This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kojima, K. K.
Right arrow Articles by Fujiwara, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kojima, K. K.
Right arrow Articles by Fujiwara, H.

 Previous Article  |  Next Article 

Molecular and Cellular Biology, September 2005, p. 7675-7686, Vol. 25, No. 17
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.17.7675-7686.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.

Eukaryotic Translational Coupling in UAAUG Stop-Start Codons for the Bicistronic RNA Translation of the Non-Long Terminal Repeat Retrotransposon SART1

Kenji K. Kojima,{dagger} Takumi Matsumoto, and Haruhiko Fujiwara*

Department of Integrated Biosciences, Graduate School of Frontier Sciences, University of Tokyo, Kashiwa-shi, Chiba 277-8562, Japan

Received 16 December 2004/ Returned for modification 19 January 2005/ Accepted 14 June 2005


arrow
ABSTRACT
 
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.


arrow
INTRODUCTION
 
In many retroviruses and retrotransposons, gag protein and pol protein are encoded in different open reading frames (ORFs) on a single RNA transcript (4, 14). They translate these two proteins as gag-pol polyproteins. Retroviruses, such as human immunodeficiency virus type 1 and human T-cell lymphotropic virus type 1, have a –1 translational frameshift signal, and Ty1, one of the yeast long terminal repeat (LTR) retrotransposons, has a +1 frameshift signal to translate gag-pol polyproteins (4, 14). The signal for retroviral –1 frameshifting in RNA is composed of two structures, a slippery sequence where the frameshifting takes place and the downstream stem-loop or pseudoknot structure (16). The general slippery sequence in retroviruses is the heptamer nucleotide sequence X XXY YYZ. The tRNA initially bound to XXY slips to bind to XXX at the P site, and the tRNA initially bound to YYZ slips to bind to YYY at the A site. The mechanism for +1 frameshifting in Ty1 is entirely different from the retroviral –1 frameshifting. A codon for low available tRNA interrupts the translation elongation and induces the tRNA slippage (15).

Non-LTR retrotransposons are considered the ancestor of LTR retrotransposons and retroviruses (25). Although more ancient classes of non-LTR retrotransposons have a single ORF, the recently branched non-LTR retrotransposons usually have two ORFs (24). The first ORF (ORF1) of the latter type of non-LTR retrotransposons encodes a gag-like protein and the second ORF (ORF2) a pol-like protein, which are similar to LTR retrotransposons and retroviruses. ORF2 encodes two essential catalytic domains, endonuclease and reverse transcriptase (Fig. 1, top). In contrast to LTR retrotransposons and retroviruses, there have been very few observations of ORF2 translation in non-LTR retrotransposons. The separate ORF2 protein (ORF2p) has been detected in L1 elements of human and rat, but the translational mechanism is still obscure (7, 12). The translational mechanism of the bicistronic RNAs in non-LTR retrotransposons remains to be determined.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 1. SART1 structure. The nucleotide and amino acid sequences near the ORF1-ORF2 overlapping region are shown below the ORF structure. The first and the second AUG codons in ORF2 are underlined. Putative stop codons are boxed.

SART1, which has two overlapping ORFs, is an active telomeric repeat-specific non-LTR retrotransposon in the silkworm Bombyx mori (37). Most of the SART1 elements among the several hundred exist as full-length genomic copies. In addition, only a single band corresponding to a full-length SART1 unit was observed in several Bombyx tissues by Northern hybridization (35). Recently, we constructed an in vivo transposition assay system of SART1 in Sf9, the moth Spodoptera frugiperda cell line, using the expression of Autographa californica nuclear polyhedrosis virus (AcNPV) (26, 29, 36). Both the silkworm, B. mori, and the moth, S. frugiperda, belong to the same order, Lepidoptera. This system reproduces native transposition features, such as frequent 5' truncation and target specificity for telomeres. Thus, the functional ORF2p of SART1 is translated in the AcNPV expression system, which will be useful for studying the translational control in non-LTR retrotransposons. In addition, we previously reported that the SART1 RNA transcribed from the polyhedrin promoter showed a band which corresponds to the full-length SART1 and the downstream polyhedrin 3' region (29). There were no clear bands below the full-length transcripts. Because both native SART1 and AcNPV-expressed SART1 are transcribed as the full-length transcripts, it is unlikely that a cryptic promoter leads to transcription of a monocistronic RNA encoding only ORF2. SART1 ORF2p seems to be translated from a single full-length, bicistronic SART1 RNA that is actively transcribed.

In SART1, ORF2 overlaps the end of ORF1 in the –1 frame, which is similar to retroviruses (Fig. 1). The first AUG codon in a reading frame of SART1 ORF2 is positioned at 3018 to 3020 and the second at 3096 to 3098 (Fig. 1, underlined). The previous stop codon in a reading frame of SART1 ORF2 is positioned at 2949 to 2951 (Fig. 1, UAG [boxed]). Therefore, the overlapping region of ORF1 and ORF2 is 64 nucleotides (nt) in length (Fig. 1, bottom). However, interestingly, the first AUG codon (3018 to 3020) in ORF2 is surrounded by two stop codons, UAA (3016 to 3018) and UGA (3019 to 3022), in an ORF1 reading frame (Fig. 1, both are boxed), which makes the stop-start fusion sequence UAAUGA (the start codon is underlined). It is of great interest to know whether such an extraordinary structure is involved in translational mechanisms in non-LTR retrotransposons.

In this study, we detected the HA (hemagglutinin influenza virus epitope)-tagged ORF2p translated from the bicistronic RNA of SART1. The ORF2p of SART1 was translated as an independent protein separately from ORF1p, which is totally different from gag-pol polyproteins in LTR retrotransposons and retroviruses. In addition, we found that the UAAUGA overlapping stop-start fusion codon at the junction between ORF1 and ORF2 and the downstream RNA secondary structure in ORF2 play an essential role for the ORF2 translation initiation of SART1. In addition, we showed that the stop codon of ORF1 and the start codon of ORF2 should be located in the neighborhood for efficient translation of SART1 ORF2 protein. These features are analogous to translational coupling observed in prokaryotes and some viruses, which suggests that SART1 ORF2 is also translated by translational coupling.


arrow
MATERIALS AND METHODS
 
Plasmid construction and recombinant AcNPV generation. Primer sequences used in this study are listed in Table 1. The SART1 ORF1-ORF2 sequence was amplified by PCR from SART1WT-pAcGHLTB (36) with primers SART1-S880-NcoI-S->G and SART1-A3670-NotI for S1-3666HA, SART1-A4303-NotI for S1-4304HA, and SART1-A6110-NotI for S1-6110HA. PCR was conducted for 30 cycles using Pfu Turbo DNA polymerase (Stratagene). The PCR products were subcloned between the NcoI and NotI sites of pAcGHLT-B (PharMingen). The HA tag sequence was amplified by PCR from pGADT7 (Clontech) using the primers pGADT7-S190-NotI and pGADT7-A2004-TAABglII. PCR was conducted for 30 cycles using Ex-Taq DNA polymerase (TaKaRa). The PCR products were subcloned between the NotI and BglII sites of pAcGHLT-B already inserted with SART1 partial sequences. Point mutations and nucleotide insertions/deletions were introduced with pairs of primers using the QuickChange mutagenesis kit (Stratagene). Double mutations were introduced using two-step site-directed mutagenesis. The mutation of each plasmid was confirmed by DNA sequencing.


