JBC Biosymposia, Inc.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M601537200 on October 11, 2006

J. Biol. Chem., Vol. 281, Issue 49, 37381-37390, December 8, 2006
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
281/49/37381    most recent
M601537200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kraunus, J.
Right arrow Articles by Baum, C.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kraunus, J.
Right arrow Articles by Baum, C.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Murine Leukemia Virus Regulates Alternative Splicing through Sequences Upstream of the 5· Splice Site*Formula

Janine Kraunus{ddagger}§, Daniela Zychlinski{ddagger}, Tilman Heise§, Melanie Galla{ddagger}, Jens Bohne{ddagger}1, and Christopher Baum{ddagger}||2

From the {ddagger}Department of Experimental Hematology, Hannover Medical School, D-30625 Hannover, Germany, the §Department of General Virology, Heinrich-Pette-Institute D-20251 Hamburg, Germany, the Bone Marrow Transplantation, University Hospital Eppendorf, D-20246 Hamburg, Germany, and the ||Division of Experimental Hematology, Cincinnati Children's Hospital, Cincinnati, Ohio 45229

Received for publication, February 17, 2006 , and in revised form, October 10, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Alternative splicing of the primary transcript plays a key role in retroviral gene expression. In contrast to all known mechanisms that mediate alternative splicing in retroviruses, we found that in murine leukemia virus, distinct elements located upstream of the 5' splice site either inhibited or activated splicing of the genomic RNA. Detailed analysis of the first untranslated exon showed that the primer binding site (PBS) activates splicing, whereas flanking sequences either downstream or upstream of the PBS are inhibitory. This new function of the PBS was independent of its orientation and primer binding but associated with a particular destabilizing role in a proposed secondary structure. On the contrary, all sequences surrounding the PBS that are involved in stem formation of the first exon were found to suppress splicing. Targeted mutations that destabilized the central stem and compensatory mutations of the counter strand clearly validated the concept that murine leukemia virus attenuates its 5' splice site by forming an inhibitory stem-loop in its first exon. Importantly, this mode of splice regulation was conserved in a complete proviral clone. Some of the mutants that increase splicing revealed an opposite effect on translation, implying that the first exon also regulates this process. Together, these findings suggest that sequences upstream of the 5' splice site play an important role in splice regulation of simple retroviruses, directly or indirectly attenuating the efficiency of splicing.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
A characteristic feature of all retroviruses is the process of reverse transcription of the RNA genome into double-stranded DNA. Following integration into the host genome, the proviral DNA functions as one expression unit, which is transcribed by cellular RNA polymerase II, yielding a single polycistronic primary transcript that serves as genomic RNA for progeny virus. Productive infection and formation of new retroviral particles require the well balanced expression of all viral genes. This is accomplished by a combination of alternative splicing (intron retention) and regulated nuclear export of the primary transcript on the RNA processing level and proteolytic cleavage and translational read-through on the post-translational level (reviewed in Refs. 1–4).

The genomic organization of all retroviruses is similar (Fig. 1A). The gag-pol open reading frame (ORF)3 encoding the inner structural proteins (Gag), and the replication enzymes (Pol) is located in the 5' half of the transcript and expressed from the unspliced genomic RNA after nuclear export. The gag-pol ORF in all primary retroviral transcripts is defined as an intron through the presence of a preceding 5' splice site (ss) in the 5'-untranslated region and a functional 3'ss located toward the end of the polymerase ORF (Fig. 1A). To express the glycoproteins (Env), which are encoded in the 3' half of the genomic RNA, the gag-pol ORF is removed by a single splice event for subsequent export of the fully spliced RNA (1, 3, 4). This one-splice event strategy creates the challenge to export intron-containing RNAs, which is typically not supported by the cell and rather results in nuclear retention and degradation of the respective RNA (5, 6). For export of their unspliced RNA, retroviruses make use of constitutive transport elements as exemplified by Mason-Pfizer monkey virus or trans-acting factors as illustrated by HIV (7–9). Murine leukemia virus (MLV), a paradigmatic gammaretrovirus, supports the export of unspliced mRNA by a yet unknown mechanism involving the so-called R region stem-loop (RSL) formed by the cap-proximal 28 bases (10).

As simple retroviruses encode no trans-acting regulators of gene expression, alternative or inefficient splicing must be regulated entirely through cis-acting RNA motifs and cellular co-factors. Such motifs may include non-consensus 3'ss, decoy 5'ss, and splice modulatory sequences such as splicing enhancers and silencers (11, 12, 16, 25, 34, 35, 41).

