ol, and engineered it to a targeted gene integration tool in zebrafish. By injecting R2Ol-encoding mRNA into zebrafish embryos, R2Ol retrotransposed precisely into the target site at high efficiency (98%) and was transmitted to the next generation at high frequency (50%). We also generated transgenic zebrafish carrying the enhanced green fluorescent protein (EGFP) reporter gene in 28S rDNA target by the R2Ol retrotransposition system.

Conclusions

Sequence-specific LINE retrotransposes into the precise sequence using target primed reverse transcription (TPRT), possibly providing an alternative and effective targeted gene knockin method in vertebrates.

Electronic supplementary material

The online version of this article (10.1186/s13100-019-0167-2) contains supplementary material, which is available to authorized users.

Evaluating different DNA binding domains to modulate L1 ORF2p-driven site-specific retrotransposition events in human cells

DNA binding domains (DBDs) have been used with great success to impart targeting capabilities to a variety of proteins creating highly useful genomic tools. We evaluated the ability of five types of DBDs and strategies (AAV Rep proteins, Cre, TAL effectors, zinc finger proteins, and Cas9/gRNA system) to target the L1 ORF2 protein to drive retrotransposition of Alu inserts to specific sequences in the human genome. First, we find that the L1 ORF2 protein tolerates the addition of protein domains both at the amino- and carboxy-terminus. Although in some instances retrotransposition efficiencies slightly diminished, all fusion proteins containing an intact ORF2 were capable of driving retrotransposition. Second, the stability of individual ORF2 fusion proteins varies and difficult to predict. Third, DBDs that require the formation of multimers for target recognition are unlikely to modify targeting of ORF2p-driven insertions. Fourth, the more components needed to assemble into a complex to drive targeted retrotransposition, the less likely the strategy will increase targeted insertions. Fifth, abundance of target sequences present in the genome will likely dictate the effectiveness and efficiency of targeted insertions. Lastly, the cleavage capabilities of Cas9 (or a Cas9 nickase variant) are unable to substitute for the L1 ORF2 endonuclease domain functions, suggestive that the endonuclease domain has alternate functions needed for retrotransposition. From these studies, we conclude that the most critical component for the modification of the human L1 ORF2 protein to drive targeted insertions is the selection of the DBD due to the varying functional requirements and impacts on protein stability.

Integration site selection by retroviruses and transposable elements in eukaryotes

Transposable elements and retroviruses are found in most genomes, can be pathogenic and are widely used as gene-delivery and functional genomics tools. Exploring whether these genetic elements target specific genomic sites for integration and how this preference is achieved is crucial to our understanding of genome evolution, somatic genome plasticity in cancer and ageing, host-parasite interactions and genome engineering applications. High-throughput profiling of integration sites by next-generation sequencing, combined with large-scale genomic data mining and cellular or biochemical approaches, has revealed that the insertions are usually non-random. The DNA sequence, chromatin and nuclear context, and cellular proteins cooperate in guiding integration in eukaryotic genomes, leading to a remarkable diversity of insertion site distribution and evolutionary strategies.

G-quadruplex structures within the 3' UTR of LINE-1 elements stimulate retrotransposition

Long interspersed nuclear elements (LINEs) are ubiquitous transposable elements in higher eukaryotes that have a significant role in shaping genomes, owing to their abundance. Here we report that guanine-rich sequences in the 3' untranslated regions (UTRs) of hominoid-specific LINE-1 elements are coupled with retrotransposon speciation and contribute to retrotransposition through the formation of G-quadruplex (G4) structures. We demonstrate that stabilization of the G4 motif of a human-specific LINE-1 element by small-molecule ligands stimulates retrotransposition.

Both the Exact Target Site Sequence and a Long Poly(A) Tail Are Required for Precise Insertion of the 18S Ribosomal DNA-Specific Non-Long Terminal Repeat Retrotransposon R7Ag

Ribosomal elements (R elements) are site-specific non-long terminal repeat (LTR) retrotransposons that target ribosomal DNA (rDNA). To elucidate how R elements specifically access their target sites, we isolated and characterized the 18S rDNA-specific R element R7Ag from Anopheles gambiae Using an in vivo and ex vivo recombinant baculovirus retrotransposition system, we found that the exact host 18S rDNA sequence at the target site is essential for the precise insertion of R7Ag. In addition, a long poly(A) tail is necessary for the accurate initiation of R7Ag reverse transcription, a novel mechanism found in non-LTR elements. We further compared the subcellular localizations of proteins in R7Ag as well as R1Bm, another R element that targets 28S rDNA. Although the open reading frame 1 proteins (ORF1ps) of both R7Ag and R1Bm localized predominantly in the cytoplasm, ORF2 proteins (ORF2ps) colocalized in the nucleus with the nucleolar marker fibrillarin. The ORF1ps and ORF2ps of both R elements colocalized largely in the nuclear periphery and to a lesser extent within the nucleus. These results suggest that R7Ag and R1Bm proteins may access nucleolar rDNA targets in an ORF2p-dependent manner.

Helitrons, the Eukaryotic Rolling-circle Transposable Elements

Helitrons, the eukaryotic rolling-circle transposable elements, are widespread but most prevalent among plant and animal genomes. Recent studies have identified three additional coding and structural variants of Helitrons called Helentrons, Proto-Helentron, and Helitron2. Helitrons and Helentrons make up a substantial fraction of many genomes where nonautonomous elements frequently outnumber the putative autonomous partner. This includes the previously ambiguously classified DINE-1-like repeats, which are highly abundant in Drosophila and many other animal genomes. The purpose of this review is to summarize what we have learned about Helitrons in the decade since their discovery. First, we describe the history of autonomous Helitrons, and their variants. Second, we explain the common coding features and difference in structure of canonical Helitrons versus the endonuclease-encoding Helentrons. Third, we review how Helitrons and Helentrons are classified and discuss why the system used for other transposable element families is not applicable. We also touch upon how genome-wide identification of candidate Helitrons is carried out and how to validate candidate Helitrons. We then shift our focus to a model of transposition and the report of an excision event. We discuss the different proposed models for the mechanism of gene capture. Finally, we will talk about where Helitrons are found, including discussions of vertical versus horizontal transfer, the propensity of Helitrons and Helentrons to capture and shuffle genes and how they impact the genome. We will end the review with a summary of open questions concerning the biology of this intriguing group of transposable elements.

Site-specific non-LTR retrotransposons

Although most of non-long terminal repeat (non-LTR) retrotransposons are incorporated in the host genome almost randomly, some non-LTR retrotransposons are incorporated into specific sequences within a target site. On the basis of structural and phylogenetic features, non-LTR retrotransposons are classified into two large groups, restriction enzyme-like endonuclease (RLE)-encoding elements and apurinic/apyrimidinic endonuclease (APE)-encoding elements. All clades of RLE-encoding non-LTR retrotransposons include site-specific elements. However, only two of more than 20 APE-encoding clades, Tx1 and R1, contain site-specific non-LTR elements. Site-specific non-LTR retrotransposons usually target within multi-copy RNA genes, such as rRNA gene (rDNA) clusters, or repetitive genomic sequences, such as telomeric repeats; this behavior may be a symbiotic strategy to reduce the damage to the host genome. Site- and sequence-specificity are variable even among closely related non-LTR elements and appeared to have changed during evolution. In the APE-encoding elements, the primary determinant of the sequence- specific integration is APE itself, which nicks one strand of the target DNA during the initiation of target primed reverse transcription (TPRT). However, other factors, such as interaction between mRNA and the target DNA, and access to the target region in the nuclei also affect the sequence-specificity. In contrast, in the RLE-encoding elements, DNA-binding motifs appear to affect their sequence-specificity, rather than the RLE domain itself. Highly specific integration properties of these site-specific non-LTR elements make them ideal alternative tools for sequence-specific gene delivery, particularly for therapeutic purposes in human diseases.