View this table:
[in this window]
[in a new window]
  		TABLE 1. List of primers

Mutant k, containing a deletion of nt 3055 to 3111, was constructed as follows. The region 880 to 3054 of SART1 was amplified by PCR from SART1-3666HA using the primers SART1-S880-NcoI-S->G and SART1-A3054-XbaI. PCR was conducted for 30 cycles using Pfu Turbo DNA polymerase. The PCR products were subcloned between the NcoI and XbaI sites of pET14b (Novagen). Bases 3112 to 3666 of SART1 and the HA tag sequence were amplified by PCR from SART1-3666HA using the primers SART1-S3112-XbaI and pGADT7-A2004-TAABglII. PCR was conducted for 30 cycles using Pfu Turbo DNA polymerase. The PCR products were subcloned between the XbaI and BglII sites of pET14b already inserted with SART1 partial sequences. The NcoI site upstream of the HA tag sequence was excluded by mutagenesis. The region including SART1 and HA tag was amplified by PCR with the primers SART1-S880-NcoI-S->G and pGADT7-A1968-TAABglII. PCR was conducted for 30 cycles using Pfu Turbo DNA polymerase, and the products were subcloned between the NcoI and BglII sites of pAcGHLT-B plasmids.

Construct I for in vivo retrotransposition assay is the same as SART1WT-pAcGHLTB (36), and construct II is the same as SART1ORF1pWT (26). Construct III was constructed as follows. The SART1 ORF2 sequence was amplified by PCR from the genomic library clone BS103 using the primers SART1-S3014-NcoI and SAX-3P-NotI. PCR was conducted for 30 cycles using Pfu Turbo DNA polymerase (Stratagene). The PCR products were subcloned between the NcoI and NotI sites of pAcGHLT-B (PharMingen). Construct IV was made as follows. The whole pAcGHLT-B plasmid except the glutathione S-transferase (GST)-His tag sequence was amplified by PCR using the 5'-phosphorylated primers pAcB-A2265 and pAcB-S2941. The PCR products were self-ligated using a DNA ligation kit version 2 (TaKaRa) at 16°C for 16 h. The portion of SART1 ORF2 and 3' untranslated region (UTR) was amplified by PCR from the genomic library clone BS103 with the primers SART1-S3014-NdeI and SART1-A6704-BglII. The PCR products were subcloned between the NdeI and BamHI sites of pGADT7. The region including SART1 ORF2-3' UTR and HA tag was amplified by PCR using the primers T7 and SART1-A6704-BglII. The PCR products were subcloned between the NcoI and BglII sites of the vector lacking the GST-His tag.

Recombinant AcNPVs were generated as described by Takahashi and Fujiwara (36).

Western blotting. Approximately 106 Sf9 cells were infected in a 24-well plate with recombinant AcNPV at a multiplicity of infection of 10 PFU per cell. At 72 h postinfection, cells were scraped, pelleted at 1,000 x g for 5 min, and extracted. Cell extracts were separated on sodium dodecyl sulfate-polyacrylamide gels and electroblotted onto Fluoro Trans W polyvinylidene difluoride membranes (Nippon Genetics). Blots were blocked with 5% nonfat dry milk (Bio-Rad) in TBS-T (50 mM Tris-HCl, 0.3 M NaCl, 0.05% Tween 20, pH 8.0). Membranes were incubated for 16 h at room temperature with mouse monoclonal anti-HA antibody (Roche) diluted 1:10,000 or mouse monoclonal anti-His antibody (Amersham) diluted 1:3,000 in TBS-T. Bound antibodies were detected using an ECL Plus Western blotting detection system (Amersham) with peroxidase-conjugated anti-mouse immunoglobulin G (Amersham) diluted 1:25,000 in TBS-T. Between steps, membranes were washed according to the manufacturer's instructions.

In vivo retrotransposition assay of SART1. The in vivo retrotransposition assay was performed essentially as described previously (36). Approximately 106 Sf9 cells were infected in a 12-well plate with SART1-containing AcNPV at a multiplicity of infection of 10 PFU per cell. In the trans-complementation experiments, Sf9 cells were infected with two recombinant AcNPVs at a multiplicity of infection of 5 PFU. At 72 h after infection, cells were scraped and pelleted at 1,000 x g for 5 min, and the total genomic DNAs were purified with PUREGENE cell and tissue DNA isolation kits (Gentra). PCR amplification was carried out with LA-Taq DNA polymerase (TaKaRa) and the primer set SART1-S6311 and CCTAA+T (Table 1) in the presence of 10 ng of Sf9 DNA. The reaction mixture was denatured at 96°C for 2 min, followed by 35 cycles of 98°C for 20 s, 62°C for 30 s, and 72°C for 30 s. The PCR products were electrophoresed on 2% agarose gels in Tris-borate-EDTA buffer and visualized by ethidium bromide staining.


arrow
RESULTS
 
Overlapping stop-start codons are distributed widely in non-LTR retrotransposons. The first AUG codon at 3018 to 3020 in SART1 ORF2 overlaps with two consecutive stop codons, UAA and UGA, in an ORF1 reading frame (Fig. 1, bottom). Many non-LTR retrotransposons have similar overlapping stop-start codons (Table 2), which are classified into three types. More than 10 non-LTR retrotransposon families in the R1, the LOA, the Jockey, and the L1 clades have the SART1-type UAAUG (UAA plus AUG) stop-start codon. The second type of overlapping stop-start codon, AUGA (AUG plus UGA), is also found in the widespread non-LTR retrotransposons, in the R1, the Tad, the Jockey, and the L1 clades. The third type, UGAUG (UGA plus AUG), is found only in JuanAg3 of the Jockey clade, among the so-far-characterized non-LTR retrotransposons. All three types of overlapping stop-start codons join downstream ORFs to upstream ORFs in the –1 frame. These non-LTR retrotransposons are distributed among all eukaryotic kingdoms. In addition, the overlapping stop-start codons are distributed in most clades of bicistronic non-LTR retrotransposons. The distribution of overlapping stop-start codons is not necessarily correlated with the phylogeny of non-LTR elements. For example, RT1 and RTAg4 in Anopheles gambiae have the AUGA overlapping stop-start codon, but their close relatives, RT2 and RTAg3 in the same insect, do not have the AUGA codon. Among 25 non-LTR retrotransposons in the R1 clade, in which the ORF1-ORF2 junction sequences have been so far characterized, five elements (SART1, HOPEBm1, HOPEBm2, R6Ag1, and R1Dm) have UAAUG and three (RT1, RTAg4, and HidaAg1) have AUGA. Another 17 retrotransposons (TRAS1, TRAS3, R1Bm, SARTPx1, Waldo-A, Waldo-B, WaldoAg1, WaldoAg2, R6Ag2, R6Ag3, RT2, RTAg3, MinoAg1, R7Ag1, R7Ag2, KagaAg1, and NotoAg1) have no overlapping stop-start codons. These facts indicate that overlapping stop-start codons were independently acquired in many non-LTR retrotransposon lineages.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Overlapping stop-start codons in non-LTR retrotransposons

