Gene 395 (2007) 116 124 www.elsevier.

com/locate/gene

A new system for analyzing LINE retrotransposition in the chicken DT40 cell line widely used for reverse genetics
Hiroshi Honda a , Kenji Ichiyanagi a , Jun Suzuki a , Takao Ono a , Hideki Koyama b , Masaki Kajikawa a , Norihiro Okada a,
a

Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259-B-21 Nagatsuta-cho, Midori-ku, Yokohama, Kanagawa 226-8501, Japan b Kihara Institute for Biological Research, Graduate School of Integrated Science, Yokohama City University, 641-12 Maioka-cho, Totsuka-ku, Yokohama, Kanagawa 244-0813, Japan Received 14 September 2006; received in revised form 14 February 2007; accepted 19 February 2007 Available online 14 March 2007

Abstract Long interspersed elements (LINEs) are autonomous transposable elements that proliferate via retrotransposition, which involves reverse transcription of LINE RNAs. It is anticipated that LINE retrotransposition requires both LINE-encoded proteins and host-encoded proteins. However, identification of the host factors, their roles, and the steps at which they act on retrotransposition are poorly understood because of the lack of an appropriate genetic system to study LINE retrotransposition in a series of mutant hosts. To construct such a genetic system, we applied the retrotransposition-indicative cassette method to DT40 cells, a chicken cell line for which a variety of isogenic mutants have been established by gene targeting. Because DT40 cells are non-adherent, we utilized a selective soft agarose medium to allow the formation of colonies of cells that had undergone LINE retrotransposition. Colony formation was completely dependent on the activities of the LINE-encoded proteins and on the presence of the essential 3 region of the LINE RNA. Moreover, the selected colonies indeed carried retrotransposed LINE copies in their chromosomes, with integration features similar to those of genomic (native) LINE copies. This method thus allows the authentic selection of LINE-retrotransposed cells and the approximate recapitulation of retrotransposition events that occur in nature. Therefore, the DT40 cell system established here provides a powerful tool for the elucidation of LINE retrotransposition pathways, the host factors involved, and their roles. 2007 Elsevier B.V. All rights reserved.
Keywords: Retrotransposon; Host factor; DNA repair; Reverse genetics

1. Introduction Long interspersed elements (LINEs) are transposable elements present in a large number of eukaryotic genomes. The LINE copies are amplified by a copy-and-paste mechanism of retrotransposition. During the process, the genomic LINE DNA is first transcribed (copied) to RNA; then this RNA is
Abbreviations: cDNA, complimentary DNA; cytomegalovirus, CMV; DSBs, double-strand breaks; EGFP, enhanced green fluorescence protein; EN, endonuclease; FBS, fetal bovine serum; G418R, G418-resistant; LINEs, long interspersed elements; L1, LINE-1; L2, LINE-2; Neo, neomycin resistance; ORFs, open reading frames; ORF1p, ORF1 protein; ORF2p, ORF2 protein; RFI, relative fluorescence intensity; RNP, ribonucleoprotein; RT, reverse transcriptase; TPRT, target-primed reverse transcription; UTR, untranslated region. Corresponding author. Tel.: +81 45 924 5742; fax: +81 45 924 5835. E-mail address: nokada@bio.titech.ac.jp (N. Okada). 0378-1119/$ - see front matter 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.gene.2007.02.017

reverse transcribed (pasted) to cDNA at the site of insertion, generating a new LINE copy. The copy numbers of LINEs often reach several thousand to one million per haploid genome. For example, the human genome contains ~ 900,000 copies of LINEs, which constitute ~ 20% of the genome (Lander et al., 2001). In addition, LINEs are thought to have a large impact on the evolution of eukaryotic genomes (Kazazian, 2004); hence, elucidating LINE dynamics and the molecular mechanisms of LINE retrotransposition is of great importance in understanding the mechanisms of genome evolution. LINEs typically encode two open reading frames (ORFs), ORF1 and ORF2, both of which are required for the LINE retrotransposition (Feng et al., 1996; Moran et al., 1996). The ORF1 protein (ORF1p) binds its own RNA to form a ribonucleoprotein (RNP) complex (Hohjoh and Singer, 1996; Kolosha and Martin, 1997). In addition, mouse LINE-1 (L1) ORF1p has a nucleic acid chaperone activity, which is also