Insertion of Retrotransposons at Chromosome Ends: Adaptive Response to Chromosome Maintenance

The telomerase complex is a specialized reverse transcriptase (RT) that inserts tandem DNA arrays at the linear chromosome ends and contributes to the protection of the genetic information in eukaryotic genomes. Telomerases are phylogenetically related to retrotransposons, encoding also the RT activity required for the amplification of their sequences throughout the genome. Intriguingly the telomerase gene is lost from the Drosophila genome and tandem retrotransposons replace telomeric sequences at the chromosome extremities. This observation suggests the versatility of RT activity in counteracting the chromosome shortening associated with genome replication and that retrotransposons can provide this activity in case of a dysfunctional telomerase. In this review paper, we describe the major classes of retroelements present in eukaryotic genomes in order to point out the differences and similarities with the telomerase complex. In a second part, we discuss the insertion of retroelements at the ends of chromosomes as an adaptive response for dysfunctional telomeres.

A New Class of SINEs with snRNA Gene-Derived Heads

Eukaryotic genomes are colonized by various transposons including short interspersed elements (SINEs). The 5' region (head) of the majority of SINEs is derived from one of the three types of RNA genes--7SL RNA, transfer RNA (tRNA), or 5S ribosomal RNA (rRNA)--and the internal promoter inside the head promotes the transcription of the entire SINEs. Here I report a new group of SINEs whose heads originate from either the U1 or U2 small nuclear RNA gene. These SINEs, named SINEU, are distributed among crocodilians and classified into three families. The structures of the SINEU-1 subfamilies indicate the recurrent addition of a U1- or U2-derived sequence onto the 5' end of SINEU-1 elements. SINEU-1 and SINEU-3 are ancient and shared among alligators, crocodiles, and gharials, while SINEU-2 is absent in the alligator genome. SINEU-2 is the only SINE family that was active after the split of crocodiles and gharials. All SINEU families, especially SINEU-3, are preferentially inserted into a family of Mariner DNA transposon, Mariner-N4_AMi. A group of Tx1 non-long terminal repeat retrotransposons designated Tx1-Mar also show target preference for Mariner-N4_AMi, indicating that SINEU was mobilized by Tx1-Mar.

Integrated mechanism for the generation of the 5' junctions of LINE inserts

To elucidate the molecular mechanism of the integration of long interspersed elements (LINEs), we characterized the 5′ ends of more than 200 LINE de novo retrotransposition events into chicken DT40 or human HeLa cells. Human L1 inserts produced 15-bp target-site duplications (TSDs) and zebrafish ZfL2-1 inserts produced 5-bp TSDs in DT40 cells, suggesting that TSD length depends on the LINE species. Further analysis of 5′ junctions revealed that the 5′-end-joining pathways of LINEs can be divided into two fundamental types—annealing or direct. We also found that the generation of 5′ inversions depends on host and LINE species. These results led us to propose a new model for 5′-end joining, the type of which is determined by the extent of exposure of 3′ overhangs generated after the second-strand cleavage and by the involvement of host factors.

DINE-1, the highest copy number repeats in Drosophila melanogaster are non-autonomous endonuclease-encoding rolling-circle transposable elements (Helentrons)

Background

The Drosophila INterspersed Elements-1 (DINE-1/INE1) transposable elements (TEs) are the most abundant component of the Drosophila melanogaster genome and have been associated with functional gene duplications. DINE-1 TEs do not encode any proteins (non-autonomous) thus are moved by autonomous partners. The identity of the autonomous partners has been a mystery. They have been allied to Helitrons (rolling-circle transposons), MITEs (DNA transposons), and non-LTR retrotransposons by different authors.

Results

We report multiple lines of bioinformatic evidence that illustrate the relationship of DINE-1 like TEs to endonuclease-encoding rolling-circle TEs (Helentrons). The structural features of Helentrons are described, which resemble the organization of the non-autonomous partners, but differ significantly from canonical Helitrons. In addition to the presence of an endonuclease domain fused to the Rep/Helicase protein, Helentrons have distinct structural features. Evidence is presented that illustrates that Helentrons are widely distributed in invertebrate, fish, and fungal genomes. We describe an intermediate family from the Phytophthora infestans genome that phylogenetically groups with Helentrons but that displays Helitron structure. In addition, evidence is presented that Helentrons can capture gene fragments in a pattern reminiscent of canonical Helitrons.

Conclusions

We illustrate the relationship of DINE-1 and related TE families to autonomous partners, the Helentrons. These findings will allow their proper classification and enable a more accurate understanding of the contribution of rolling-circle transposition to the birth of new genes, gene networks, and genome composition.

L1 retrotransposition: The snap-velcro model and its consequences

LINE-1 (L1) elements are the only active and autonomous transposable elements in humans. The core retrotransposition machinery is a ribonucleoprotein particle (RNP) containing the L1 mRNA, with endonuclease and reverse transcriptase activities. It initiates reverse transcription directly at genomic target sites upon endonuclease cleavage. Recently, using a direct L1 extension assay (DLEA), we systematically tested the ability of native L1 RNPs to extend DNA substrates of various sequences and structures. We deduced from these experiments the general rules guiding the initiation of L1 reverse transcription, referred to as the snap-velcro model. In this model, L1 target choice is not only mediated by the sequence specificity of the endonuclease, but also through base-pairing between the L1 mRNA and the target site, which permits the subsequent L1 reverse transcription step. In addition, L1 reverse transcriptase efficiently primes L1 DNA synthesis only when the 3′ end of the DNA substrate is single-stranded, suggesting so-far unrecognized DNA processing steps at the integration site.

RNA-Mediated Gene Duplication and Retroposons: Retrogenes, LINEs, SINEs, and Sequence Specificity

A substantial number of “retrogenes” that are derived from the mRNA of various intron-containing genes have been reported. A class of mammalian retroposons, long interspersed element-1 (LINE1, L1), has been shown to be involved in the reverse transcription of retrogenes (or processed pseudogenes) and non-autonomous short interspersed elements (SINEs). The 3′-end sequences of various SINEs originated from a corresponding LINE. As the 3′-untranslated regions of several LINEs are essential for retroposition, these LINEs presumably require “stringent” recognition of the 3′-end sequence of the RNA template. However, the 3′-ends of mammalian L1s do not exhibit any similarity to SINEs, except for the presence of 3′-poly(A) repeats. Since the 3′-poly(A) repeats of L1 and Alu SINE are critical for their retroposition, L1 probably recognizes the poly(A) repeats, thereby mobilizing not only Alu SINE but also cytosolic mRNA. Many flowering plants only harbor L1-clade LINEs and a significant number of SINEs with poly(A) repeats, but no homology to the LINEs. Moreover, processed pseudogenes have also been found in flowering plants. I propose that the ancestral L1-clade LINE in the common ancestor of green plants may have recognized a specific RNA template, with stringent recognition then becoming relaxed during the course of plant evolution.