SART1 ORF2 is translated separately from ORF1, not by frameshifting. The primary question is whether SART1 ORF2p is translated by ribosomal frameshifting or is translated from the first AUG codon in ORF2. We expressed SART1 in an AcNPV expression system. A GST-His6 tag was fused at the N terminus of ORF1p, and an HA tag was fused at the C terminus of ORF2p (Fig. 2A). The sequence of the overlapping region of ORF1 and ORF2 in respective constructs is the same as in the native SART1. We made three constructs differing at the C-terminal truncation of ORF2p. S1-3666HA has ORF2 truncated just after the endonuclease domain (Fig. 2A, 1). S1-4304HA has ORF2 truncated just before the reverse transcriptase domain (Fig. 2A, 2). S1-6110HA has almost full-length ORF2 (Fig. 2A, 3). In addition, we made a construct in which ORF1 and truncated ORF2 were fused (Fig. 2A, 4). RNA including both ORF1 and ORF2 was presumed to be transcribed from the AcNPV polyhedrin promoter, based on the observation of our previous report (29). Because the SART1 5' UTR was replaced by the GST-His tag sequence in the AcNPV-expressed SART1, the length of SART1 RNA expressed by AcNPV was 127 nt shorter than the native SART1.



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 2. ORF2 translation assay by Western blotting. (A) Structures of AcNPV-expressed constructs. ZF, zinc finger; EN, endonuclease; RT, reverse transcriptase; GST, glutathione S-transferase; HA, hemagglutinin influenza virus epitope tag. (B) AcNPV-expressed ORF1p visualized by Western blotting with anti-His antibody. Numbers of lanes indicate constructs shown in panel A. (C) AcNPV-expressed ORF2p visualized by Western blotting with anti-HA antibody.



View larger version (35K):
[in this window]
[in a new window]
  		FIG. 3. Summary of ORF2 translation assay. Red filled circles indicate actual start codons, a red open circle indicates a non-AUG start site, and black open circles indicate in-frame AUG codons which were not used for translation initiation. The distances of AUG codons from the actual stop codon are shown below the ORF structures. Mutations are summarized at the middle, and the detailed sequences are at the bottom. Mutagenized nucleotides are in lowercase. Putative start codons are underlined, and actual start codons are red. Putative stop codons are boxed, and actual stop codons are in purple boxes. Results of the ORF2 translation assay are summarized at the right. ORF2p expression is indicated as the efficiency relative to the wild type (a). Minuses indicate no expression. MW, molecular mass.



View larger version (67K):
[in this window]
[in a new window]
  		FIG. 4. Results of ORF2 translation assay. (A) AcNPV-expressed ORF1p visualized by Western blotting with anti-His antibody. Letters of lanes indicate constructs shown in Fig. 3. The molecular weight of ORF1p of each construct is shown below. (B) AcNPV-expressed ORF2p visualized by Western blotting with anti-HA antibody. The arrow indicates the ORF2p bands.

GST-His-tagged ORF1p was expressed in three constructs at the expected size (106 kDa), and the GST-His-tagged ORF1-ORF2 fusion protein (133 kDa) was expressed in S1-F3666HA (Fig. 2B). We observed only a single strong band in each lane. Faint bands in lanes 2 and 4 were probably degraded products. The HA-tagged ORF2p bands were observed in both S1-3666HA and S1-4304HA but not in S1-6110HA (Fig. 2C). Even when we increased the exposure time, we could not detect ORF2p bands in S1-6110HA (data not shown). It is possible that no band for the full-length ORF2p (S1-6110HA) (Fig. 2C, lane 3) was due to transcriptional disruption reported in the human L1 (9). The molecular masses of ORF2p of S1-3666HA (Fig. 2C, lane 1) and S1-4304HA (Fig. 2C, lane 2) were close to the estimated sizes, 27 kDa and 56 kDa, respectively, assuming that the first AUG in ORF2 was used for the initiation of translation. If ORF1-ORF2 fusion proteins were expressed, the estimated molecular masses were 133 kDa for S1-3666HA (Fig. 2C, lane 1) and 162 kDa for S1-4304HA (Fig. 2C, lane 2). However, we could not find any bands above 100 kDa (Fig. 2C, lanes 1 and 2). We detected ORF1-ORF2 fusion proteins at the expected size (133 kDa) in S1-F3666HA (Fig. 2C, lane 4). If ORF1-ORF2 fusion proteins were translated and processed in S1-3666HA and the band in lane 1 represented proteins after proteolysis, the ORF1-ORF2 fusion proteins in S1-F3666HA would also be processed (Fig. 2C, lanes 1 and 4). Therefore, ribosomal frameshifting and proteolysis are unlikely mechanisms for the production of ORF2p.

The first AUG in ORF2 is used as a start codon. To investigate whether the first AUG in ORF2 is the actual start codon of ORF2, we introduced a substitution of the first AUG in ORF2 to ACG (Fig. 3, b) or AGC (Fig. 3, c) in S1-3666HA (Fig. 2A, 1). ACG has a weak activity for translation initiation in eukaryotic cells (30), and thus the order of translation initiation potency is AUG > ACG > AGC. When ACG is used for translation initiation, ACG is recognized by tRNA bound to methionine, as with AUG (30). The ORF1p was detected as an expected 106-kDa band in all three constructs (Fig. 4A, a to c). The substitution from AUG to ACG reduced, but did not completely diminish, the ORF2 translation, whereas the substitution from AUG to AGC appeared to eliminate it (Fig. 4B, a to c). To exclude the possibility of unexpected mutations in plasmid construction and recombinant AcNPV generation, we sequenced the mutant AcNPV genome. There were no unexpected mutations in ORF2 (data not shown). The reduction of ORF2 translation in mutants indicated that the first AUG in ORF2 is used as a start codon. The translation efficiency of ORF2 accorded with the potency of translation initiation of each codon. Judging from the ORF2p/ORF1p ratio quantified by densitometry, the translation efficiency of ORF2 in mutant b was approximately 4% of the wild type (a) (the results are summarized in Fig. 3). Unexpectedly, in mutant b, ORF2p was translated from the mutated start codon ACG (3018), not from the second in-frame AUG (3096) codon 78 nt downstream from the native start codon (Fig. 3, b). If the translation started at the second AUG codon, the expected size of ORF2p would be 24 kDa, 3 kDa smaller. There was no band below 27 kDa (Fig. 4B, b).