H. Honda et al. / Gene 395 (2007) 116124

117

required for L1 retrotransposition (Martin et al., 2005). Thus, ORF1p may not only be a structural component but also has an additional function(s) in retrotransposition. The ORF2 protein (ORF2p) has an endonuclease (EN) and reverse transcriptase (RT) domain and forms a large RNP complex with the LINE RNA and ORF1p (Kulpa and Moran, 2006; Matsumoto et al., 2006). During retrotransposition, the EN domain nicks a target site on chromosomal DNA, and the 3 hydroxyl group generated by the cleavage is subsequently utilized by the RT domain to prime the reverse transcription of the RNA in the complex by a mechanism called target-primed reverse transcription (TPRT) (Luan et al., 1993; Cost et al., 2002) (see Fig. 1). The late stages of LINE retrotransposition (after TPRT) involve a second DNA cleavage on the target DNA, removal of the LINE RNA, synthesis of the second (sense) strand cDNA, and ligation of the LINE cDNA and target DNA. However, these reactions have not been precisely resolved. It is anticipated that host-encoded proteins are involved in these late stages of LINE retrotransposition. For example, Gilbert et al. (2005) have proposed that products of TPRT are recognized and processed by the host DNA repair machinery. Indeed, overexpression of the human L1-encoded protein(s) in cultured cells induces double-strand breaks (DSBs) recognized by host factors on the chromosomal DNA (Gasior et al., 2006), suggesting the involvement of the host DNA repair machinery in L1 retrotransposition. Bioinformatic studies also have suggested host protein involvement in joining of the LINE and target DNA (Zingler et al., 2005; Ichiyanagi et al., 2007). Previous genetic studies (Morrish et al., 2002; Gasior et al., 2006) identified a few host mutations that affect retrotransposition efficiency from mutant cell lines generated by random mutagenesis or established from a human patient. However, a more comprehensive understanding of the roles of host proteins in LINE retrotransposition will require a systematic study examining the effects of a number of single host mutations and combinations of mutations on LINE retrotransposition. To

facilitate such studies, it would be beneficial to identify cell lines to which gene targeting can be easily applied. DT40 cells, a chicken B lymphocyte cell line, is one such candidate because genes of interest can be efficiently knocked out in this cell line (Buerstedde and Takeda, 1991). Indeed, a variety of mutant DT40 lines have been produced, including those in which DNA repair genes are knocked out (Winding and Berchtold, 2001; Yamazoe et al., 2004). Furthermore, because of the relatively stable phenotype and karyotype of DT40 cells, targeting of multiple genes is feasible as well, thus facilitating the analysis of their genetic interactions. In this study, we established a method to determine the frequency of LINE retrotransposition in DT40 cells using retrotransposition-indicative genes. 2. Materials and methods 2.1. DNA vectors The vector pEGFP-N1 (Clontech) harbors the gene encoding enhanced green fluorescence protein (EGFP), and the vector pGL3-promoter (Promega) harbors the firefly luciferase gene. The vector pBZ2-5 carries a mneoI-marked wild-type zebrafish LINE ZfL2-5 copy under the control of the cytomegalovirus (CMV) promoter (Sugano et al., 2006). The vectors p132.49 and pBB4 are retrotransposition-deficient mutant derivatives of pBZ2-5, with a point mutation in the RT domain (D689Y) and deletion of the essential 3 region of the LINE, respectively (Sugano et al., 2006). The vector p131.11 is a derivative of pBZ2-5 that was constructed in this study, in which the active site of the EN domain was disrupted by site-directed mutagenesis (E72A). The vector pEGFPFLAG-1, used for monitoring transfection efficiencies in the LINE retrotransposition experiments, was constructed by recloning the EcoRINotI fragment of pEGFP-N1 (containing the entire EGFP gene) into the same restriction sites of pFLAG-CMV-5a (SIGMA) to create a vector that carries the EGFP gene but not the neomycinresistance (Neo) gene. The ZfL2-2 vector, designated as pZfL22/mEGFPi, which is marked with the EGFP-based retrotransposition-indicative gene, was constructed by replacement of the mneoI region in pBZ2-5 with the EGFP-intron region in pBS-L1RP-EGFP (Ostertag et al., 2000). 2.2. Cell culture, transfection, and reporter gene expression monitoring Chicken DT40 cells were purchased from RIKEN Biosource Center (number RCB1464) and were cultured in RPMI medium 1640 (Invitrogen), supplemented with 10% fetal bovine serum (FBS), 1% chicken serum and 50 M 2-mercaptoethanol in a humidified atmosphere with 5% CO2 at the temperatures indicated. For transfection, exponentially growing DT40 cells (~ 7 106 cells) were harvested and gently resuspended in 200 l of medium. Cells were then mixed with 15 g of the vector DNA indicated (or 15 g each of two vectors for cotransfection experiments), which had been purified using the

Fig. 1. A model for retrotransposition of ZfL2-2 and other LINEs. The original copy of ZfL2-2 is represented as an open rectangle on the top left, with a single ORF (gray rectangle) that carries the endonuclease (EN) and reverse transcriptase (RT) domains. The 3 region of ZfL2-2 essential for retrotransposition is indicated by a slashed box. Bold arrows indicate the progression of the LINE retrotransposition steps: transcription, translation, RNP formation, target DNA cleavage, target-primed reverse transcription (TPRT), and later steps, which are not yet elucidated.