Parallel relaxation of stringent RNA recognition in plant and mammalian L1 retrotransposons

L1 elements are mammalian non-long terminal repeat retrotransposons, or long interspersed elements (LINEs), that significantly influence the dynamics and fluidity of the genome. A series of observations suggest that plant L1-clade LINEs, just as mammalian L1s, mobilize both short interspersed elements (SINEs) and certain messenger RNA by recognizing the 3'-poly(A) tail of RNA. However, one L1 lineage in monocots was shown to possess a conserved 3'-end sequence with a solid RNA structure also observed in maize and sorghum SINEs. This strongly suggests that plant LINEs require a particular 3'-end sequence during initiation of reverse transcription. As one L1-clade LINE was also found to share the 3'-end sequence with a SINE in a green algal genome, I propose that the ancestral L1-clade LINE in the common ancestor of green plants may have recognized the specific RNA template, with stringent recognition then becoming relaxed during the course of plant evolution.

Low dependency of retrotransposition on the ORF1 protein of the zebrafish LINE, ZfL2-1

The zebrafish long interspersed element (LINE), ZfL2-1, which belongs to the L2 clade, contains two open reading frames, ORF1 and ORF2. ORF1 encodes a protein containing a coiled-coil motif and an esterase domain, whereas ORF2 encodes a protein containing an endonuclease and a reverse transcriptase domain. To elucidate the functional significance of ORF1 in retrotransposition, we constructed many variants of ZfL2-1 and examined their retrotransposition ability. We concluded: 1) the ORF1 protein is not essential for ZfL2-1 retrotransposition in cultured cells; 2) the translation of ORF1 is required for the translation of ORF2; and 3) ORF2 translation probably occurs via suppression of the ORF1 stop codon, the efficiency of which is influenced by the context of the sequence juxtaposed to the 3' side of the stop codon. These results offer a new perspective on the evolution of the L2 clade LINEs.

Self-interaction, nucleic acid binding, and nucleic acid chaperone activities are unexpectedly retained in the unique ORF1p of zebrafish LINE

Long interspersed elements (LINEs) are mobile elements that comprise a large proportion of many eukaryotic genomes. Although some LINE-encoded open reading frame 1 proteins (ORF1ps) were suggested to be required for LINE mobilization through binding to their RNA, their general role is not known. The ZfL2-1 ORF1p, which belongs to the esterase-type ORF1p, is especially interesting because it has no known RNA-binding domain. Here we demonstrate that ZfL2-1 ORF1p has all the canonical activities associated with known ORF1ps, including self-interaction, nucleic acid binding, and nucleic acid chaperone activities. In particular, we showed that its chaperone activity is reversible, suggesting that the chaperone activities of many other ORF1ps are also reversible. From this discovery, we propose that LINE ORF1ps play a general role in LINE integration by forming a complex with LINE RNA and rearranging its conformation.

Coevolution of telomeric repeats and telomeric repeat-specific non-LTR retrotransposons in insects

In the telomeres of the silkworm Bombyx mori, telomeric repeat-specific non-long terminal repeat (LTR) retrotransposon SARTBm1 is accumulated in the TTAGG telomeric repeats. Here, we identify novel telomeric repeat-specific non-LTR retrotransposons, SARTTc family, from the red flour beetle Tribolium castaneum in the unconventional TCAGG telomeric repeats. To compare the sequence specificity of SARTBm1 and SARTTc1, we developed a comparable ex vivo retrotransposition assay. Both SARTBm1 and SARTTc1 preferred the telomeric sequence of their hosts, suggesting that the target specificity of these retrotransposons coevolved with their host's telomeric repeats. Swapping experiment indicated that the endonuclease domain is involved in recognizing the target sequence. Moreover, SARTBm1 proteins could retrotranspose 3'untranslated region (UTR) sequence of SARTTc1 as well as their own 3'UTR, whereas SARTTc1 proteins could only retrotranspose their own 3'UTRs. These results provide insights to the mechanism and divergence of sequence specificity and 3'UTR recognition in non-LTR retrotransposons.

Evolutionary conservation of the functional modularity of primate and murine LINE-1 elements

LINE-1 (L1) retroelements emerged in mammalian genomes over 80 million years ago with a few dominant subfamilies amplifying over discrete time periods that led to distinct human and mouse L1 lineages. We evaluated the functional conservation of L1 sequences by comparing retrotransposition rates of chimeric human-rodent L1 constructs to their parental L1 counterparts. Although amino acid conservation varies from ∼35% to 63% for the L1 ORF1p and ORF2p, most human and mouse L1 sequences can be functionally exchanged. Replacing either ORF1 or ORF2 to create chimeric human-mouse L1 elements did not adversely affect retrotransposition. The mouse ORF2p retains retrotransposition-competency to support both Alu and L1 mobilization when any of the domain sequences we evaluated were substituted with human counterparts. However, the substitution of portions of the mouse cys-domain into the human ORF2p reduces both L1 retrotransposition and Alu trans-mobilization by 200–1000 fold. The observed loss of ORF2p function is independent of the endonuclease or reverse transcriptase activities of ORF2p and RNA interaction required for reverse transcription. In addition, the loss of function is physically separate from the cysteine-rich motif sequence previously shown to be required for RNP formation. Our data suggest an additional role of the less characterized carboxy-terminus of the L1 ORF2 protein by demonstrating that this domain, in addition to mediating RNP interaction(s), provides an independent and required function for the retroelement amplification process. Our experiments show a functional modularity of most of the LINE sequences. However, divergent evolution of interactions within L1 has led to non-reciprocal incompatibilities between human and mouse ORF2 cys-domain sequences.

Independently derived targeting of 28S rDNA by A- and D-clade R2 retrotransposons: Plasticity of integration mechanism

Restriction-like endonuclease (RLE) bearing non-LTR retrotransposons are site-specific elements that integrate into the genome through a target primed reverse transcription mechanism (TPRT). R2 elements have been used as a model system for investigating non-LTR retrotransposon integration. We previously demonstrated that R2 retrotransposons require two subunits of the element-encoded multifunctional protein to integrate-one subunit bound upstream of the insertion site and one bound downstream. R2 elements have been phylogenetically categorized into four clades: R2-A, B, C and D, that diverged from a common ancestor more than 850 million years ago. All R2 elements target the same sequence within 28S rDNA. The amino-terminal domain of R2Bm, an R2-D clade element, contains a single zinc finger and a Myb motif that are responsible for binding R2 protein downstream of the insertion site. Target site recognition is of interest as it is the first step in the integration reaction and may help elucidate evolutionary history and integration mechanism. The amino-terminal domain of R2-A clade members contains three zinc fingers and a Myb motif. We show here that R2Lp, an R2-A clade member, uses its amino-terminal DNA binding motifs to bind upstream of the insertion site. Because the R2-A and R2-D clade elements recognize 28S rDNA differently, we conclude the A- and D-clades represent independent targeting events to the 28S site. Our results also indicate a certain plasticity of insertional mechanics exists between the two clades.