In MLV, the 5'ss matches to almost 100% the cellular consensus sequence (Fig. 1B) and splices to a 3'ss within the pol reading frame (13) (Fig. 1B). The 3'ss misses two relatively important nucleotides surrounding the AG. The polypyrimidine tract (PPT), a second key feature of all 3'ss, is of suboptimal length (10 nucleotides long when compared with 13 residues in average) but not interrupted by weakening purines as most PPTs of HIV (14). Interestingly, a recent study showed that in the context of MLV-based retroviral vectors, the 3'ss can be replaced by very efficient counterparts derived from the human EF1{alpha} gene (15). Although the infectious titer was reduced by about 1 order of magnitude, unspliced genomic RNA was still formed, arguing for the existence of additional splice inhibitory sequences.


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 1.
Genomic organization, RNA species, and splice sites of MLV. A, MLV proviral genome and mRNA species. The genome is flanked by the long terminal repeats (LTRs, consisting of U3, R, and U5). The ORFs are shown as gray boxes. The transcription start site (cap) and the polyadenylation site (pA) mark the length of the genomic RNA. In addition, the PBS, the packaging signal ({Psi}), and the splice sites (5'ss/3'ss) are shown. Below the mRNA species are depicted: the genomic RNA corresponding to the full-length unspliced primary transcript and the spliced, subgenomic RNA. B, comparison of cellular and MLV consensus (Con.)5' and 3'ss. Dark to light gray marks the exon/intron junction, and vertical lines show matched nucleotides. The length of the PPT is indicated.

 
Recently, we have observed that self-inactivating MLV vectors display complete splicing of the retroviral intron when the promoter was placed in between the primer binding site (PBS) and the 5'ss (16). Using a number of mutants, we here demonstrate that distinct sequence elements upstream of the 5'ss either promote or inhibit splicing. Interestingly, this regulation correlates with the stability of the RNA secondary structure proposed by Mougel et al. (17). These data suggest a novel mechanism of splice regulation in retroviruses, directly or indirectly regulating the activity of the 5'ss through upstream sequences.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Plasmids—The retroviral vector plasmids were derived from pSF91 (18). Mutant vectors lacking the U5 region of the 5' long terminal repeat located 70–145 bp downstream of the cap site (pSF91delU5) or lacking the PBS located 146–163 bp downstream of the cap site (pSF91delPBS) were generated in a three-fragment ligation using PCR-based deletion strategies (primers are provided in the supplemental table). Using pSF91 as a template, two separate PCR products were generated. For the deletion of U5, the forward primer (forward, 5'-CAG ATG GTC CCC AGA TGC-3', position –150 regarding cap) was used in combination with a reverse primer immediately upstream of U5. The 3' primer (reverse, 5'-ACG CTG AAC TTG TGG CCG-3') of the 3' PCR product is located downstream of an NcoI site (+706 regarding cap). This primer was used in combination with a forward primer annealing just downstream of the U5 region. The two products were then digested with XbaI (5' fragment) and NcoI (3' fragment), leaving a blunt end in the middle. The two products were ligated into XbaI/NcoI digested pSF91. An analogous strategy was used for the deletion of the PBS.

All other vector plasmids including deletion, antisense, and other mutants were derived from pSF91 by overlap PCR. Again, pSF91 was used as template. The outer 5' and 3' primers are mentioned above. The inner primers bridge the deletion (in the case of pSF91del33–182: sense primer, 5'-AGT CGC CCG G-GA CCA CCG ACC CCC CCG CCG-3'; antisense primer, 5'-GTC GGT GGT C-CC GGG CGA CTC AGT CAA TCG-3', overlap is underlined, hyphen marks the deletion) or carry the suitable point mutations (in the case of pSF91mlvPBS: sense primer, 5'-CAT TTG GGGGCTCGT CCG GGA TTT GGA GAC CCC TG-3'; antisense 5'-CCA AAT CCCGGACGA GCC CCC AAA TGA AAG ACC CC-3', overlap is underlined, mutations are bold). The overlapping PCR product was digested with XbaI and NcoI and ligated into the pSF91 backbone. For the deletions or mutations named sm1, sm2, sm3, and sm3comp, a different 3' primer (5'-AAT GGG CCA CAA AAC GGG CCC CCG A-3') was used, including an ApaI site. The final overlap PCR product was then cloned via XbaI and ApaI. All used primers (FW, forward; RV, reverse) are listed in the supplemental table, indicating name, sequence, and final vector construct. To transfer the PBS and the deletion 3'PBS to a complete proviral clone, an EcoRI/PstI fragment of pMOVGFP (kind gift from B. Schnierle, Paul-Ehrlich-Institute, Langen, Germany) was subcloned into pBluescript (Stratagene). Within this subclone, the AflII/PstI fragment (–464 to +563, regarding the cap site as +1) was exchanged to the corresponding pSF91 sequence and to that of SF91delPBS and SF91del3'PBS via AflII and PstI, yielding pMOVSFGFP, pMOVSFdelPBSGFP, and pMOVSFdel3'PBSGFP, respectively. Correct deletions or nucleotide replacements were confirmed by sequencing.