The overlapping stop-start codon is essential for ORF2 translation. The next question is whether the overlapping stop-start codon is important for ORF2 translation. We made three other mutants. In mutant d, the substitution from UAA at position 3016 to 3018 to CAA formed another overlapping stop-start codon, AUGA (AUG plus UGA) at 3018 to 3021 (Fig. 3, d). The same overlapping stop-start codon, AUGA, is observed in some non-LTR retrotransposons (Table 2). In mutant e, the second stop codon, UGA at 3019 to 3021, was changed to UGG in addition to the first stop codon mutation, and the start codon of ORF2 was not mutated (Fig. 3, e). In this mutant, the stop codon of ORF1 is positioned at 3085 to 3087, 69 nt downstream from the native stop codon and 67 nt downstream from the native start codon of ORF2. If the 69-nt-downstream stop codon in mutant e is used, ORF1p is expected to be 108 kDa, which is 2.64 kDa larger than ORF1p in constructs a to d. In mutant f, two consecutive stop codons and the start codon were all destroyed (Fig. 3, f). In this mutant, the stop codon of ORF1 is also 69 nt downstream from the native stop codon, as in mutant e. The molecular weight of ORF1p in f is also expected to be 108 kDa. As the native start codon is mutated, AUG at 3096 to 3099 becomes the first AUG codon in ORF2 in mutant f, 8 nt downstream from the stop codon of ORF1.

The ORF1p bands in d to f could be detected (Fig. 4A, a to f). The strong ORF2p band was observed in d, but no bands were detected in e and f (Fig. 4B, d to f), even though the start codon of ORF2 was intact in e (Fig. 3, e). The molecular weight of ORF2p in d is equal to that in a, which supports that ORF2p in d is also translated from the first AUG codon at 3018 in ORF2. Judging from the ORF2p/ORF1p ratio quantified by densitometry, the translation efficiency of ORF2 in d was approximately 13% of wild type. This result indicates that another stop-start codon, AUGA, was used in d in place of UAAUG.

Greater distance between the stop codon and the start codon diminishes ORF2 translation. The stop codon of ORF1 and the start codon of ORF2 are at their closest when they form overlapping stop-start codons. We considered that ORF2 was not translated in mutant e because the stop codon of ORF1 was 67 nt from the start codon of ORF2 (Fig. 3, e). Therefore, we further analyzed the effect of the distance between the stop codon of ORF1 and the start codon of ORF2. In mutant g, the stop codon of ORF1 is positioned at 2803 to 2805, 215 nt upstream from the start codon of ORF2, and there are no differences around the start codon of ORF2 (Fig. 3, g). In mutant h, the first AUG in ORF2 was mutated, and a new in-frame AUG was created at 2967 to 2969, 49 nt upstream from the stop codon of ORF1 (Fig. 3, h). The sequence near the stop codon of ORF1 in h is the same as that in c.

Although the ORF1p bands were observed in mutants g and h at the expected sizes of 98 kDa and 106 kDa, respectively (Fig. 4A, g and h), we could not detect ORF2p (Fig. 4B, g and h). Because the sequence around the first AUG in ORF2 of g is identical to the wild type (Fig. 3, g), the premature translation termination of ORF1 considerably reduced ORF2 translation. In addition, the artificial upstream AUG in h could not substitute for the native start codon.

Overlapping stop-start codon is necessary but not sufficient for ORF2 translation. Next, we examined whether the overlapping stop-start codon is sufficient for ORF2 translation, introducing an artificial stop-start codon, UAAUG, 51 nt upstream from the native UAAUG. In both mutant i and mutant j, UAA at 2965 to 2967 is the stop codon of ORF1 and AUG at 2967 to 2969 is the first AUG in ORF2 (Fig. 3, i and j). There are no mutations near the native start codon at 3018 to 3020 in i. However, in mutant i, UAA at 3016 to 3018 is not the stop codon of ORF1, and AUG at 3018 to 3020, the start codon of ORF2 in wild-type SART1, is not the first AUG codon in ORF2. In mutant j, the native start codon, AUG at 3018 to 3020, is changed to AGC, which is the same mutation as in c and h. If artificial UAAUG were sufficient for the translation of ORF2, the 29-kDa ORF2p would be expressed.

We detected the slightly smaller ORF1p bands in both mutants (Fig. 4A, i and j), but we could not detect clear ORF2p bands in either construct (Fig. 4B, i and j). Thus, the overlapping stop-start codon is essential but is not enough for the translation of ORF2. These results showed that an artificial overlapping stop-start codon cannot replace the native overlapping stop-start codon. There is another factor necessary for the translation of ORF2.

Downstream RNA secondary structures also affect ORF2 translation. In retroviruses, the frameshift signal is composed of a slippery sequence and a downstream stem-loop RNA secondary structure (16). Although there are no stable stem-loop structures in the ORF1-ORF2 overlapping region (2952 to 3015) of SART1, the RNA secondary structure prediction program at GeneBee-NET server (6) predicts a very stable stem-loop structure at position 3055 to 3111, downstream of the overlapping stop-start codon UAAUG (Fig. 5A, left). We predicted another possibility for a pseudoknot structure manually (Fig. 5A, right). Sequences 3055 to 3079 and 3088 to 3111 are predicted to form a stem, and this stem formation is necessary for both the stem-loop and the pseudoknot. If CUGCUG at 3081 to 3086 in the loop annealed to CAGCAG at 3023 to 3028, the stem-loop structure would change into the pseudoknot structure.



View larger version (43K):
[in this window]
[in a new window]
  		FIG. 5. ORF2 translation assay for the downstream RNA secondary structure. (A) Predicted secondary structure downstream from the overlapping stop-start codon of SART1. Mutagenized sequences are shown below. (B) Conserved stem-loop structures of three related retrotransposons. Two bold lines in the SART1 structure indicate sequences assumed to be annealed in the pseudoknot model. (C) AcNPV-expressed ORF1p visualized by Western blotting with anti-His antibody. Letters of the lanes indicate constructs shown in panel A and Fig. 3. (D) AcNPV-expressed ORF2p visualized by Western blotting with anti-HA antibody.

We also predicted RNA stem-loop structures just downstream of UAAUG in HOPEBm1 and HOPEBm2 (Fig. 5B), which are close relatives of SART1 (17) and have the SART1-type overlapping stop-start codons (Table 2 and Fig. 5B). The spacing between UAAUG and the stem-loops in HOPEBm1 and HOPEBm2 is 34 nt, equal to SART1. The nucleotides forming stems are highly conserved among these three retrotransposons. However, HOPEBm1 and HOPEBm2 cannot form the pseudoknot structures that could be predicted in SART1 (Fig. 5B).