Creation of a novel telomere-cutting endonuclease based on the EN domain of telomere-specific non-long terminal repeat retrotransposon, TRAS1

Background

The ends of chromosomes, termed telomeres consist of repetitive DNA. The telomeric sequences shorten with cell division and, when telomeres are critically abbreviated, cells stop proliferating. However, in cancer cells, by the expression of telomerase which elongates telomeres, the cells can continue proliferating. Many approaches for telomere shortening have been pursued in the past, but to our knowledge, cutting telomeres in vivo has not so far been demonstrated. In addition, there is lack of information on the cellular effects of telomere shortening in human cells.

Results

Here, we created novel chimeric endonucleases to cut telomeres by fusing the endonuclease domain (TRAS1EN) of the silkworm's telomere specific non-long terminal repeat retrotransposon TRAS1 to the human telomere-binding protein, TRF1. An in vitro assay demonstrated that the TRAS1EN-TRF1 chimeric endonucleases (T-EN and EN-T) cut the human (TTAGGG)_n repeats specifically. The concentration of TRAS1EN-TRF1 chimeric endonucleases necessary for the cleavage of (TTAGGG)_n repeats was about 40-fold lower than that of TRAS1EN alone. When TRAS1EN-TRF1 endonucleases were introduced into human U2OS cancer cells using adenovirus vectors, the enzymes localized at telomeres of nuclei, cleaved and shortened the telomeric DNA by double-strand breaks. When human U2OS and HFL-1 fibroblast cells were infected with EN-T recombinant adenovirus, their cellular proliferation was suppressed for about 2 weeks after infection. In contrast, the TRAS1EN mutant (H258A) chimeric endonuclease fused with TRF1 (ENmut-T) did not show the suppression effect. The EN-T recombinant adenovirus induced telomere shortening in U2OS cells, activated the p53-dependent pathway and caused the senescence associated cellular responses, while the ENmut-T construct did not show such effects.

Conclusions

A novel TRAS1EN-TRF1 chimeric endonuclease (EN-T) cuts the human telomeric repeats (TTAGGG)_n specifically in vitro and in vivo. Thus, the chimeric endonuclease which is expressed from an adenoviral vector can suppress cell proliferation of cancer cells.

Unique functions of repetitive transcriptomes

Repetitive sequences occupy a huge fraction of essentially every eukaryotic genome. Repetitive sequences cover more than 50% of mammalian genomic DNAs, whereas gene exons and protein-coding sequences occupy only ~3% and 1%, respectively. Numerous genomic repeats include genes themselves. They generally encode "selfish" proteins necessary for the proliferation of transposable elements (TEs) in the host genome. The major part of evolutionary "older" TEs accumulated mutations over time and fails to encode functional proteins. However, repeats have important functions also on the RNA level. Repetitive transcripts may serve as multifunctional RNAs by participating in the antisense regulation of gene activity and by competing with the host-encoded transcripts for cellular factors. In addition, genomic repeats include regulatory sequences like promoters, enhancers, splice sites, polyadenylation signals, and insulators, which actively reshape cellular transcriptomes. TE expression is tightly controlled by the host cells, and some mechanisms of this regulation were recently decoded. Finally, capacity of TEs to proliferate in the host genome led to the development of multiple biotechnological applications.

The telomere-specific non-LTR retrotransposons SART1 and TRAS1 are suppressed by Piwi subfamily proteins in the silkworm, Bombyx mori

The telomere structures in Bombyx mori are thought to be maintained mainly by the transposition of the specialized telomeric retroelements SART and TRAS. The silkworm genome has telomeric TTAGG repeats and telomerase, but this telomerase displays little or no activity. Here, we report that the transcription of the telomeric retroelements SART1 and TRAS1 is suppressed by the silkworm Piwi subfamily proteins BmAgo3 and Siwi. The silkworm Piwi subfamily was found to be expressed predominantly in the gonads and early embryo, as in other model organisms, but in BmN4 cultured cells, these proteins formed granules that were separate from the nuage, which is a different behaviour pattern. The expression of TRAS1 was increased in BmN4 cells when BmAgo3 or Siwi were silenced by RNAi. Our results suggest that B. mori Piwi proteins are involved in regulating the transposition of telomeric retroelements, and that the functional piRNA pathway is conserved in BmN4 cultured cells.

The diversity of retrotransposons and the properties of their reverse transcriptases

A number of abundant mobile genetic elements called retrotransposons reverse transcribe RNA to generate DNA for insertion into eukaryotic genomes. Four major classes of retrotransposons are described here. First, the long-terminal-repeat (LTR) retrotransposons have similar structures and mechanisms to those of the vertebrate retroviruses. Genes that may enable these retrotransposons to leave a cell have been acquired by these elements in a number of animal and plant lineages. Second, the tyrosine recombinase retrotransposons are similar to the LTR retrotransposons except that they have substituted a recombinase for the integrase and recombine into the host chromosomes. Third, the non-LTR retrotransposons use a cleaved chromosomal target site generated by an encoded endonuclease to prime reverse transcription. Finally, the Penelope-like retrotransposons are not well understood but appear to also use cleaved DNA or the ends of chromosomes as primer for reverse transcription. Described in the second part of this review are the enzymatic properties of the reverse transcriptases (RTs) encoded by retrotransposons. The RTs of the LTR retrotransposons are highly divergent in sequence but have similar enzymatic activities to those of retroviruses. The RTs of the non-LTR retrotransposons have several unique properties reflecting their adaptation to a different mechanism of retrotransposition.

Determinants for DNA target structure selectivity of the human LINE-1 retrotransposon endonuclease

The human LINE-1 endonuclease (L1-EN) is the targeting endonuclease encoded by the human LINE-1 (L1) retrotransposon. L1-EN guides the genomic integration of new L1 and Alu elements that presently account for ∼28% of the human genome. L1-EN bears considerable technological interest, because its target selectivity may ultimately be engineered to allow the site-specific integration of DNA into defined genomic locations. Based on the crystal structure, we generated L1-EN mutants to analyze and manipulate DNA target site recognition. Crystal structures and their dynamic and functional analysis show entire loop grafts to be feasible, resulting in altered specificity, while individual point mutations do not change the nicking pattern of L1-EN. Structural parameters of the DNA target seem more important for recognition than the nucleotide sequence, and nicking profiles on DNA oligonucleotides in vitro are less well defined than the respective integration site consensus in vivo. This suggests that additional factors other than the DNA nicking specificity of L1-EN contribute to the targeted integration of non-LTR retrotransposons.