Cells, Transfections, and Reporter Assays—293T cells were grown in Dulbecco's modified Eagle's medium (Biochrom, Berlin, Germany) supplemented with 10% fetal calf serum, 2 mM glutamine, and 1 mM sodium pyruvate including antibiotics. 7 x 105 293T cells/well were seeded in a 6-well plate. For transfection, the medium was exchanged, and 25 µM chloroquine (Sigma, Taufkirchen, Germany) was added. Retroviral vector DNA (0.9 µg) was transfected using the calcium phosphate precipitation method (19). Medium was exchanged 6 h after transfection, and the cells were harvested after 48 h. Transfection efficiencies ranging between 60 and 80% and protein expression were assessed by flow cytometry in a FACScalibur (BD Biosciences, Heidelberg, Germany) using CellQuest software (BD Biosciences).

Retroviral vector particles were produced by cotransfection of 0.9 µg of retroviral plasmid pSF91 with expression plasmids for MLV gag-pol (1.5 µg) and ecotropic envelope (0.3 µg) into 7 x 105 Phoenix GP (G. Nolan, Stanford University, Palo Alto, CA) cells. In the case of pSF91artPBS, plasmids coding for artificial tRNAs (tRNA-x2-Lys, tRNA-x2-Pro (20)) were co-transfected. Supernatants containing the viral particles were collected 48 h after transfection, filtered through a 0.22-µm filter, and used to transduce 1 x 105 target cells in serial dilutions for titer determination. Transduction was assisted by adding 4 µg/ml protamine sulfate and centrifugation for 60 min at 400 x g and 25–32 °C. Cells were grown for another 2 days before the percentage of enhanced green fluorescent protein (eGFP)-positive cells was determined by flow cytometry. Further analysis was limited to those experiments where less than 30% of target cells were productively transduced.

RNA Preparation and Northern Blot—For preparation of nuclear and cytoplasmic RNA, 8 x 106 cells were collected 48 h after transfection and treated according to the protocol of Weil et al. (21). Briefly, the cells were resuspended in 500 µl of Nonidet P-40 lysis buffer (0.5% Nonidet P-40, 0.14 M NaCl, 10 mM Tris, pH 8.4, 1.5 mM MgCl2, 10 mM EDTA, pH 8.0) for 5 min at 0 °C. After centrifugation at 470 x g for 5 min at 4 °C, the supernatant containing the cytoplasmic fraction was harvested. The nuclear pellet was washed twice with lysis buffer. RNA was extracted from total cells or nuclear and cytoplasmic fractions using the RNA Instapure reagent according to the manufacturer's protocol (Eurogentec, Brussels, Belgium).

For Northern blot, 5–10 µg of RNA were separated at 2 V/cm in 1% agarose gels after denaturing RNA samples with glyoxal (6%) and Me2SO (50%). Subsequently, RNAs were transferred to Biodyne B membrane (0.45 µm, Pall) by capillary transfer and UV cross-linked (Stratalinker, Stratagene). Specific probes used for hybridization corresponded to the cDNAs of eGFP, GAPDH, cytochrome c oxidase II, and the env fragment of MLV. Probes (25 ng) were radiolabeled using the Prime-It II kit (Stratagene, Amsterdam, The Netherlands) to an activity of at least 5 x 108 cpm/µg and separated from unincorporated nucleotides on spin columns (Molecular probes, Göttingen, Germany). DNA template used for the in vitro transcription to generate antisense RNA probes specific for the GAPDH intron B was raised by PCR on genomic DNA using reverse primer 5'-GGA CTA GTT AAT ACG ACT CAC TAT AGG GTG CGG TGG AGA TCT G-3' containing the T7 RNA polymerase promoter sequence (shown in bold) and forward primer 5'-CAA GGA GAG CTC AAG GTC-3'. Transcription reactions were carried out with 0.5 µl of PCR product in a final volume of 20 µl in transcription buffer (Promega, Mannheim, Germany) containing 0.31 mM ATP, CTP, and GTP, 0.25 µM UTP, 5.0 µM [{alpha}-32P]UTP (800 Ci/mmol; Hartmann Analytic, Braunschweig, Germany), 5 mM dithiothreitol, 20 units of RNasin (Promega), and 20 units of T7 RNA polymerase (Promega). The reaction was terminated by adding 10 µg of yeast tRNA and 1 unit of DNaseI (Promega) and incubated for 15 min at 37 °C. Unincorporated nucleotides were removed as above. Hybridization solutions had a final activity of 106 cpm/ml. Membranes were washed, sealed, and exposed to x-ray films (Kodak X-omat-AR, Kodak, Stuttgart, Germany) or quantified by PhosphorImager analysis (Fuji, Düsseldorf, Germany; Amersham Biosciences, Freiburg, Germany) analysis.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
The MLV-derived Vector SF91 Shows Balanced Splicing— The MLV-based vector SF91 (18) (Fig. 2A) was used to investigate the potential splice regulatory role of sequences located upstream of the 5'ss, namely R, U5, PBS, and a short region downstream of PBS (here referred to as 3'PBS). The SF91 vector is derived from MLV by deleting the gag-pol ORF and introducing the env 3'ss including its PPT and branch point sequences 3' of the packaging signal ({Psi}) followed by the original gag ATG, which allows translation of a reporter gene such as eGFP. Importantly, the vector SF91 contains a 460-bp spliceable intron in the 5'-untranslated region, which uses the same 5'ss and 3'ss as the proviral gag-pol intron. For a detailed analysis of splicing, vector plasmids were transiently transfected into human 293T cells. These were harvested 48 h after transfection to prepare nuclear and cytoplasmic RNA. 5 or 10 µgof each RNA fraction were analyzed by Northern blotting.