We made four constructs to investigate whether stem-loop or pseudoknot structures are important for ORF2 translation. In mutant k, the downstream stem-loop structure (3055 to 3111) was entirely deleted from S1-3666HA (Fig. 5A, k). In this construct, the downstream region from UAAUG can form neither the stem-loop nor the pseudoknot. The expected size of ORF2p in k was 24 kDa. In mutants l and m, the presumed pseudoknot would be weakened but the stem-loop structure would not (Fig. 5A, l and m). The presumed pseudoknot and the stem-loop structure would be as stable in mutant n as in the wild type because two sets of mutated nucleotides can form base pairs (Fig. 5A, n).

We observed ORF1p bands of expected sizes in all constructs (Fig. 5C). The ORF2p band appeared to be completely eliminated in k and reduced in l, m, and n (Fig. 5D). Judging from the ORF2p/ORF1p ratio quantified by densitometry, the translation efficiency of ORF2 was low in l and m (approximately 6 to 10% of the wild type) but was recovered in n (approximately 20%), even though the efficiency was still lower than in the wild type (Fig. 5C and D). These results suggested that the stem formation is essential for the translation of ORF2 and that annealing between CUGCUG at 3081 to 3086 in the loop and CAGCAG at 3023 to 3028 is involved in the translation efficiency. It is also possible that the primary sequence is important for the ORF2 translation instead of the pseudoknot formation.

ORF2p acts as an independent functional unit. As described above, SART1 ORF2 is translated separately from ORF1, which suggests that ORF1p and ORF2p can be supplied for retrotransposition independently. To investigate this possibility, we performed the in vivo retrotransposition assay reported previously (26, 29, 36). We used four constructs (Fig. 6A). Construct I contains a full-length SART1. Construct II includes only ORF1. Constructs III and IV have only ORF2 and 3' UTR with either a GST-His tag (III) or an HA tag (IV) at the N terminus of ORF2. If ORF1p and ORF2p can be supplied independently, coinfection of AcNPV encoding only ORF1 (II) and AcNPV encoding ORF2 and 3' UTR (III or IV) should lead to retrotransposition.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 6. In vivo retrotransposition assay. (A) Diagram of constructs. (B) AcNPV-expressed ORF1p (I and II) or ORF2p (III) visualized by Western blotting with anti-His antibody (top) and AcNPV-expressed ORF2p (IV) visualized by Western blotting with anti-HA antibody (bottom). Letters of lanes indicate constructs shown in panel A. (C) PCR amplification of the 3' junctions of the retrotransposed SART1. PCR was conducted with a SART1 internal primer and a telomeric repeat primer. The expected size of PCR products is approximately 400 bp (indicated by the arrowhead).

Expression of ORF1p and/or ORF2p was confirmed by Western blotting with anti-His antibody (ORF1p in constructs I and II and ORF2p in construct III) or anti-HA antibody (ORF2p in construct IV) (Fig. 6B). Interestingly, when we coinfected constructs II and III with the same multiplicity of infection (5 PFU per cell each), the ORF2p signal was much weaker than the ORF1p signal (Fig. 6B, II + III). It indicates that ORF2 is less efficiently translated than ORF1 by nature.

In our in vivo retrotransposition assay system, retrotransposition can be detected by PCR with a SART1 internal primer (SART1-S6311) and a telomeric repeat-specific primer (CCTAA+T) (26, 29, 36). We detected the PCR products, which represented the retrotransposition of SART1 into the telomeric repeats in Sf9 cells, when construct I was infected (Fig. 6C, I). Infection of either construct II alone or construct III alone could not lead to retrotransposition (Fig. 6C, II and III), but coinfection resulted in retrotransposition (Fig. 6C, II + III). We exchanged the N-terminal tag from a GST-His tag to an HA tag in order to confirm the ability of trans-complementation of ORF1p and ORF2p (Fig. 6C, II, II + IV, and IV). These results showed that ORF1p and ORF2p can be supplied for retrotransposition independently and indicated that ORF2p translated from the first AUG in ORF2 is functional.


arrow
DISCUSSION
 
SART1 uses internal translation initiation, not ribosomal frameshifting. In this study, we revealed that SART1 ORF2p is translated from the start codon AUG at the middle of SART1 RNA. Ilves et al. detected the rat L1 ORF2p translated in vitro by autoradiography, but they could not clarify how the separate ORF2p is translated (12). Recently, Ergün et al. detected the in vivo human L1 ORF2p by immunohistochemistry with anti-ORF2p antibody, but they did not approach the ORF2 translational mechanism (7). One of the difficulties in detecting ORF2p is due to the low expression of ORF2p even in the context of high-copy plasmids and overexpression. Actually, ORF2p expression (construct III) was weaker than ORF1p expression (II), even in the monocistronic RNA constructs (Fig. 6, II + III). We used a baculovirus expression system to produce a large amount of SART1 bicistronic RNA and could detect ORF2p by Western blotting (Fig. 2C). Even using this baculovirus expression system, we could not detect the full-length ORF2p (S1-6110HA) (Fig. 2C, 3). Using the truncated ORF2 constructs with HA tag, we clarified the internal translation initiation for SART1 ORF2p. Many LTR retrotransposons and retroviruses apply ribosomal frameshifting to translate their gag-pol polyproteins. In contrast, some non-LTR retrotransposons, including SART1, adopt quite a different mechanism for the translation of their bicistronic RNA from LTR retrotransposons and retroviruses.

Translational mechanism of SART1 ORF2: translational coupling. Several eukaryotic mechanisms for internal translation initiation have been reported (20). The scanning model predicts that the ribosome enters at the 5' terminus of mRNA and migrates on mRNA from 5' to 3'. One mechanism for internal translation initiation is leaky scanning, in which the ribosome bypasses AUG codons in a weak context for translation initiation (20). As there are 24 AUG codons before the ORF2 translation initiation site, leaky scanning can be excluded for the translational mechanism of SART1 ORF2. Another possible mechanism is nonlinear movement of the ribosome from the 5' end of RNA to the first AUG in ORF2 according to ribosome shunting (8). If SART1 used ribosome shunting, the shunting ribosome would translate ORF2 and the remaining ribosome would translate ORF1, and the ORF2 translation would not be affected by the position of ORF1 translation termination. Thus, the above result that the premature translation termination of ORF1 diminished ORF2p translation (Fig. 3 and 4, g and i) excludes the possibility of ribosome shunting.

Internal ribosome entry represents another way to initiate translation far downstream of the 5' end of RNA (32). Internal ribosome entry needs a specific RNA secondary structure, named an internal ribosome entry site (IRES). Insect picorna-like viruses have IRESs in their intercistronic region to translate their bicistronic RNA (38), although most viral and cellular IRESs are positioned in the 5' untranslated regions (20, 32). The independence of the expression of a downstream ORF from translation of a preceding ORF represents a key feature of internal ribosome entry. Therefore, the disappearance of ORF2p by premature translation termination of ORF1 (Fig. 3 and 4, g and i) rejects the possibility of internal ribosome entry for the mechanism of SART1 ORF2 translation.