Characterization of the sequence specificity of the R1Bm endonuclease domain by structural and biochemical studies

R1Bm is a long interspersed element (LINE) inserted into a specific sequence within 28S rDNA of the silkworm genome. Of two open reading frames (ORFs) of R1Bm, ORF2 encodes a reverse transcriptase (RT) and an endonuclease (EN) domain which digests specifically both top and bottom strand of the target sequence in 28S rDNA. To elucidate the sequence specificity of EN domain of R1Bm (R1Bm EN), we examined the cleavage tendency for the target sequences, and found that 5'-A(G/C)(A/T)!(A/G)T-3' is the consensus sequence (! = cleavage site). We also determined the crystal structure of R1Bm EN at 2.0 A resolution. Its structure was basically similar to AP endonuclease family, but had a special beta-hairpin at the edge of the DNA binding surface, which is a common feature among EN of LINEs. Point-mutations on the DNA binding surface of R1Bm EN significantly decreased the cleavage activities, but did not affect the sequence recognition in most residues. However, two mutants Y98A and N180A had altered cleavage patterns, suggesting an important role of these residues (Y98 and N180) for the sequence recognition of R1Bm EN. In addition, Y98A mutant showed another cleavage pattern, that implies de novo design of novel sequence-specific EN.

A novel target-specific gene delivery system combining baculovirus and sequence-specific long interspersed nuclear elements

Transposable elements are valuable for somatic and germ-line transformation. However, long interspersed nuclear elements (LINEs) have not been used because of poor information on the transposition mechanism. We have developed a novel gene delivery system combining baculovirus AcNPV and two silkworm LINEs, SART1 and R1, which integrate into specific sequences of telomeric repeats and 28S ribosomal DNA, respectively. When two LINEs containing the enhanced green fluorescent protein gene recombined into AcNPV were infected into fifth instar larvae of the silkworm, we observed target-specific retrotransposition of LINEs at 72h post-infection, using polymerase chain reaction amplification and sequencing. Telomere- and 28S rDNA-specific transposition occurred in all nine tissues tested, including the ovary and testis. This is the first demonstration of site-specific gene delivery in living larvae. Insertion efficiencies were dependent on the virus titer for injection and the host strains of Bombyx mori. Using this system, we successfully detected the intergeneration transmission of retrotransposed sequences. In addition, AcNPV-mediated SART1 also transposed into telomere of another lepidopteran, Orgyia recens, suggesting that this system is useful for a wide variety of AcNPV-infectious insects. Site-specific gene delivery by virus-mediated LINE will be a potential gene therapy tool to avoid harmful unexpected insertions.

In vivo gene transfer into the honeybee using a nucleopolyhedrovirus vector

The honeybee Apis mellifera L. is a social insect and one of the most industrially important insects. We examined whether a baculovirus-mediated retrotransposon is applicable to in vivo transfer of exogenous genes to the honeybees. Honeybee larvae and pupae were injected with two types of recombinant Autographa californica nucleopolyhedrovirus (AcNPV) vectors, one that includes the enhanced green fluorescent protein gene (egfp) as a reporter to be inserted into the honeybee genome, and another that includes the reverse transcriptase gene responsible for the insertion. Fluorescence was observed in most of the viral-injected larvae and pupae. Reverse transcription-polymerase chain reaction and immunoblotting confirmed egfp mRNA and eGFP expression in these honeybees, although egfp insertion into the honeybee genome was not confirmed. These results indicate that AcNPV vectors can be used for the transfer and transient expression of an exogenous gene in the larval and pupal honeybees.

RNA from the 5' end of the R2 retrotransposon controls R2 protein binding to and cleavage of its DNA target site

Non-LTR retrotransposons insert into eukaryotic genomes by target-primed reverse transcription (TPRT), a process in which cleaved DNA targets are used to prime reverse transcription of the element's RNA transcript. Many of the steps in the integration pathway of these elements can be characterized in vitro for the R2 element because of the rigid sequence specificity of R2 for both its DNA target and its RNA template. R2 retrotransposition involves identical subunits of the R2 protein bound to different DNA sequences upstream and downstream of the insertion site. The key determinant regulating which DNA-binding conformation the protein adopts was found to be a 320-nt RNA sequence from near the 5' end of the R2 element. In the absence of this 5' RNA the R2 protein binds DNA sequences upstream of the insertion site, cleaves the first DNA strand, and conducts TPRT when RNA containing the 3' untranslated region of the R2 transcript is present. In the presence of the 320-nt 5' RNA, the R2 protein binds DNA sequences downstream of the insertion site. Cleavage of the second DNA strand by the downstream subunit does not appear to occur until after the 5' RNA is removed from this subunit. We postulate that the removal of the 5' RNA normally occurs during reverse transcription, and thus provides a critical temporal link to first- and second-strand DNA cleavage in the R2 retrotransposition reaction.

Essential domains for ribonucleoprotein complex formation required for retrotransposition of telomere-specific non-long terminal repeat retrotransposon SART1

Non-long terminal repeat (LTR) retrotransposons are major components of the higher eukaryotic genome. Most of them have two open reading frames (ORFs): ORF2 encodes mainly the endonuclease and reverse transcriptase domains, but the functional features of ORF1 remain largely unknown. We used telomere-specific non-LTR retrotransposon SART1 in Bombyx mori and clarified essential roles of the ORF1 protein (ORF1p) in ribonucleoprotein (RNP) formation by novel approaches: in vitro reconstitution and in vivo/in vitro retrotransposition assays using the baculovirus expression system. Detailed mutation analyses showed that each of the three CCHC motifs at the ORF1 C terminus are essential for SART1 retrotransposition and are involved in packaging the SART1 mRNA specifically into RNP. We also demonstrated that amino acid residues 555 to 567 and 285 to 567 in the SART1 ORF1p are crucial for the ORF1p-ORF1p and ORF1p-ORF2p interactions, respectively. The loss of these domains abolishes protein-protein interaction, leading to SART1 retrotransposition deficiency. These data suggest that systematic formation of RNP composed of ORF1p, ORF2p, and mRNA is mainly mediated by ORF1p domains and is a common, essential step for many non-LTR retrotransposons encoding the two ORFs.

Identification and characterization of the telomerase reverse transcriptase of Bombyx mori (silkworm) and Tribolium castaneum (flour beetle)

Chromosomal ends of most eukaryotes are composed of simple telomeric repeats. Arthropod telomeres are generally constituted by TTAGG pentanucleotide repeats; however, some insect species including Drosophila melanogaster do not have telomeric repeats. In contrast, the domestic silkworm Bombyx mori contains TTAGG-type telomeric repeats, but the telomerase activity has not been detected in all investigated tissues. To search for a cause of unusual telomere structure in insects, we here identified telomerase reverse transcriptase (TERT) subunit from the domestic silkworm B. mori and the flour beetle Tribolium castaneum. This is the first report of telomerase genes from arthropods. The domestic silkworm TERT gene (BmoTERT) and the flour beetle TERT gene (TcasTERT) both did not have the N-terminal GQ motif. Comparison between cDNA and genomic DNA of BmoTERT revealed that it includes no introns. BmoTERT contains five ATG codons in its 5'UTR, which could reduce the translation of BmoTERT proteins. Also, Northern hybridization indicated that BmoTERT is transcribed at a very low level. These unique features of BmoTERT possibly explain the undetectable Bombyx telomerase activity.