The vector SF91 showed balanced splicing with a predominance of the unspliced RNA in the nucleus (Fig. 2B, lane 2, left panel), whereas the cytoplasmic compartment revealed an accumulation of spliced message, most likely caused by more efficient export of processed RNA (Fig. 2B, lane 2, right panel). Due to this imbalance in the export kinetics of spliced and unspliced RNAs, we investigated both RNA fractions, assuming that mutants with inhibited splicing would be underrepresented in total RNA and in the cytoplasmic fraction. To control the RNA fractionation procedure, two probes were devised. The first recognized intron B of GAPDH (Fig. 2B, Intron), whose signal should be confined to the nuclear fraction. Using the intron probe, we detected two specific bands at 6.8 kb and about 13 kb in the nucleus, corresponding to the pre-mRNA (NCBI accession number NC_000012 [GenBank] .10) and a longer transcript, which was observed previously (22); these transcripts were much weaker or absent in the cytoplasm (Fig. 2B, right panel). Another band at 5 kb was detectable in both nuclear and cytoplasmic fractions (Fig. 2B). The size matched to the large ribosomal RNA (28 S), and since this was also seen with other probes, we regarded it as unspecific. The second probe was a fragment of cytochrome c oxidase II (CytC), a gene exclusively transcribed in the mitochondria, thus serving as a cytoplasmic marker. Due to incomplete removal of mitochondria from the nucleus, the nuclear fraction was never fully devoid of CytC RNA (Fig. 2B, left panel). In summary, we established a simplified system to study retroviral splice regulation and nuclear export of MLV RNA.


Figure 2
View larger version (52K):
[in this window]
[in a new window]

  		FIGURE 2.
Sequences of the untranslated first exon change the splicing pattern of the MLV-derived vector SF91. A, in SF91 (top), coding sequences from MLV are replaced with eGFP using the original gag ATG site as the translational start site. The viral 3'ss, including branch point and PPT, is cloned upstream of eGFP. Thereby, a shortened intron comprising 460 nucleotides harboring identical splices sites as the MLV provirus is reconstituted. The region of the untranslated first exon (cap site to 5'ss) is enlarged below. The mutant SF91del33–182 (middle) lacks the bases +33 to +182 regarding cap as +1. The mutant SF91delRSL lacks the bases of the RSL that has been shown to be involved in accumulation of unspliced RNA in the cytoplasm (10). LTR, long terminal repeat. B, Northern blots from fractionated lysates of transfected 293T cells, prepared as described under "Experimental Procedures." The blots were hybridized with a radiolabeled eGFP-specific probe (upper panel). Rehybridization with a GAPDH-specific probe served as loading control (second panel). The efficiency of fractionation was checked with a probe specific for an intron of GAPDH (Intron) as a nuclear marker (third panel) and with a cytochrome c oxidase II-specific probe (CytC) as cytoplasmic marker (lower panel). The unspliced and spliced RNAs are identified on the right. Molecular mass standards (M) in kb are shown on the left. Mock, mock-transfected.