118

H. Honda et al. / Gene 395 (2007) 116124

QIAfilter Plasmid Mega Kit (Qiagen), and placed in a 3-mm cuvette. Transfection was carried out by electroporation at 250 V and 960 F on the GENE Pulser (Bio-Rad). To examine the reporter expression from pEGFP-N1 at different temperatures, the electroporated cells were diluted into 10 ml of medium and then further diluted by fourfold. Duplicate 10-ml cultures were then incubated at 33 C and 37 C in 100-mm dishes. The fraction of EGFP-expressing cells was determined at each time point using the FACSCalibur Flow Cytometry System (Becton-Dickinson). The threshold value of relative fluorescence intensity (RFI) per cell, which discriminates EGFP-positive and -negative cells, was determined by analyzing the RFI in cells of an untransfected control culture. The EGFPpositive region (M1) in Fig. 2B was determined as the region in which only 0.05% of the control cells were assigned. To examine the reporter expression from pGL3-promoter, DT40 cells were cotransfected with pGL3-promoter and pEGFPN1. At each time point, cells were sampled, the luciferase activity was measured by the Steady-Glo Luciferase Assay System (Promega) with the TopCount NXT (PerkinElmer), and EGFPexpressing cells were counted as described above. For normal-

ization, the relative luminescence units (RLU) of the luciferase were divided by the number of EGFP-expressing cells in the culture assayed. 2.3. Retrotransposition assay DT40 cells were cotransfected with pEGFPFLAG-1 and one of the LINE expression vectors, pBZ2-2, p132.49, pBB4, p131.11, or pJM102/L1.3 (Sassaman et al., 1997). After transfection, cells were diluted (see Section 2.2) and incubated at 33 C for 3 days to allow the LINEs to retrotranspose. After this post-transfection incubation, the numbers of EGFP-positive and -negative cells were counted by flow cytometry to monitor the transfection efficiency. For gelation and selection, 5 ml of the culture (diluted with medium to 1.42.0 105 cells/ml, giving a total of 0.71.0 106 cells) in a 100-mm dish was mixed with an equal volume of agarose-RPMI medium 1640 containing 0.24% agarose, 30% FBS, 3% chicken serum, and 3.2 mg/ml G418, which had been pre-incubated at 45 C (Adachi et al., 2001). After an 11-day incubation at 37 C to allow the formation of visible colonies, G418-resistant (G418R) colonies

Fig. 2. Expression of reporter genes in DT40 cells. (A) The fractions of EGFP-expressing cells in transiently transfected cultures. Cells electroporated with pEGFP-N1 were incubated at 33 C (diamonds) and 37 C (triangles) in two independent experiments (closed and open symbols). The percentage of EGFP-expressing cells was monitored by flow cytometry for 3 days after transfection. (B) Histograms of relative fluorescence intensity (RFI) of EGFP in individual transfected cells. After a 74hour incubation at 33 C (upper) or 37 C (lower) after electroporation, the RFI of EGFP in each cell was measured by flow cytometry. M1 indicates the region in which cells were assigned as EGFP-positive based on a negative control experiment (see Section 2.2). (C) Maintenance of luciferase activities in DT40 cells transfected with a transiently introduced vector. Cells electroporated with pEGFP-N1 and pGL3-promoter were incubated for 3 days at 33 C (diamonds) or 37 C (triangles), and relative luminescence units (RLU) of luciferase per EGFP-expressing cell were determined at each time point. Results of two independent experiments are shown (closed and open symbols). (D) Growth curves of DT40 cells. DT40 cells at a density of 5 105 cells/ml were incubated at 33 C (closed squares) or 37 C (open squares) for 48 h. At each time point, cell densities were monitored and plotted using a semi-log scale. The error bars indicate standard deviations of data from four independent experiments.

H. Honda et al. / Gene 395 (2007) 116124

119

were counted. Retrotransposition frequency was calculated as the number of G418R colonies per EGFP-positive cell. For the EGFP-based retrotransposition-indicative cassette, DT40 cells were transfected with pZfL2-2/mEGFPi, and the post-transfection incubation was carried out as described above. The cells were then observed under a fluorescence microscope to assess whether any cells expressed active EGFP. 2.4. DNA analysis of the genomes of G418R clones Four independent G418R colonies were picked and cultured in new liquid medium at 37 C for 4 to 10 days (until reaching confluency). Genomic DNA was then isolated from each culture using the DNeasy Tissue Kit (Qiagen). PCR was carried out to examine the structure of the mneoI cassette integrated in these genomic DNAs with the NeoBglF1 and NeoBamR1 primers (see Supplementary Table 1). The products were analyzed on a 1% agarose gel. The genomic DNA of untransfected DT40 cells was used as a negative control. The sequences of the 3 LINE-target junctions were determined by cassette PCR (Siebert et al., 1995). Genomic DNA (~1 g) from two of the G418R clones was digested with EcoRV and SmaI and ligated to a cassette that had been prepared by annealing of the two oligonucleotides CasL-1 and CasBlu-2. The ligation products were used as templates for nested PCR. The first PCR was performed with the mneoI-specific primer F1 and the cassette-specific primer AP1, and the second PCR was performed with the ZfL2-2-specific primer F3 and another cassette-specific primer, AP2. The products of the second PCR were separated on a 1% agarose gel and analyzed by Southern hybridization with a 32P-labeled ZfL2-2 probe (P1) to identify specifically amplified products. We cloned and sequenced two of the PCR products. The chromosomal sites of the two insertions were identified by BLAST search (Altschul et al., 1990) against the sequenced chicken genome (Hillier et al., 2004). To determine the sequences of the 5 junctions of these integrants, we designed primers specific to the upstream (5) regions of the insertions based on the chicken genome sequenceW1F and W2F for the integrants on chromosome 24 and 15, respectively. The 5 junction regions were amplified from the genomic DNA by PCR with NeoBamR1 and the respective chromosome-specific primers. The amplified fragments were directly sequenced. We also designed primers specific to the 3 regions of the insertions W1R and W2R, respectively. Using W1F and W1R (or W2F and W2R), the linearity between the two independently determined flanking regions in these integrants was confirmed by PCR analysis. 3. Results 3.1. Reduced temperature improves maintenance of gene expression from non-replicating vectors in DT40 cells In human cells, analysis of LINE retrotransposition with the retrotransposition-indicative cassette mneoI in a well-established assay utilizes the episomal replication of the LINE expression vector (Moran et al., 1996). Such an episomal vector