BmStart1, a novel carotenoid-binding protein isoform from Bombyx mori, is orthologous to MLN64, a mammalian cholesterol transporter

Carotenoid-binding protein (CBP) from the silkworm Bombyx mori is an essential molecule for carotenoid dependent cocoon pigmentation. We identified a novel isoform of CBP, Start1 of B. mori (BmStart1). BmStart1 contains a membrane-spanning MENTAL domain in its N-terminus and a lipid-binding START domain in its C-terminus. This domain architecture is identical to the mammalian MLN64 and Start1 of Drosophila melanogaster (DmStart1), both of which have been implicated to function in cholesterol transport and regulation of steroidogenesis. BmStart1 is expressed in both white and yellow cocoon strains of B. mori, while CBP is only detected in the yellow cocoon strain. BmStart1 mRNA abundance in the prothoracic gland, the main ecdysteroidogenic tissue, positively correlates with changes in the hemolymph ecdysteroid level. Genomic sequence analysis revealed that BmStart1 and CBP are generated from the same gene locus by alternative splicing. Splice site comparison and homology search indicate that BmStart1 is orthologous to both MLN64 and DmStart1. This study implies that alternative splicing of the BmStart1/CBP gene generates unique protein isoforms whose endogenous ligands, sterol or carotenoid, are structurally different.

Telomere-specific non-LTR retrotransposons and telomere maintenance in the silkworm, Bombyx mori

Most insects have telomeres that consist of pentanucleotide (TTAGG) telomeric repeats, which are synthesized by telomerase. However, all species in Diptera so far examined and several species in other orders of insect have lost the (TTAGG)n repeats, suggesting that some of them recruit telomerase-independent telomere maintenance. The silkworm, Bombyx mori, retains the TTAGG motifs in the chromosomal ends but expresses quite a low level of telomerase activity in all stages of various tissues. Just proximal to a 6-8-kb stretch of the TTAGG repeats in B. mori, more than 1000 copies of non-LTR retrotransposons, designated TRAS and SART families, occur among the telomeric repeats and accumulate. TRAS and SART are abundantly transcribed and actively retrotransposed into TTAGG telomeric repeats in a highly sequence-specific manner. They have three possible mechanisms to ensure specific integration into the telomeric repeats. This article focuses on the telomere structure and telomere-specific non-LTR retrotransposons in B. mori and discusses the mechanisms for telomere maintenance in this insect.

An extraordinary retrotransposon family encoding dual endonucleases

Retrotransposons commonly encode a reverse transcriptase (RT), but other functional domains are variable. The acquisition of new domains is the dominant evolutionary force that brings structural variety to retrotransposons. Non-long-terminal-repeat (non-LTR) retrotransposons are classified into two groups by their structure. Early branched non-LTR retrotransposons encode a restriction-like endonuclease (RLE), and recently branched non-LTR retrotransposons encode an apurinic/apyrimidinic endonuclease-like endonuclease (APE). In this study, we report a novel non-LTR retrotransposon family Dualen, identified from the Chlamydomonas reinhardtii genome. Dualen encodes two endonucleases, RLE and APE, with RT, ribonuclease H, and cysteine protease. Phylogenetic analyses of the RT domains revealed that Dualen is positioned at the midpoint between the early-branched and the recently branched groups. In the APE tree, Dualen was branched earlier than the I group and the Jockey group. The ribonuclease H domains among the Dualen family and other non-LTR retrotransposons are monophyletic. Phylogenies of three domains revealed the monophyly of the Dualen family members. The domain structure and the phylogeny of each domain imply that Dualen is a retrotransposon conserving the domain structure just after the acquisition of APE. From these observations, we discuss the evolution of domain structure of non-LTR retrotransposons.

APE-type non-LTR retrotransposons: determinants involved in target site recognition

Non-long terminal repeat (Non-LTR) retrotransposons represent a diverse and widely distributed group of transposable elements and an almost ubiquitous component of eukaryotic genomes that has a major impact on evolution. Their copy number can range from a few to several million and they often make up a significant fraction of the genomes. The members of the dominating subtype of non-LTR retrotransposons code for an endonuclease with homology to apurinic/apyrimidinic endonucleases (APE), and are thus termed APE-type non-LTR retrotransposons. In the last decade both the number of identified non-LTR retrotransposons and our knowledge of biology and evolution of APE-type non-LTR retrotransposons has increased tremendously.

Eukaryotic translational coupling in UAAUG stop-start codons for the bicistronic RNA translation of the non-long terminal repeat retrotransposon SART1

Most eukaryotic cellular mRNAs are monocistronic; however, many retroviruses and long terminal repeat (LTR) retrotransposons encode multiple proteins on a single RNA transcript using ribosomal frameshifting. Non-long terminal repeat (non-LTR) retrotransposons are considered the ancestor of LTR retrotransposons and retroviruses, but their translational mechanism of bicistronic RNA remains unknown. We used a baculovirus expression system to produce a large amount of the bicistronic RNA of SART1, a non-LTR retrotransposon of the silkworm, and were able to detect the second open reading frame protein (ORF2) by Western blotting. The ORF2 protein was translated as an independent protein, not as an ORF1-ORF2 fusion protein. We revealed by mutagenesis that the UAAUG overlapping stop-start codon and the downstream RNA secondary structure are necessary for efficient ORF2 translation. Increasing the distance between the ORF1 stop codon and the ORF2 start codon decreased translation efficiency. These results are different from the eukaryotic translation reinitiation mechanism represented by the yeast GCN4 gene, in which the probability of reinitiation increases as the distance between the two ORFs increases. The translational mechanism of SART1 ORF2 is analogous to translational coupling observed in prokaryotes and viruses. Our results indicate that translational coupling is a general mechanism for bicistronic RNA translation.

R2 target-primed reverse transcription: ordered cleavage and polymerization steps by protein subunits asymmetrically bound to the target DNA

R2 elements are non-long terminal repeat retrotransposons that specifically insert into 28S rRNA genes of many animal groups. These elements encode a single protein with reverse transcriptase and endonuclease activities as well as specific DNA and RNA binding properties. In this report, gel shift experiments were conducted to investigate the stoichiometry of the DNA, RNA, and protein components of the integration reaction. The enzymatic functions associated with each of the protein complexes were also determined, and DNase I digests were used to footprint the protein onto the target DNA. Additionally, a short polypeptide containing the N-terminal putative DNA-binding motifs was footprinted on the DNA target site. These combined findings revealed that one protein subunit binds the R2 RNA template and the DNA 10 to 40 bp upstream of the insertion site. This subunit cleaves the first DNA strand and uses that cleavage to prime reverse transcription of the R2 RNA transcript. Another protein subunit(s) uses the N-terminal DNA binding motifs to bind to the 18 bp of target DNA downstream of the insertion site and is responsible for cleavage of the second DNA strand. A complete model for the R2 integration reaction is presented, which with minor modifications is adaptable to other non-LTR retrotransposons.