Another possibility of internal translation initiation is translation reinitiation after the translation of an upstream ORF. Several examples of translation reinitiation in eukaryotes have been reported (18, 23), but the mechanisms of reinitiation seem distinct from SART1 internal translation initiation by two points. First, reinitiation is efficient when the upstream ORF is less than 30 codons long (18). In the well-known reinitiation system for the yeast GCN4, each of four upstream ORFs consists of only three or four codons (10). In contrast, SART1 ORF2p is translated after the translation of ORF1p, which is 712 amino acids in length. Second, reinitiation is most efficient when the upstream ORF terminates some distance before the start codon of the downstream ORF (18, 23). Reinitiation of the yeast GCN4 is inefficient when the stop codon of the upstream ORF is 56 nt upstream from AUG of the downstream ORF but efficient when the stop codon is 176 nt or 201 nt upstream from AUG in a natural (nonstarvation) condition (10). The distance between the upstream ORF and downstream ORF is necessary for ribosomes to bind the eIF2-GTP-Met-tRNAiMet ternary complex (10). In the translation of SART1 ORF2, the first AUG of ORF2 at 3018 to 3020 served as a site for translation initiation only when it was 2 nt downstream (Fig. 3 and 4, a) or 1 nt upstream (d) from the stop codon of ORF1 but did not do so when it was 67 nt upstream (e), 215 nt downstream (g), or 53 nt downstream (i). The change of translation efficiency following the change of distance between two ORFs in SART1 ORF2 translation is opposite to that in reinitiation.

The overlapping stop-start codon UAAUG plays a key role for the ORF2 translation of SART1. Such an overlapping stop-start codon is also observed in bacterial genes. The UGAUG stop-start codon is observed at the cistronic junction between trpB and trpA in Escherichia coli (3), and UAAUG is situated between the coat gene and the lysis gene in GA RNA phage (13). Expression of these genes is translationally coupled through their overlapping stop-start codons. AUGA also has a capacity for effective translational coupling (33). In the case of translational coupling, the following ORF is translated exclusively by the ribosome that translates the preceding ORF. In addition, increasing the distance between the stop codon of the preceding ORF and the start codon of the following ORF decreases the translation of the following ORF (33). The effect of the distance between two ORFs of SART1 is similar to that in translational coupling.

Although translational coupling is observed mainly in prokaryotes, the existence of eukaryotic translational coupling has been proved by some experiments and the observation of translational mechanisms of several RNA viruses (2, 11, 28, 31). Peabody and Berg showed the existence of translational coupling machinery in mammalian cells using artificial bicistronic RNA (31). Influenza B virus RNA segment 7 contains a UAAUG overlapping stop-start codon, which plays a central role in the translation of the downstream ORF (11). Termination-dependent translation reinitiation with short intercistronic sequences was reported for two viruses, human respiratory syncytial virus (HRSV) (2) and rabbit hemorrhagic disease virus (RHDV) (28). Although these viruses do not contain overlapping stop-start codons, the translational mechanism of downstream ORF could be similar to that of SART1 and influenza virus because some of their related viruses have overlapping stop-start codons. Feline calicivirus and Manchester virus, which are related to RHDV, and pneumonia virus of mice, which is related to HRSV, contain AUGA overlapping stop-start codons (2, 28). Norwalk virus has UAAUG (28). In RHDV, the substitution from AUG to ACG or AUC resulted in about one-fourth of the wild-type expression level of the downstream ORF; in contrast, the substitution to UGU reduced it to nearly zero (28). In our experiments, the substitution from AUG to ACG reduced the SART1 ORF2 translation, whereas the substitution from AUG to AGC appeared to eliminate it (Fig. 4B, a to c). The effects of substitutions at AUG are similar between SART1 and RHDV.

Associated with the mechanism of translational coupling, the downstream RNA secondary structure reinforces ORF2 translation in SART1. The most likely role of the RNA secondary structure of SART1 is blocking the movement of ribosomes to allow sufficient time for recognizing a start codon. The spacing between the start codon and the stem-loop is 34 nt and between the start codon and the presumed pseudoknot is 2 nt (Fig. 5A). It was reported that a downstream RNA secondary structure facilitates translation initiation most efficiently when it is separated from the AUG codon by 14 nt (19). This indicates that the downstream RNA secondary structure of SART1 could not directly lead ribosomes to recognize the ORF2 start codon. Adhin and van Duin proposed that the loosened ribosome may slip forward or backward to locate a reinitiation site in prokaryotes (1). This ribosomal scanning-like movement has a range of action of more than 40 nt. Kozak reported that the backward movement of eukaryotic ribosomes after the translation of short ORFs is limited (21), but it was the case of translation termination reinitiation, not the case of translational coupling. There remains the possibility that the downstream RNA secondary structure of SART1 obstructs the ribosome forward movement and leads the backward movement in order to recognize the ORF2 start codon.

Do other non-LTR retrotransposons use translational coupling? HOPEBm1 and HOPEBm2 are the close relatives of SART1 and have overlapping stop-start codons (UAAUG) and downstream RNA secondary structures (Table 2 and Fig. 5B). Translational coupling is likely to have been acquired in the common ancestor of SART1, HOPEBm1, and HOPEBm2. Some non-LTR retrotransposons that are phylogenetically distant from SART1 also have overlapping stop-start codons (Table 2). These retrotransposons could also use translational coupling for their ORF2 translation. However, not all non-LTR retrotransposons have overlapping stop-start codons. For example, SARTPx1, another close relative of SART1, has ORF2 overlapping ORF1 in the –1 frame but does not have an overlapping stop-start codon (17). Some non-LTR retrotransposons, like human L1 or I factor of Drosophila, have two separate ORFs. In these elements, internal translation initiation was indicated because there was no detection of ORF1-ORF2 fusion proteins (5, 12, 27). Many non-LTR retrotransposons do not have overlapping stop-start codons, but most of them encode methionines at the beginning of endonuclease domains (data not shown). We consider that the ORF2 protein of many non-LTR retrotransposons could be translated from the first AUG in ORF2. However, we could not exclude the possibility of ribosomal frameshifting for the ORF2 translation of some non-LTR retrotransposons. TRAS1 and TRAS3, which belong to a family of the silkworm telomeric repeat-specific retrotransposons other than SART1, have ORF2 overlapping ORF1 in the +1 frame, and their first AUG in ORF2 is positioned inside the conserved endonuclease domains (22). We further characterized the ORF1-ORF2 junction sequences of other TRAS family retrotransposons, TRAS4, TRAS5, and TRAS6. However, no TRAS family retrotransposons have methionines at the beginning of endonuclease domains (unpublished data). The TRAS family is likely to apply ribosomal frameshifting for their ORF2 translation, although further analysis is necessary to ascertain this possibility.