is, however, not available for chicken DT40 cells at present. We considered that transient transfection with a non-replicating vector would be applicable for retrotransposition detection in DT40 cells as well as in rodent cells (Morrish et al., 2002). Because LINE retrotransposition requires transcription and translation (Fig. 1), the level and maintenance of LINE expression are important factors for efficient retrotransposition. Therefore, we initially examined the effects of incubation temperature on expression from a transiently introduced DNA vector. DT40 cells were transfected with the pEGFP-N1 reporter, which encodes EGFP, and the fraction of EGFPexpressing cells was monitored by FACS analysis for three days (Fig. 2A). About 20 to 35% of the electroporated cells expressed EGFP at 26 h after electroporation; the fraction of total cells in the M1 region approximates the transfection efficiency (Fig. 2B). EGFP expression was detected even at 74 h after electroporation, with about 15 to 30% of cells expressing EGFP (Fig. 2A). The decay slopes of the EGFP-expressing cells were not affected by the incubation temperatures. However, the RFI of EGFP in the transfected cells was altered by the incubation temperature (Fig. 2B); most cells incubated at 33 C had an RFI of 100 to 10,000 or more, whereas the majority of cells incubated at 37 C had RFIs of 10 to 1000. Therefore, although the fraction of EGFP-positive cells did not differ, the expression level in individual cells was significantly diminished by incubation at 37 C after transfection. We examined this temperature effect more quantitatively by monitoring luciferase activity from the transiently introduced reporter pGL3-promoter vector. At 26 h after electroporation, the luciferase activity was detected in cultures incubated at both 33 C and 37 C (Fig. 2C). The RLUs of luciferase per transfected cell were 3 to 4 times lower when incubated at 37 C compared with 33 C. Moreover, luciferase activity was barely detectable in the culture at 37 C at 50 and 74 h after electroporation, whereas the culture incubated at 33C had detectable activity even at 74 h after electroporation. The growth rate of DT40 cells was approximately two times lower at 33 C than at 37 C (Fig. 2D). Because the nonreplicating vectors become diluted through cell division, the slower growth at 33 C would attenuate the rate of dilution, thus maintaining a higher level of expression from the vector for a longer period. Therefore, in the following retrotransposition studies, we incubated transfected cells at 33 C for 3 days to allow better expression of the genetically marked LINEs. 3.2. Detection of ZfL2-2 retrotransposition in DT40 cells ZfL2-2, which is also known as CR1-2_DR (Kapitonov and Jurka, 2003), is a zebrafish LINE that resembles human LINE-2 (and thus is classified into the L2 clade of LINEs) (Sugano et al., 2006). ZfL2-2 encodes a single protein containing EN and RT domains (Fig. 3A). The 3 untranslated region (UTR) of this element contains a sequence essential for retrotransposition, and it likely forms a secondary structure when transcribed into RNA; the UTR ends with a repeated hexanucleotide, (AAATGT)n (Kapitonov and Jurka, 2003; Kajikawa et al., 2005; Sugano et al., 2006). We have shown that this zebrafish

120

H. Honda et al. / Gene 395 (2007) 116124

LINE retains retrotransposition activity in human HeLa cells (Sugano et al., 2006); therefore, we used ZfL2-2 to test if retrotransposition of genetically marked LINEs is also detectable in heterologous chicken DT40 cells.