Retrotransposable elements on the W chromosome of the silkworm, Bombyx mori

The sex chromosomes of the silkworm, Bombyxmori, are designated ZW(XY) for females and ZZ(XX) for males. The W chromosome of B. mori does not recombine with the Z chromosome and autosomes and no genes for morphological characters have been mapped to the W chromosome as yet. Furthermore, femaleness is determined by the presence of a single W chromosome, regardless of the number of autosomes or Z chromosomes. To understand these interesting features of the W chromosome, it is necessary to analyze the W chromosome at the molecular biology level. Initially to isolate DNA sequences specific for the W chromosome as randomly amplified polymorphic DNA (RAPD) markers, we compared the genomic DNAs between males and females by PCR with arbitrary 10-mer primers. To the present, we have identified 12 W-specific RAPD markers, and with the exception of one RAPD marker, all of the deduced amino acid sequences of these W-specific RAPD markers show similarity to previously reported amino acid sequences of retrotransposable elements from various organisms. After constructing a genomic DNA lambda phage library of B. mori we obtained two lambda phage clones, one containing the W-Kabuki RAPD sequence and one containing the W-Samurai RAPD sequence and found that these DNA sequences comprised nested structures of many retrotransposable elements. To further analyze the W chromosome, we obtained 14 W-specific bacterial artificial chromosome (BAC) clones from three BAC libraries and subjected these clones to shotgun sequencing. The resulting assembly of sequences did not produce a single contiguous sequence due to the presence of many retrotransposable elements. Therefore, we coupled PCR with shotgun sequencing. Through these analyses, we found that many long terminal repeat (LTR) and non-LTR retrotransposons, retroposons, DNA transposons and their derivatives, have accumulated on the W chromosome as strata. These results strongly indicate that retrotransposable elements are the main structural component of the W chromosome.

Functional roles of 3'-terminal structures of template RNA during in vivo retrotransposition of non-LTR retrotransposon, R1Bm

R1Bm is a non-LTR retrotransposon found specifically within 28S rRNA genes of the silkworm. Different from other non-LTR retrotransposons encoding two open reading frames (ORFs), R1Bm structurally lacks a poly (A) tract at its 3' end. To study how R1Bm initiates reverse transcription from the poly (A)-less template RNA, we established an in vivo retrotransposition system using recombinant baculovirus, and characterized retrotransposition activities of R1Bm. Target-primed reverse transcription (TPRT) of R1Bm occurred from the cleavage site generated by endonuclease (EN). The 147 bp of 3'-untranslated region (3'UTR) was essential for efficient retrotransposition of R1Bm. Even using the complete R1Bm element, however, reverse transcription started from various sites of the template RNA mostly with 5'-UG-3' or 5'-UGU-3' at their 3' ends, which are presumably base-paired with 3' end of the EN-digested 28S rDNA target sequence, 5'-AGTAGATAGGGACA-3'. When the downstream sequence of 28S rDNA target was added to the 3' end of R1 unit, reverse transcription started exactly from the 3' end of 3'UTR and retrotransposition efficiency increased. These results indicate that 3'-terminal structure of template RNA including read-through region interacts with its target rDNA sequences of R1Bm, which plays important roles in initial process of TPRT in vivo.

An Entamoeba histolytica LINE/SINE pair inserts at common target sites cleaved by the restriction enzyme-like LINE-encoded endonuclease

The non-long-terminal-repeat (non-LTR) retrotransposons (also called long interspersed repetitive elements [LINEs]) are among the oldest retroelements. Here we describe the properties of such an element from a primitive protozoan parasite, Entamoeba histolytica, that infects the human gut. This 4.8-kb element, called EhLINE1, is present in about 140 copies dispersed throughout the genome. The element belongs to the R4 clade of non-LTR elements. It has a centrally located reverse transcriptase domain and a restriction enzyme-like endonuclease (EN) domain at the carboxy terminus. We have cloned and expressed a 794-bp fragment containing the EN domain in Escherichia coli. The purified protein could nick supercoiled pBluescript DNA to yield open circular and linear DNAs. The conserved PDX(12-14)D motif was required for activity. Genomic sequences flanking the sites of insertion of EhLINE1 and the putative partner short interspersed repetitive element (SINE), EhSINE1, were analyzed. Both elements resulted in short target site duplications (TSD) upon insertion. A common feature was the presence of a short T-rich stretch just upstream of the TSD in most insertion sites. By sequence analysis an empty target site in the E. histolytica genome, known to be occupied by EhSINE1, was identified. When a 176-bp fragment containing the empty site was used as a substrate for EN, it was prominently nicked on the bottom strand at the precise point of insertion of EhSINE1, showing that this SINE could use the LINE-encoded endonuclease for its insertion. The nick on the bottom strand was toward the right of the TSD, which is uncommon. The lack of strict target site-specificity of the restriction enzyme-like EN encoded by EhLINE1 is also exceptional. A model for retrotransposition of EhLINE1/SINE1 is presented.

Essential motifs in the 3' untranslated region required for retrotransposition and the precise start of reverse transcription in non-long-terminal-repeat retrotransposon SART1

Non-long-terminal-repeat (non-LTR) retrotransposons amplify their copies by reverse transcribing mRNA from the 3' end, but the initial processes of reverse transcription are still unclear. We have shown that a telomere-specific non-LTR retrotransposon of the silkworm, SART1, requires the 3' untranslated region (3' UTR) for retrotransposition. With an in vivo retrotransposition assay, we identified several novel motifs within the 3' UTR involved in precise and efficient reverse transcription. Of 461 nucleotides (nt) of the 3' UTR, the central region, from nt 163 to nt 295, was essential for SART1 retrotransposition. Of five putative stem-loops formed in RNA for the SART1 3' UTR, the second stem-loop (nt 159 to 221) is included in this region. Loss of the 3' region (nt 296 to 461) in the 3' UTR and the poly(A) tract resulted in decreased and inaccurate reverse transcription, which starts mostly from several telomeric repeat-like GGUU sequences just downstream of the second stem-loop. These results suggest that short telomeric repeat-like sequences in the 3' UTR anneal to the bottom strand of (TTAGG)(n) repeats. We also demonstrated that the mRNA for green fluorescent protein (GFP) could be retrotransposed into telomeric repeats when the GFP coding region is fused with the SART1 3' UTR and SART1 open reading frame proteins are supplied in trans.

Crystal structure of the endonuclease domain encoded by the telomere-specific long interspersed nuclear element, TRAS1

The telomere-specific long interspersed nuclear element, TRAS1, encodes an endonuclease domain, TRAS1-EN, which specifically cleaves the telomeric repeat targets (TTAGG)n of insects and (TTAGGG)n of vertebrates. To elucidate the sequence-specific recognition properties of TRAS1-EN, we determined the crystal structure at 2.4-A resolution. TRAS1-EN has a four-layered alpha/beta sandwich structure; its topology is similar to apurinic/apyrimidinic endonucleases, but the beta-hairpin (beta10-beta11) at the edge of the DNA-binding surface makes an extra loop that distinguishes TRAS1-EN from cellular apurinic/apyrimidinic endonucleases. A protein-DNA complex model suggests that the beta10-beta11 hairpin fits into the minor groove, enabling interaction with the telomeric repeats. Mutational studies of TRAS1-EN also indicated that the Asp-130 and beta10-beta11 hairpin structure are involved in specific recognition of telomeric repeats.