 
Distinct Elements in the Untranslated First Exon Regulate Splicing and RNA Export—Next we focused on the region upstream of the 5'ss and its role in splice regulation. Following earlier experiments (16), we deleted bases +33 to +182 (with the cap site being +1) upstream of the 5'ss, resulting in the vector SF91del33–182 (Fig. 2A). When analyzed in parallel with SF91, SF91del33–182 showed an increased splicing efficiency in the nuclear fraction (Fig. 2B, lane 3, left panel). Accordingly, the cytoplasm contained almost exclusively spliced RNA, whereas unspliced RNA could only be detected after prolonged exposure (data not shown). This experiment revealed that the deleted sequences, located upstream of the 5'ss, play a major role in retroviral RNA processing. This rather large deletion also seems to lower the RNA level (Fig. 2B, lanes 2 and 3, left panel). In contrast, smaller deletions that also increased the efficiency of splicing did not affect RNA levels (see below).

To examine the role of the RSL, previously shown to promote the appearance of unspliced RNA in the cytoplasm (10), we constructed another vector lacking the bases 4–31 of the R region (SF91delRSL, Fig. 2A) and thus the entire RSL. When compared with SF91, deletion of the RSL enhanced nuclear splicing by 2.5-fold (as determined by PhosphorImager analysis). In addition, deleting the RSL inhibited the accumulation of unspliced RNA in the cytoplasm (Fig. 2B, lanes 2 and 4). The ratio of spliced/unspliced RNA in the cytoplasm was 2.3 for SF91 and changed to 8.6 for SF91delRSL (3.8-fold enhancement). These data confirm that the RSL is involved in RNA export (10); moreover, our data suggest that the RSL participates in balancing the retroviral splice reaction.

Splicing Is Blocked in Vectors Lacking the PBS—To study the role of the remaining sequences in the untranslated first exon in more detail, we deleted parts of this region separately. The vectors as shown in Fig. 3A were SF91delR2 (R deletion 3' of the RSL), SF91delU5 (deletion of U5), SF91delPBS (deletion of the PBS), and SF91del3'PBS (deletion of bases +164 to +182, immediately downstream of the PBS). Following transient transfection of 293T cells and fractionation of nuclear and cytoplasmic RNA, Northern blots were performed as described above. Deletions of R2, U5, or 3'PBS produced similar amounts of total RNA but strongly enhanced the accumulation of spliced message already in the nucleus (Fig. 3B, lanes 3, 4, and 7, left panel). The overloaded lanes 4 and 7 revealed that unspliced RNA was still detectable with these deletion mutants. In contrast, splicing was nearly blocked when deleting the PBS (Fig. 3B, lane 6, left panel). Another deletion mutant with the combined deletion of PBS sequences and downstream sequences +164 to +182 (SF91delPBS-3'PBS, Fig. 3A) demonstrated a similar phenotype as SF91del3'PBS, in which only the bases +164 to +182 were deleted (Fig. 3B, compare lanes 5 and 7). Therefore, the latter mutation was dominant over the deletion of the PBS.

For all these mutants except the one with the deleted PBS, we detected only spliced RNA in the cytoplasmic fraction (Fig. 3B, right panel, compare lane 6 with lanes 3, 4, 5, and 7). This reflected the strength of the nuclear splicing reaction. As overexposure of the blot showed unspliced RNA in these cases (data not shown), a strong effect of the deleted sequences on RNA export was unlikely.

The correctness of the transcripts was proven by RT-PCR and subsequent sequence analysis (data not shown). The additional band observed in the nucleus was refractory to sequence analysis due to PCR-intramolecular hybridization. However, scoring the first untranslated exon in a splice site prediction program (available from the Berkeley Drosophila Genome Project) revealed a cryptic 5'ss at 262 nucleotides (3' to the major 5'ss, consistent with the observed band). Interestingly, splicing at this site was not inhibited by deletion of the PBS (Fig. 3B, lane 6, left panel). In summary, these data revealed an exceptional role of the PBS being the only element within the first untranslated exon that promotes retroviral splicing from the 5'ss, whereas the surrounding sequences inhibit this process.