The LINE expression vector originally used for experiments with HeLa cells, pBZ2-5 (Sugano et al., 2006), carries an active genomic copy of ZfL2-2 marked with mneoI (Fig. 3B). In the mneoI assay, the ZfL2-2 element is placed under the control of the CMV promoter, and the Neo gene is inserted in the antisense orientation immediately downstream of the ZfL2-2 ORF. This Neo gene is disrupted by the insertion of an intron sequence in the antisense orientation. Thus, the intron can be spliced from the LINE transcript but not from the Neo gene transcript. Transcription, splicing of the marked ZfL2-2 RNA, reverse transcription of the spliced RNA, and integration of the resulting cDNA into a chromosomal site restore an intact Neo gene, allowing phenotypic selection for retrotransposition (Fig. 3B) (Sugano et al., 2006). The pBZ2-5 DNA was cotransfected with the EGFP vector (pEGFPFLAG-1) into DT40 cells, and the cells were incubated at 33 C for 3 days to allow the LINE to retrotranspose (Fig. 3C). After this post-transfection incubation, the transfection efficiency was monitored by EGFP expression, and the cells were seeded into medium in the presence of G418. Because DT40 cells do not adhere to the dish, we used soft agarose plates and counted the number of G418R colonies that formed. Incubation for 11 days at 37 C yielded approximately 150 colonies per dish (Fig. 3D left), suggestive of LINE retrotransposition. The number of G418R colonies per EGFP-expressing cell, which was taken as the retrotransposition frequency, was ~ 1.8 10 3. Retrotransposition frequency was decreased by 1.5-fold by the post-transfection incubation at 37 C, presumably reflecting the lower level of LINE expression. To confirm that the G418R colonies arose via LINE retrotransposition, we first examined the effects of LINE mutations on colony formation. A mutation altering the active
Fig. 3. ZfL2-2 retrotransposition assay in DT40 cells. (A) A schematic of zebrafish LINE ZfL2-2. The 5 UTR, ORF and 3 UTR are indicated. The EN and RT domains in the ORF are shown as black rectangles. The slashed box represents the 3 tail required for ZfL2-2 retrotransposition. (B) Phenotypic selection of retrotransposition by use of mneoI. The neomycin-resistance gene (light gray rectangles indicated by N and eo) in the mneoI cassette is flanked by a promoter (P) and a polyadenylation signal (A) (top row). pCMV, the cytomegalovirus promoter. SVpA, the SV40 polyadenylation signal. After transcription and polyadenylation of the LINE RNA (second row), the intron in mneoI is excised (third row). Reverse transcription of the RNA and cDNA integration into a new site results in regeneration of the Neo gene (bottom row), thus conferring G418 resistance to the host cell. (C) Retrotransposition assay in DT40 cells. The two vectors, pBZ2-5 and pEGFPFLAG-1 (upper left), were cointroduced into DT40 cells (represented by small open circles in the leftmost large circle). After post-transfection incubation for 3 days at 33 C, transfection efficiencies were monitored by determining the fraction of cells expressing EGFP (represented by small light and dark gray circles in the middle). Among the transfected cells, only those that had restored the Neo gene by retrotransposition (represented by small dark gray circles) can form colonies in a soft agarose plate containing G418 (in the rightmost circle). The panel shows an image of a colony of G418R DT40 cells formed in the agarose plate. (D) Retrotransposition-dependent G418R colony formation. Images show 100mm dishes with G418R colonies. Cells were transfected with pBZ2-5 (WT), p132.49 (RTm, an RT mutant), or p131.11 (ENm, an EN mutant). N, the number of independent experiments; TE, the mean values ( standard deviation) for transfection efficiency; NC, the mean number (standard deviation) of G418R colonies per 100-mm dish; RF, the mean values (standard deviation) for retrotransposition frequency.

H. Honda et al. / Gene 395 (2007) 116124

121

site of the RT domain of the protein (D689Y) and deletion of the 3 UTR each disrupt the retrotransposition activity of ZfL2-2 in HeLa cells (Sugano et al., 2006). When these mutant LINE vectors were introduced into DT40 cells, no G418R colonies appeared (Fig. 3D middle and data not shown). In addition, when a point mutation was introduced into the active site of the EN domain, again no G418R colonies appeared (Fig. 3D right). These results indicate that the formation of G418R colonies is completely dependent on the retrotransposition activity of the LINE that carries mneoI. In addition, G418R colonies were also formed by transient transfection of the mneoI-marked pJM102/L1.3 vector, which carries a human L1 element (Sassaman et al., 1997). The retrotransposition frequency of human L1 in DT40 cells was

calculated as ~ 5.9 10 3, which is ~ 3 times higher than that of ZfL2-2. Thus, it is likely that retrotransposition of any type of active LINE is investigable in DT40 cells. 3.3. Analysis of ZfL2-2 integrants in the genomic DNA of G418R DT40 cells To further study the nature of the genetic alterations in the G418R cells, genomic DNA samples were prepared from four independent G418R clones. First, we tested for the presence of the intron-excised derivative of the mneoI cassette in their genomes. The genomic DNAs prepared were analyzed by PCR such that both the original (2.1 kb) and intron-excised (1.3 kb) mneoI could be amplified (Fig. 4A). The intron-less band was

Fig. 4. ZfL2-2 integrants in chromosomal DNA from transfected DT40 cells. (A) Detection of the mneoI cassette DNA in G418R DT40 cells. The DNA fragment of the mneoI cassette was amplified by PCR with the primers NeoBglF1 and NeoBamR1 (left) and analyzed on an agarose gel (right). pBZ2-5 DNA (lane 1), genomic DNA of untransfected DT40 cells (lane 2), and genomic DNAs isolated from four independent G418R clones (lanes 36). Lane M, 1-kb plus marker DNAs. (B) The structures of ZfL2-2 retrotransposition products. Sequences of the integration site pre- and post-integration and schematic representations of the integrants on chromosomes 24 (Integrant 1) and 15 (Integrant 2) are shown. The sequences of the chicken genome and ZfL2-2 DNA are shown in uppercase and lowercase letters, respectively. The sequence of the ZFL2-2 3 terminal repeat is shown in bold italics. The 5 and 3 microhomologous sequences are underlined.