Secondary structure models of the 3' untranslated regions of diverse R2 RNAs

The RNA structure of the 3' untranslated region (UTR) of the R2 retrotransposable element is recognized by the R2-encoded reverse transcriptase in a reaction called target primed reverse transcription (TPRT). To provide insight into structure-function relationships important for TPRT, we have created alignments that reveal the secondary structure for 22 Drosophila and five silkmoth 3' UTR R2 sequences. In addition, free energy minimization has been used to predict the secondary structure for the 3' UTR R2 RNA of Forficula auricularia. The predicted structures for Bombyx mori and F. auricularia are consistent with chemical modification data obtained with beta-ethoxy-alpha-ketobutyraldehyde (kethoxal), dimethyl sulfate, and 1-cyclohexyl-3-(2-morpholinoethyl)carbodiimide metho-p-toluene sulfonate. The structures appear to have common helices that are likely important for function.

Footprint of the retrotransposon R2Bm protein on its target site before and after cleavage

R2 elements are non-long terminal repeat (non-LTR) retrotransposons that specifically integrate into the 28 S rRNA genes of their host. These elements encode a single open reading frame with a genome-specific endonuclease and a reverse transcriptase that uses the cleaved chromosomal target site to prime reverse transcription. Cleavage of the DNA strand that is used to prime reverse transcription is an efficient process that occurs in the presence or absence of RNA. Cleavage of the second DNA strand is much less efficient and requires RNA. Reverse transcription occurs before second strand cleavage and only if the RNA bound to the protein contains the 3' untranslated region of the R2 element. Thus a complex series of protein interactions with the DNA and conformational changes in the protein are likely to occur during this retrotransposition reaction. Here, we conduct electrophoretic mobility-shift assays and DNase I footprint studies on the binding of the R2 protein to the DNA target in the presence and absence of RNA both before and after first strand cleavage. While the total expanse of the protein footprint on the DNA eventually covers five helical turns, before cleavage the footprint only extends from 17 bp to 40 bp upstream of the cleavage site. This footprint is the same in the presence and absence of RNA. We hypothesize that the active site of the endonuclease domain is analogous to type IIS restriction enzymes in that it is located on a flexible domain that is not tightly bound to the cleavage site. After first strand cleavage the protein footprint extends beyond the cleavage site. We suggest that this increased protection after cleavage is the RT domain that is positioned over the free DNA end to begin reverse transcription on the nicked DNA substrate.

Integration of the 5' end of the retrotransposon, R2Bm, can be complemented by homologous recombination

R2Bm is a non-long-terminal-repeat (non-LTR) retrotransposon that was identified at a specific target site in the 28S rRNA genes of the silkworm, Bombyx mori. Although in vitro analysis has revealed that the 3' end of R2Bm is integrated into the target site by means of target-primed reverse transcription (TPRT), the mechanism of the 5' end integration is not well understood. We established a novel in vivo system to assay the insertion mechanism of R2Bm using a cultured cell line, C65, and a baculovirus, AcNPV, as host and vector, respectively. The 3' end of R2Bm integrated at the target site in the rRNA genes of C65 cells when an AcNPV containing both the full-length 3' UTR and the entire open reading frame (ORF) of R2Bm was introduced while the 5' end integration was incorrect. The 5' end of R2Bm was integrated, however, when the 28S gene sequence upstream of the R2Bm target site was added to the R2Bm sequence. Thus, in our assay, homologous sequences were likely essential for the successful integration of the entire R2Bm into the host cell genome. We also demonstrated that the failure to integrate caused by a frame-shifted ORF was rescued by co-infection with a helper virus that contained only the R2Bm ORF. This indicates that R2 retrotransposition can be complemented in trans. These findings suggest that the host's mechanism for DNA repair may be necessary for the integration of the 5' end of R2Bm and that R2Bm protein may only have the ability to integrate the 3' end of the element by TPRT.

End-to-end template jumping by the reverse transcriptase encoded by the R2 retrotransposon

The reverse transcriptase encoded by the non-long terminal repeat retrotransposon R2 has been shown to be able to jump from the 5'-end of one RNA template (the donor) to the 3'-end of a second RNA template (the acceptor) in the absence of preexisting sequence identity between the two templates. These jumps between RNA templates have similarity to the end-to-end template jumps described for the RNA-directed RNA polymerases encoded by certain RNA viruses. Here we describe for the first time the mechanism by which such end-to-end template jumps can occur. Most template jumps by the R2 reverse transcriptase are brought about by the enzyme's ability to add nontemplated (overhanging) nucleotides to the cDNA when it reaches the end of the donor RNA. The enzyme then anneals these overhanging nucleotides to sequences at the 3'-end of the acceptor RNA. The annealing is most efficient if it involves the terminal nucleotide(s) of the acceptor RNA but can occur to sites at least 5 nucleotides from the 3'-end. These end-to-end jumps are similar to steps proposed to be part of the integration reaction of non-long terminal repeat retrotransposons and can explain chimeric integration products derived from multiple RNA templates.

Targeted nuclear import of open reading frame 1 protein is required for in vivo retrotransposition of a telomere-specific non-long terminal repeat retrotransposon, SART1

Non-long terminal repeat (non-LTR) retrotransposons, most of which carry two open reading frames (ORFs), are abundant mobile elements that are distributed widely among eukaryotes. ORF2 encodes enzymatic domains, such as reverse transcriptase, that are conserved in all retroelements, but the functional roles of ORF1 in vivo are little understood. We show with green fluorescent protein-ORF1 fusion proteins that the ORF1 proteins of SART1, a telomeric repeat-specific non-LTR retrotransposon in Bombyx mori, are transported into the nucleus to produce a dotted localization pattern. Nuclear localization signals N1 (RRKR) and N2 (PSKRGRG) at the N terminus and a highly basic region in the center of SART1 ORF1 are involved in nuclear import and the dotted localization pattern in the nucleus, respectively. An in vivo retrotransposition assay clarified that at least three ORF1 domains, N1/N2, the central basic domain, and CCHC zinc fingers are required for SART1 retrotransposition. The nuclear import activity of SART1 ORF1 makes it clear that the ORF1 proteins of non-LTR retrotransposons work mainly in the nucleus, in contrast to the cytoplasmic action of Gag proteins of LTR elements. The functional domains found here in SART1 ORF1 will be useful for developing a more efficient and target-specific LINE-based gene delivery vector.

Site-specific selfish genes as tools for the control and genetic engineering of natural populations

Site-specific selfish genes exploit host functions to copy themselves into a defined target DNA sequence, and include homing endonuclease genes, group II introns and some LINE-like transposable elements. If such genes can be engineered to target new host sequences, then they can be used to manipulate natural populations, even if the number of individuals released is a small fraction of the entire population. For example, a genetic load sufficient to eradicate a population can be imposed in fewer than 20 generations, if the target is an essential host gene, the knockout is recessive and the selfish gene has an appropriate promoter. There will be selection for resistance, but several strategies are available for reducing the likelihood of it evolving. These genes may also be used to genetically engineer natural populations, by means of population-wide gene knockouts, gene replacements and genetic transformations. By targeting sex-linked loci just prior to meiosis one may skew the population sex ratio, and by changing the promoter one may limit the spread of the gene to neighbouring populations. The proposed constructs are evolutionarily stable in the face of the mutations most likely to arise during their spread, and strategies are also available for reversing the manipulations.