Figure 3
View larger version (36K):
[in this window]
[in a new window]

  		FIGURE 3.
The splice enhancing PBS is surrounded by splice inhibitory elements. A, a schematic illustration of the first untranslated exon (cap to 5'ss) of SF91 (top) and the deletion mutants. Sequences are deleted as indicated: SF91delR2 (+33-+69 deleted); SF91delU5 (+70-+145 deleted); SF91delPBS-3'PBS (+146-+182 deleted); SF91delPBS (+146-+163 deleted); SF91del3'PBS (+164-+182 deleted). In SF91mlvPBS and SF91mlvPBSdel3'PBS, the PBS sequence (annealing to tRNAGln in SF91) is mutated in the PBS sequence original used in MLV for tRNAPro. On the right, splicing efficiency as determined in B is shown. B, Northern blot analysis of RNA from transfected 293T cells. Using the eGFP-specific probe (upper panel), the blots represent the ratio of unspliced and spliced RNA in the nuclear (left) and cytoplasmic (right) fraction. GAPDH-, intron-, and CytC-specific probes served as controls for loading and fractionation. Molecular mass standards (M) in kb are shown on the left. Mock, mock-transfected. C, translation efficiency of the untranslated region mutants as indicated by mean fluorescence intensity (MFI) of eGFP determined in unsorted cell pools comprising 10,000 single events. Median and standard deviation of 3–9 independent experiments are given. rel., relative.

 
Deletion of U5 and 3'PBS Impairs Translational Utilization—Taking advantage of eGFP encoded by the vectors, we also determined the influence of the 5'-untranslated region deletions on the protein level in transfected 293T cells. Although the deletion of U5 and 3'PBS produced almost exclusively spliced RNA in the cytoplasm, the mean fluorescence intensity was markedly decreased (Fig. 3C). Both U5 and 3'PBS deletions reduced the mean fluorescence intensity by 80% when compared with the parental vector SF91. Thus, in these deletion mutants, the amount of spliced RNA in the cytoplasm did not correlate with translational utilization (compare Fig. 3B, right panel with Fig. 3C). However, deleting the PBS reduced GFP expression by 40%, which correlated with the reduced amount of spliced RNA. We extended these observations to 293T-based packaging cells, which stably produce MLV Gag-Pol. The presence of Gag did not alter the translational regulation (data not shown). Moreover, identical data were obtained in transfected HeLa cells (data not shown). These data suggest that the sequences involved in splice regulation also contribute to translational control of the retroviral transcript.

The Splice-promoting Effect of the PBS Is Independent of Its Genotype and Primer Binding—To find out whether the role of the PBS in splice regulation is dependent on the type of the corresponding tRNA and is therefore dependent on the PBS sequence or whether primer binding itself regulates splicing, we developed another set of PBS mutants. MLV typically contains a PBS with a specificity for the proline tRNA. Our vectors contain a PBS with a specificity for the glutamine tRNA as this PBS does not inhibit transcription in primitive embryonic and hematopoietic stem cells (23, 24). Exchanging the PBS for tRNAGln used in SF91 to the PBS for tRNAPro usually found in MLV (five point mutations within the 18 bases of the PBS) yielded SF91mlvPBS (Fig. 3A). In the transient transfection assay, SF91 and SF91mlvPBS showed similar splicing patterns, as shown in Fig. 3B (lanes 2 and 8, left and right panel). The independence of the kind of PBS and corresponding tRNA in splice regulation was also confirmed by a mutant containing the MLV-PBS followed by a deletion of +164 to +182 (SF91mlvPBSdel3'PBS, Fig. 3A). When compared with its SF91 counterpart (SF91del3'PBS), a similar splice alteration was observed in both cases (Fig. 3B, compare lanes 7 and 9, left and right panel). These data showed that the dominant splice inhibitory effect of the region 3' of the PBS is independent of the type of the neighboring PBS and that the experimental system provided highly reproducible results.

Annealing of the tRNA to the PBS takes place during virion assembly (25). However, it is still possible that tRNA precursors, which are present in the nucleus (26), might anneal to the PBS of the pre-mRNA. To rule out any effects of primer tRNA binding in splice regulation, we designed vectors containing an artificial PBS (SF91artPBS), which does not bind any cellular tRNA, adapting the approach developed by Pedersen and colleagues (20), as described in Ref. 27. With this method, it is possible to examine splicing in the absence or presence of a bound tRNA primer. We found that the potential annealing of a matching tRNA did not alter splicing efficiency, strongly suggesting that binding of a tRNA primer is not important for splice regulation (data not shown).

Evidence for a Structural Role of the PBS in Splice Enhancement—The previous experiments imply that the role of the PBS in balanced splicing is not dependent on tRNA specificity or primer binding. To analyze whether the role of the PBS in splice regulation is largely sequence-independent, we developed a mutant that carries the PBS in antisense orientation (SF91asPBS, Fig. 4A). Indeed, this construct showed an identical splice phenotype as the wild type SF91 (Fig. 4B, compare lanes 3 and 5). This was confirmed by quantification of the RNA ratios (Fig. 5). Thus, the phenotype of the PBS deletion, namely a block in splicing, was reverted when introducing the PBS in antisense (Fig. 4B, lanes 4 and 5, and Fig. 5).