122

H. Honda et al. / Gene 395 (2007) 116124

indeed amplified from all four genomic DNAs, whereas no intron-positive band was amplified. These data indicate that the intron-excised mneoI cassette was indeed inserted and fixed somewhere in the genomes of the G418R clones and that the donor LINE vector DNA was lost during the course of incubation. Second, to study the features of the retrotransposition products in these cells in detail, we determined the sequences of two retrotransposition sites. The 3 junctions of the integrants in two of the clones were amplified by cassette PCR (see Section 2.4). Southern blot analysis of the PCR products revealed that these clones carried multiple insertions (data not shown), which also has been observed with the mneoI cassette in mammalian cells (Wei et al., 2000; Symer et al., 2002). Sequencing of two of the bands revealed that the LINE was inserted in chromosomes 24 and 15 of the chicken genome. We thus designed chromosomal position-specific primers for both integrations to amplify fragments containing the upstream (5) region of the target sequence and the inserted copy of ZfL2-2 (see Section 2.4). The integrant on chromosome 24 (Fig. 4B, Integrant 1) carried a 5-truncated ZfL2-2 copy of 4.2 kb (including the Neo gene) but lacked the intron sequence. Moreover, the inserted copy ended with the 3 terminal repeat. These features strongly suggest that this product resulted from TPRT-initiated retrotransposition rather than another type of DNA recombination. This ZfL2-2 copy shared microhomologous sequences with the target sequence at both the 5 and 3 junctions (3 and 5 bp, respectively). Thus, a target sequence of 2 bp, at the longest, was duplicated at each LINE end. The integrant on chromosome 15 (Fig. 4B, Integrant 2) also contained a 5-truncated ZfL2-2 copy that ended with the 3 terminal repeat (the length of the insertion was 0.7 kb) and had microhomologous sequences of 2 and 4 bp at the 5 and 3 junctions, respectively. In this instance, a target sequence of at least 8 bp was truncated. Importantly, all of the features of the two insertions agreed well with those observed in native ZfL2-2 integrants in the zebrafish genome (Ichiyanagi et al., 2007), suggesting that retrotransposition in DT40 cells proceeds via pathways similar to those utilized in nature. 3.4. Detection of retrotransposition using an EGFP-based retrotransposition-indicative cassette We also tested if LINE retrotransposition in DT40 cells could be detected by use of an EGFP-based retrotranspositionindicative cassette (Ostertag et al., 2000). Like mneoI, this cassette is inactivated by the insertion of an antisense intron and can only be reactivated by retrotransposition, driving the host cell to express EGFP. DT40 cells were transfected with pZfL22/mEGFPi, a ZfL2-2 vector containing the EGFP cassette, and the culture was incubated at 33 C for 3 days. After this incubation, EGFP-expressing cells were detected by fluorescence microscopy; however, the fraction of EGFP-expressing cells of the total cells transfected was estimated to be only ~ 10 6 to 10 5, which was too low to be quantified by flow cytometry. In addition, when the temperature for the posttransfection incubation was shifted to 37 C, no EGFPexpressing cells were observed in ~ 107 transfectants, consistent

with the results with mneoI that incubation at 33 C yielded better retrotransposition frequency. Thus, the EGFP cassette, although detectable at 33 C, is not as appropriate as the mneoI retrotransposition-indicative cassette for quantitative studies of LINE retrotransposition. 4. Discussion In this study, we constructed a new system to detect LINE retrotransposition in DT40 cells via the mneoI cassette. The difficulty arising from the absence of an episomal vector that stably provides LINE transcripts was solved by transient transfection of a non-replicating vector followed by lowtemperature incubation to maintain a relatively high level of LINE expression during the period required for a detectable number of cells to undergo retrotransposition. The second problem in the previous methodology was the non-adherent property of DT40 cells, but this was overcome by using the soft agarose technique (Adachi et al., 2001). Thus, our method enabled reproducible frequencies of detection of G418R DT40 colonies produced via LINE retrotransposition. The establishment of this DT40 cell system will allow us to progress to the next stage of LINE mobility studies. The involvement of host-encoded proteins in LINE retrotransposition is very poorly understood, although circumstantial evidence suggests such involvement (Morrish et al., 2002; Gilbert et al., 2005; Zingler et al., 2005; Gasior et al., 2006; Ichiyanagi et al., 2007). DT40 cells are genetically modifiable by gene targeting, such that mutant derivatives of the gene(s) of interest can be constructed by knockout (Winding and Berchtold, 2001; Yamazoe et al., 2004). Therefore, effects of a series of host mutations on LINE retrotransposition can be studied by use of isogenic cell lines to identify host-encoded proteins involved in LINE retrotransposition. Moreover, even cell lines carrying mutations in multiple genes can be constructed for DT40 cells; therefore, effects of combinations of host mutations can be examined by conventional genetic studies to identify genetic interactions and epistasis groups for the LINE mobility process. These studies will help to delineate the LINE retrotransposition pathways, especially the late steps following the TPRT reaction. In the course of the study, we also detected retrotransposition of human L1 in DT40 cells. L1 is distantly related to ZfL2-2 in the LINE phylogeny (Ohshima and Okada, 2005). L1 and ZfL22 show interesting differences in sequence and mobility. First, L1 carries two ORFs, like many other LINEs, whereas ZfL2-2 has a single ORF. Second, ZfL2-2 retrotransposition requires a specific sequence in its 3 UTR like many other LINEs (Luan and Eickbush, 1995; Okada et al., 1997; Kajikawa and Okada, 2002; Osanai et al., 2004; Sugano et al., 2006), whereas L1 retrotransposition does not require such a sequence, but rather the polyadenosine added at the end of the RNA serves as a template for reverse transcription (Moran et al., 1996; Cost et al., 2002; Kulpa and Moran, 2006). Thus, studies of retrotransposition of these LINEs in wild-type and mutant DT40 cell lines could reveal general and specific aspects of the modes of LINE retrotransposition.