Translational coupling for eukaryotic cellular genes. SART1 is encoded on the host genome. This is different from other examples of eukaryotic translational coupling, all of which are RNA viruses. SART1 RNA is considered to be transcribed by cellular transcriptional machinery and translated by cellular translational machinery like host genes. Eukaryotic translational coupling is likely to be also used by cellular bicistronic genes. Mouse embryonic RNA splicing variant of glutamic acid decarboxylase (GAD) is a strong candidate for cellular bicistronic RNA translated by translational coupling. The bicistronic RNA of GAD contains a UGAUG overlapping stop-start codon, and its downstream ORF is translated in vivo (34). UGAUG in GAD is analogous to UAAUG in SART1. This bicistronic RNA is produced through the developmentally regulated alternative splicing of a single exon. It is likely that other cellular genes have bicistronic mRNA forms that are translated by translational coupling.


arrow
ACKNOWLEDGMENTS
 
This work was supported by grants from the Ministry of Education, Science and Culture of Japan and by a Grant-in-Aid from the Research for the Future Program of Japan and Research Fellowships for Young Scientists of the Japan Society for the Promotion of Science (JSPS).


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Integrated Biosciences, Graduate School of Frontier Sciences, University of Tokyo, Bioscience Building, 501, Kashiwa-shi, Chiba 277-8562, Japan. Phone: 81-4-7136-3659. Fax: 81-4-7136-3660. E-mail: haruh{at}k.u-tokyo.ac.jp. Back

{dagger} Present address: Bioinformatics Center, Institute for Chemical Research, Kyoto University, Uji, Kyoto 611-0011, Japan. Back


arrow
REFERENCES
 
    1
  1. Adhin, M. R., and J. van Duin. 1990. Scanning model for translational reinitiation in eubacteria. J. Mol. Biol. 213:811-818.[CrossRef][Medline]
    2. 2
  2. Ahmadian, G., J. S. Randhawa, and A. J. Easton. 2000. Expression of the ORF-2 protein of the human respiratory syncytial virus M2 gene is initiated by a ribosomal termination-dependent reinitiation mechanism. EMBO J. 19:2681-2689.[CrossRef][Medline]
    3. 3
  3. Aksoy, S., C. L. Squires, and C. Squires. 1984. Translational coupling of the trpB and trpA genes in the Escherichia coli tryptophan operon. J. Bacteriol. 157:363-367.[Abstract/Free Full Text]
    4. 4
  4. Belcourt, M. F., and P. J. Farabaugh. 1990. Ribosomal frameshifting in the yeast retrotransposon Ty: tRNAs induce slippage on a 7 nucleotide minimal site. Cell 62:339-352.[CrossRef][Medline]
    5. 5
  5. Bouhidel, K., C. Terzian, and H. Pinon. 1994. The full-length transcript of the I factor, a LINE element of Drosophila melanogaster, is a potential bicistronic RNA messenger. Nucleic Acids Res. 22:2370-2374.[Abstract/Free Full Text]
    6. 6
  6. Brodskii, L. I., V. V. Ivanov, I. L. Kalaidzidis, A. M. Leontovich, V. K. Nikolaev, S. I. Feranchuk, and V. A. Drachev. 1995. GeneBee-NET: an Internet-based server for analyzing biopolymers structure. Biokhimiia 60:1221-1230. (In Russian.)[Medline]
    7. 7
  7. Ergün, S., C. Buschmann, J. Heukeshoven, K. Dammann, F. Schnieders, H. Lauke, F. Chalajour, N. Kilic, W. H. Strätling, and G. G. Schumann. 2004. Cell type-specific expression of LINE-1 open reading frames 1 and 2 in fetal and adult human tissues. J. Biol. Chem. 279:27753-27763.[Abstract/Free Full Text]
    8. 8
  8. Fütterer, J., Z. Kiss-László, and T. Hohn. 1993. Nonlinear ribosome migration on cauliflower mosaic virus 35S RNA. Cell 73:789-802.[CrossRef][Medline]
    9. 9
  9. Han, J. S., S. T. Szak, and J. D. Boeke. 2004. Transcriptional disruption by the L1 retrotransposon and implications for mammalian transcriptomes. Nature 429:268-274.[CrossRef][Medline]
    10. 10
  10. Hinnebusch, A. G. 1997. Translational regulation of yeast GCN4. A window on factors that control initiator-tRNA binding to the ribosome. J. Biol. Chem. 272:21661-21664.[Free Full Text]
    11. 11
  11. Horvath, C. M., M. A. Williams, and R. A. Lamb. 1990. Eukaryotic coupled translation of tandem cistrons: identification of the influenza B virus BM2 polypeptide. EMBO J. 9:2639-2647.[Medline]
    12. 12
  12. Ilves, H., O. Kahre, and M. Speek. 1992. Translation of the rat LINE bicistronic RNAs in vitro involves ribosomal reinitiation instead of frameshifting. Mol. Cell. Biol. 12:4242-4248.[Abstract/Free Full Text]
    13. 13
  13. Inokuchi, Y., A. Hirashima, Y. Sekine, L. Janosi, and A. Kaji. 2000. Role of ribosome recycling factor (RRF) in translational coupling. EMBO J. 19:3788-3798.[CrossRef][Medline]
    14. 14
  14. Jacks, T., M. D. Power, F. R. Masiarz, P. A. Luciw, P. J. Barr, and H. E. Varmus. 1988. Characterization of ribosomal frameshifting in HIV-1 gag-pol expression. Nature 331:280-283.[CrossRef][Medline]
    15. 15
  15. Kawakami, K., S. Pande, B. Faiola, D. P. Moore, J. D. Boeke, P. J. Farabaugh, J. N. Strathern, Y. Nakamura, and D. J. Garfinkel. 1993. A rare tRNA-Arg(CCU) that regulates Ty1 element ribosomal frameshifting is essential for Ty1 retrotransposition in Saccharomyces cerevisiae. Genetics 135:309-320.[Abstract]
    16. 16
  16. Kim, Y. G., S. Maas, and A. Rich. 2001. Comparative mutational analysis of cis-acting RNA signals for translational frameshifting in HIV-1 and HTLV-2. Nucleic Acids Res. 29:1125-1131.[Abstract/Free Full Text]
    17. 17
  17. Kojima, K. K., and H. Fujiwara. 2003. Evolution of target specificity in R1 clade non-LTR retrotransposons. Mol. Biol. Evol. 20:351-361.[Abstract/Free Full Text]
    18. 18
  18. Kozak, M. 1987. Effects of intercistronic length on the efficiency of reinitiation by eucaryotic ribosomes. Mol. Cell. Biol. 7:3438-3445.[Abstract/Free Full Text]
    19. 19
  19. Kozak, M. 1990. Downstream secondary structure facilitates recognition of initiator codons by eukaryotic ribosomes. Proc. Natl. Acad. Sci. USA 87:8301-8305.[Abstract/Free Full Text]
    20. 20
  20. Kozak, M. 1999. Initiation of translation in prokaryotes and eukaryotes. Gene 234:187-208.[CrossRef][Medline]
    21. 21
  21. Kozak, M. 2001. Constraints on reinitiation of translation in mammals. Nucleic Acids Res. 29:5226-5232.[Abstract/Free Full Text]
    22. 22
  22. Kubo, Y., S. Okazaki, T. Anzai, and H. Fujiwara. 2001. Structural and phylogenetic analysis of TRAS, telomeric repeat-specific non-LTR retrotransposon families in Lepidopteran insects. Mol. Biol. Evol. 18:848-857.[Abstract/Free Full Text]
    23. 23
  23. Luukkonen, B. G., W. Tan, and S. Schwartz. 1995. Efficiency of reinitiation of translation on human immunodeficiency virus type 1 mRNAs is determined by the length of the upstream open reading frame and by intercistronic distance. J. Virol. 69:4086-4094.[Abstract]
    24. 24
  24. Malik, H. S., W. D. Burke, and T. H. Eickbush. 1999. The age and evolution of non-LTR retrotransposable elements. Mol. Biol. Evol. 16:793-805.[Abstract]
    25. 25
  25. Malik, H. S., and T. H. Eickbush. 2001. Phylogenetic analysis of ribonuclease H domains suggests a late, chimeric origin of LTR retrotransposable elements and retroviruses. Genome Res. 11:1187-1197.[Abstract/Free Full Text]
    26. 26
  26. Matsumoto, T., H. Takahashi, and H. Fujiwara. 2003. 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. Mol. Cell. Biol. 24:105-122.
    27. 27
  27. McMillan, J. P., and M. F. Singer. 1993. Translation of the human LINE-1 element, L1Hs. Proc. Natl. Acad. Sci. USA 90:11533-11537.[Abstract/Free Full Text]
    28. 28
  28. Meyers, G. 2003. Translation of the minor capsid protein of a calicivirus is initiated by a novel termination-dependent reinitiation mechanism. J. Biol. Chem. 278:34051-34060.[Abstract/Free Full Text]
    29. 29
  29. Osanai, M., H. Takahashi, K. K. Kojima, M. Hamada, and H. Fujiwara. 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. 24:7902-7913.[Abstract/Free Full Text]
    30. 30
  30. Peabody, D. S. 1989. Translation initiation at non-AUG triplets in mammalian cells. J. Biol. Chem. 264:5031-5035.[Abstract/Free Full Text]
    31. 31
  31. Peabody, D. S., and P. Berg. 1986. Termination-reinitiation occurs in the translation of mammalian cell mRNAs. Mol. Cell. Biol. 6:2695-2703.[Abstract/Free Full Text]
    32. 32
  32. Pestova, T. V., V. G. Kolupaeva, I. B. Lomakin, E. V. Pilipenko, I. N. Shatsky, V. I. Agol, and C. U. Hellen. 2001. Molecular mechanisms of translation initiation in eukaryotes. Proc. Natl. Acad. Sci. USA 98:7029-7036.[Abstract/Free Full Text]
    33. 33
  33. Sprengel, R., B. Reiss, and H. Schaller. 1985. Translationally coupled initiation of protein synthesis in Bacillus subtilis. Nucleic Acids Res. 13:893-909.[Abstract/Free Full Text]
    34. 34
  34. Szabo, G., Z. Katarova, and R. Greenspan. 1994. Distinct protein forms are produced from alternatively spliced bicistronic glutamic acid decarboxylase mRNAs during development. Mol. Cell. Biol. 14:7535-7545.[Abstract/Free Full Text]
    35. 35
  35. Takahashi, H., and H. Fujiwara. 1999. Transcription analysis of the telomeric repeat-specific retrotransposons TRAS1 and SART1 of the silkworm Bombyx mori. Nucleic Acids Res. 27:2015-2021.[Abstract/Free Full Text]
    36. 36
  36. Takahashi, H., and H. Fujiwara. 2002. Transplantation of target site specificity by swapping the endonuclease domains of two LINEs. EMBO J. 21:408-417.[CrossRef][Medline]
    37. 37
  37. Takahashi, H., S. Okazaki, and H. Fujiwara. 1997. A new family of site-specific retrotransposons, SART1, is inserted into telomeric repeats of the silkworm, Bombyx mori. Nucleic Acids Res. 25:1578-1584.[Abstract/Free Full Text]
    38. 38
  38. Wilson, J. E., M. J. Powell, S. E. Hoover, and P. Sarnow. 2000. Naturally occurring dicistronic cricket paralysis virus RNA is regulated by two internal ribosome entry sites. Mol. Cell. Biol. 20:4990-4999.[Abstract/Free Full Text]