In great contrast to the deletion of the PBS, deleting the 19 bases downstream of the PBS (mutant SF91del3'PBS) strongly enhanced splicing (Fig. 3B, compare lanes 6 and 7; confirmed in Fig. 4B, lanes 4 and 6, and quantified in Fig. 5). To investigate whether the role of the sequence downstream of the PBS in splice regulation is dependent on its sequence, we cloned a mutant that contained this region in antisense orientation (SF91as3'PBS; Fig. 4A). In this case, the deletion and antisense orientation produced similar phenotypes, i.e. increased splicing (Fig. 4B, lanes 6 and 7; quantified in Fig. 5). Identical results were obtained when transfecting murine SC-1 fibroblasts, revealing that the observed effects are not species-specific (data not shown).

We conclude that the role of the PBS in splice regulation is largely independent of its sequence. Instead, our data suggest that the PBS modulates splicing by forming a spacer within a larger structural framework.


Figure 4
View larger version (42K):
[in this window]
[in a new window]

  		FIGURE 4.
Role of the PBS and downstream regions in RNA processing. A, a drawing of the first untranslated exon of SF91. Black arrows (PBS) and white arrows (3'PBS) indicate the orientation of the two elements. SF91asPBS (middle) carries the PBS sequence in antisense orientation. The other mutant (SF91as3'PBS, bottom) carries the 19 bases downstream of the PBS in antisense orientation. The right column shows the splicing efficiency as determined in B. B, Northern blot data were obtained as described before. eGFP-transcript ratios (upper panel) of the wild type vector SF91 were directly compared with the transcripts of the PBS deletion (SF91delPBS) and the antisense (SF91asPBS) mutant. Additionally, the deletion mutant of the sequence 3' of the PBS (SF91del3'PBS) was compared with its antisense counterpart SF91as3'PBS. The last lane shows the combined deletion of both regions PBS and 3'PBS. Control probes (lower panels) were used as indicated. Molecular mass standards (M) in kb are shown on the left. Mock, mock-transfected.

 
The Splice Regulation Mediated by the PBS and Adjacent Sequences Also Applies to a Complete MLV Proviral Clone—To test whether the above results apply to an entire proviral genome, we transferred the two deletion mutants with the most prominent phenotypes, namely the PBS deletion (inhibits splicing) and the 3'PBS deletion (increases splicing), into a replication-competent MLV construct called MOVGFP (28). MOVGFP contains the entire MLV proviral genome with the eGFP gene inserted into the proline-rich region of env. We cloned the different leader regions also containing a part of the SF91 U3 (SF) into the proviral plasmid MOVGFP yielding a MOVSFGFP and the respective deletion mutants, MOVSF-delPBSGFP and MOVSFdel3'PBSGFP. Based on analysis of total RNA, introducing the SF91 leader into MOVGFP slightly increased splicing of the env RNA (Fig. 6, compare lanes 4 and 3). Deleting the PBS led to accumulation of unspliced RNA, as predicted by the above studies (Fig. 6, lane 5). In contrast, deleting the region 3' of the PBS severely reduced the amount of genomic RNA and generated much higher levels of spliced env RNA (Fig. 6, lane 6). As an additional control, an unmodified wild type MLV was used (Fig. 6, lane 2). Thus, the data obtained in the context of the subgenomic vector system (Figs. 2, 3, 4, 5) were predictive for RNA processing of a complete provirus. The data also show that the distance to the splice acceptor is not a critical parameter of the splice inhibitory effects of the sequences located upstream of the 5'ss.


Figure 5
View larger version (25K):
[in this window]
[in a new window]

  		FIGURE 5.
Quantification of the splicing ratios. Northern blots containing constructs from Figs. 2, 3, 4 were quantified using PhosphorImager technology. The relative amounts of spliced and unspliced RNA are shown. Values represent the median of 3–5 independent experiments.

 
To test whether the env splice acceptor is required in this context, we designed vectors lacking this sequence, the preceding PPT, and the branch point; these constructs still allow cryptic splicing to a minor splice acceptor located in the packaging signal. In this context (data not shown) as well as in the vectors containing the authentic env splice acceptor (Fig. 4) or the proviral context shown above (Fig. 6), deleting sequences 3' of the PBS enhanced splicing. Together, these data reveal that the splice regulation mediated by sequences upstream of the 5'ss is independent of the size of the intron and the type of the splice acceptor. This leaves the 5'ss as the most likely interaction partner of the splice regulatory sequences located in the first untranslated exon.