H. Honda et al. / Gene 395 (2007) 116124

123

The retention of LINE retrotransposition activity in heterologous hosts (Eickbush et al., 2000; Kajikawa and Okada, 2002; Sugano et al., 2006, etc) suggests that LINE retrotransposition involves host proteins highly conserved between native and experimental model hosts. On the other hand, the features of the 5 junctions in ZfL2-2 copies that retrotransposed in human cells have been shown to resemble those of L1 integrants in human rather than ZfL2-2 integrants in zebrafish (Ichiyanagi et al., 2007). This argues in favor of the idea that the step(s) to join the 5 junction depends on host cell (or organism) components rather than on the LINEs themselves. In other words, the retrotransposition intermediates may be processed by the available host DNA repair system(s). The relative availability of each of these DNA repair systems may differ among hosts, resulting in different pathways of retrotransposition. Recently, a human pre-B cell line, Nalm-6, was shown to be highly proficient for gene targeting (Adachi et al., 2006), although not many knockout cell lines have been constructed. If mutant Nalm-6 cells were compared to mutant DT40 cells in the retrotransposition assay described here, the effects of equivalent mutations on LINE retrotransposition in human and chicken could be used to advance our understanding of the adaptability of LINE mobility pathways as well as host strategies to adapt to these mobile elements. This information could improve our understanding of the effects of LINEs on genomic evolution. Acknowledgments We thank Dr. John Moran for generously providing the plasmid pJM102/L1.3. We thank Drs. Eric Ostertag and Haig Kazazian for generously providing the plasmid pBS-L1RPEGFP containing the EGFP retrotransposition-indicative cassette. We thank Mr. Katsumi Yamaguchi for his assistance in the retrotransposition assay in DT40 cells. Mr. Takayuki Ishino (Ishiyaku Publishers, Inc.) is acknowledged for his helpful suggestions on the DT40 experiments. This work was supported by a Grant-in-Aid to N.O. from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.gene.2007.02.017. References
Adachi, N., Ishino, T., Ishii, Y., Takeda, S., Koyama, H., 2001. DNA ligase IVdeficient cells are more resistant to ionizing radiation in the absence of Ku70: implications for DNA double-strand break repair. Proc. Natl. Acad. Sci. U. S. A. 98, 1210912113. Adachi, N., et al., 2006. The human pre-B cell line Nalm-6 is highly proficient in gene targeting by homologous recombination. DNA Cell Biol. 25, 1924. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403410. Buerstedde, J.M., Takeda, S., 1991. Increased ratio of targeted to random integration after transfection of chicken B cell lines. Cell 67, 179188. Cost, G.J., Feng, Q., Jacquier, A., Boeke, J.D., 2002. Human L1 element targetprimed reverse transcription in vitro. EMBO J. 21, 58995910.