Molecular and Cellular Biology, September 2005, p. 7675-7686, Vol. 25, No. 17
0270-7306/05/$08.00+0     doi:10.1128/MCB.25.17.7675-7686.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Jang, S. A, Sung, B. H., Cho, J. H., Kim, S. C. (2009). Direct Expression of Antimicrobial Peptides in an Intact Form by a Translationally Coupled Two-Cistron Expression System. Appl. Environ. Microbiol. 75: 3980-3986 [Abstract] [Full Text]  
  • Guo, L.-h., Sun, L., Chiba, S., Araki, H., Suzuki, N. (2009). Coupled termination/reinitiation for translation of the downstream open reading frame B of the prototypic hypovirus CHV1-EP713. Nucleic Acids Res 37: 3645-3659 [Abstract] [Full Text]  
  • Kirilyuk, A., Tolstonog, G. V., Damert, A., Held, U., Hahn, S., Lower, R., Buschmann, C., Horn, A. V., Traub, P., Schumann, G. G. (2008). Functional endogenous LINE-1 retrotransposons are expressed and mobilized in rat chloroleukemia cells. Nucleic Acids Res 36: 648-665 [Abstract] [Full Text]  
  • Poyry, T. A.A., Kaminski, A., Connell, E. J., Fraser, C. S., Jackson, R. J. (2007). The mechanism of an exceptional case of reinitiation after translation of a long ORF reveals why such events do not generally occur in mammalian mRNA translation. Genes Dev. 21: 3149-3162 [Abstract] [Full Text]  
  • 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: 5168-5179 [Abstract] [Full Text]  
  • Li, P. W.-L., Li, J., Timmerman, S. L., Krushel, L. A., Martin, S. L. (2006). The dicistronic RNA from the mouse LINE-1 retrotransposon contains an internal ribosome entry site upstream of each ORF: implications for retrotransposition. Nucleic Acids Res 34: 853-864 [Abstract] [Full Text]  
  • Alisch, R. S., Garcia-Perez, J. L., Muotri, A. R., Gage, F. H., Moran, J. V. (2006). Unconventional translation of mammalian LINE-1 retrotransposons. Genes Dev. 20: 210-224 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kojima, K. K.
Right arrow Articles by Fujiwara, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kojima, K. K.
Right arrow Articles by Fujiwara, H.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»