Eickbush, D.G., Luan, D.D., Eickbush, T.H., 2000. Integration of Bombyx mori R2 sequences into the 28S ribosomal RNA genes of Drosophila melanogaster. Mol. Cell. Biol. 20, 213223. Feng, Q., Moran, J.V., Kazazian Jr., H.H., Boeke, J.D., 1996. Human L1 retrotransposon encodes a conserved endonuclease required for retrotransposition. Cell 87, 905916. Gasior, S.L., Wakeman, T.P., Xu, B., Deininger, P.L., 2006. The human LINE-1 retrotransposon creates DNA double-strand breaks. J. Mol. Biol. 357, 13831393. Gilbert, N., Lutz, S., Morrish, T.A., Moran, J.V., 2005. Multiple fates of L1 retrotransposition intermediates in cultured human cells. Mol. Cell. Biol. 25, 77807795. Hillier, L.W., et al., 2004. Sequence and comparative analysis of the chicken genome provide unique perspectives on vertebrate evolution. Nature 432, 695716. Hohjoh, H., Singer, M.F., 1996. Cytoplasmic ribonucleoprotein complexes containing human LINE-1 protein and RNA. EMBO J. 15, 630639. Ichiyanagi, K., Nakajima, R., Kajikawa, M., Okada, N., 2007. Novel retrotransposon analysis reveals multiple mobility pathways dictated by hosts. Genome Res. 17, 3341. Kajikawa, M., Ichiyanagi, K., Tanaka, N., Okada, N., 2005. Isolation and characterization of active LINE and SINEs from the eel. Mol. Biol. Evol. 22, 673682. Kajikawa, M., Okada, N., 2002. LINEs mobilize SINEs in the eel through a shared 3 sequence. Cell 111, 433444. Kapitonov, V.V., Jurka, J., 2003. The esterase and PHD domains in CR1-like non-LTR retrotransposons. Mol. Biol. Evol. 20, 3846. Kazazian Jr., H.H., 2004. Mobile elements: drivers of genome evolution. Science 303, 16261632. Kolosha, V.O., Martin, S.L., 1997. In vitro properties of the first ORF protein from mouse LINE-1 support its role in ribonucleoprotein particle formation during retrotransposition. Proc. Natl. Acad. Sci. U. S. A. 94, 1015510160. Kulpa, D.A., Moran, J.V., 2006. Cis-preferential LINE-1 reverse transcriptase activity in ribonucleoprotein particles. Nat. Struct. Mol. Biol. 13, 655660. Lander, E.S., et al., 2001. Initial sequencing and analysis of the human genome. Nature 409, 860921. Luan, D.D., Eickbush, T.H., 1995. RNA template requirements for target DNAprimed reverse transcription by the R2 retrotransposable element. Mol. Cell. Biol. 15, 38823891. Luan, D.D., Korman, M.H., Jakubczak, J.L., Eickbush, T.H., 1993. Reverse transcription of R2Bm RNA is primed by a nick at the chromosomal target site: a mechanism for non-LTR retrotransposition. Cell 72, 595605. Martin, S.L., et al., 2005. LINE-1 retrotransposition requires the nucleic acid chaperone activity of the ORF1 protein. J. Mol. Biol. 348, 549561. Matsumoto, T., Hamada, M., Osanai, M., Fujiwara, H., 2006. Essential domains for ribonucleoprotein complex formation required for retrotransposition of telomere-specific non-long terminal repeat retrotransposon SART1. Mol. Cell. Biol. 26, 51685179. Moran, J.V., Holmes, S.E., Naas, T.P., DeBerardinis, R.J., Boeke, J.D., Kazazian Jr., H.H., 1996. High frequency retrotransposition in cultured mammalian cells. Cell 87, 917927. Morrish, T.A., et al., 2002. DNA repair mediated by endonuclease-independent LINE-1 retrotransposition. Nat. Genet. 31, 159165. Ohshima, K., Okada, N., 2005. SINEs and LINEs: symbionts of eukaryotic genomes with a common tail. Cytogenet. Genome Res. 110, 475490. Okada, N., Hamada, M., Ogiwara, I., Ohshima, K., 1997. SINEs and LINEs share common 3 sequences: a review. Gene 205, 229243. Osanai, M., Takahashi, H., Kojima, K.K., Hamada, M., Fujiwara, H., 2004. Essential motifs in the 3 untranslated region required for retrotransposition and the precise start of reverse transcription in non-long-terminal-repeat retrotransposon SART1. Mol. Cell. Biol. 24, 79027913. Ostertag, E.M., Prak, E.T., DeBerardinis, R.J., Moran, J.V., Kazazian Jr., H.H., 2000. Determination of L1 retrotransposition kinetics in cultured cells. Nucleic Acids Res. 28, 14181423. Sassaman, D.M., et al., 1997. Many human L1 elements are capable of retrotransposition. Nat. Genet. 16, 3743. Siebert, P.D., Chenchik, A., Kellogg, D.E., Lukyanov, K.A., Lukyanov, S.A., 1995. An improved PCR method for walking in uncloned genomic DNA. Nucleic Acids Res. 23, 10871088.

124

H. Honda et al. / Gene 395 (2007) 116124 Yamazoe, M., Sonoda, E., Hochegger, H., Takeda, S., 2004. Reverse genetic studies of the DNA damage response in the chicken B lymphocyte line DT40. DNA Repair (Amst) 3, 11751185. Zingler, N., et al., 2005. Analysis of 5 junctions of human LINE-1 and Alu retrotransposons suggests an alternative model for 5 -end attachment requiring microhomology-mediated end-joining. Genome Res. 15, 780789.

Sugano, T., Kajikawa, M., Okada, N., 2006. Isolation and characterization of retrotransposition-competent LINEs from zebrafish. Gene 365, 7482. Symer, D.E., et al., 2002. Human L1 retrotransposition is associated with genetic instability in vivo. Cell 110, 327338. Wei, W., Morrish, T.A., Alisch, R.S., Moran, J.V., 2000. A transient assay reveals that cultured human cells can accommodate multiple LINE-1 retrotransposition events. Anal. Biochem. 284, 435438. Winding, P., Berchtold, M.W., 2001. The chicken B cell line DT40: a novel tool for gene disruption experiments. J. Immunol. Methods 249, 116.