Detailed Mapping of the Sequence Downstream of the PBS Validates a Structural Model for Splice Regulation—To analyze whether splicing in MLV depends on sequence or structure of the first untranslated exon, we focused on the validated secondary structure (17) and introduced further mutations into the region 3' to the PBS (Fig. 7A). First, the deletion 3'PBS was divided into smaller deletions named sm1 and sm2 (stem mutants 1 and 2). We then analyzed total RNA by Northern blot. The parental construct SF91 was compared with SF91delPBS and SF91del3'PBS (Fig. 7B, lanes 2, 3, and 4, upper panel), reproducing the effects on splicing from previous analyses performed with fractionated RNA. The deletion sm1, which compromises the first half of del3'PBS, shifted the balance toward the spliced RNA (Fig. 7B, compare lanes 4 and 5, upper panel). The deletion of the second half of del3'PBS (sm2) displayed the same phenotype as the complete del3'PBS (Fig. 7B, compare lanes 4 and 6, upper panel). Since a potential binding motif for hnRNPA1, a splice repressor, coincides with sm2 (AGGGA), we destroyed it by mutation to ACCGA (sm3; Fig. 7A, boxed motif). This mutant again showed a similar phenotype as sm2 and del3'PBS (Fig. 7B, lanes 7, 6, and 4), suggesting that potential hnRNPA1 binding at this site is not involved in MLV splice regulation. To ultimately test whether the sequence motif or the secondary structure is important, we introduced compensatory mutations into the corresponding bases on the ascending side of the stem (sm3comp, Fig. 7A, boxed motif). Importantly, these mutations restored the wild type splicing phenotype (Fig. 7B, compare lanes 8 and 2). A lower RNA amount for sm3 was only observed in this particular experiment and not in others (Fig. 7B, two extra lanes). PhosphorImager analysis confirmed that the splice inhibitory effect mainly resides in the lower part of the stem (del3'PBS, sm2 and sm3) and that the compensation of sm3 restores the wild type balance (Fig. 7B, lower panel). By RT-PCR and sequencing, all mutants were shown to splice to the correct 3'ss (data not shown). These findings strongly suggest that alternative splicing in MLV is regulated via a secondary structure upstream of the 5'ss.


Figure 6
View larger version (55K):
[in this window]
[in a new window]

  		FIGURE 6.
Sequences upstream of the 5'ss regulate splicing in a complete MLV proviral clone. Northern blot analysis of 293 T cells transfected with an MLV wild type plasmid (MLV wt) and a variant carrying the eGFP gene in the proline-rich region (MOVGFP; 28). The SF91 leader (SF) and two deletion mutants, mainly delPBS and del3'PBS, were cloned into MOVGFP, resulting in MOVSFGFP, MOVSFdelPBSGFP, and MOVSFdel3'PBSGFP. 10 µg of total RNA were separated on a denaturing agarose gel. MLVs specific RNAs were detected with a probe corresponding to the env ORF. Rehybridization with a GAPDH-specific probe served as loading control (bottom panel). Molecular mass standards in kb are given on the left, and the RNA species are identified on the right. Mock, mock-transfected.

 


Figure 7
View larger version (34K):
[in this window]
[in a new window]

  		FIGURE 7.
The secondary structure upstream of the 5'ss regulates alternative splicing in MLV. A, a schematic illustration of the secondary structure of the MLV first untranslated exon according to the structure by Mougel et al. (17). The RSL (first 32 bases), the U5 region, the PBS, the region downstream of the PBS (3'PBS), and the 5'ss are indicated. The large gray arrow points to the invariant GU dinucleotide at the 5'ss also being the first two nucleotides of the intron. Downstream of the 5'ss follows the packaging signal ({Psi}). Key sequences are shown in detail; other stems are represented as thick black lines, and loops are represented as thick black half-open circles. For the RSL, the PBS, the consensus 5'ss, and the mutated nucleotides in sm3/sm3comp, the important nucleotides are emphasized by gray circles. B, upper panel, Northern blot of total RNA from the indicated constructs. RNA species are indicated on the right, and molecular size markers (M) are noted in kb on the left. A GAPDH-specific probe was used as a loading control. The two extra lanes represent a different experiment for the two most important mutations. The lower panel shows the quantification of the experiment depicted in the upper panel by PhosphorImager analysis. Spliced and unspliced RNA are indicated as in Fig. 5. Mock, mock-transfected.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The present study investigated the impact of sequences located upstream of the 5'ss of MLV, a paradigmatic gamma-retrovirus, on retroviral splice regulation. Our data reveal that sequences surrounding and including the PBS are crucial determinants of alternative splicing. These sequences show a higher degree of conservation with other gammaretroviruses (such as gibbon ape leukemia virus, 84% identity) when compared with other regions (such as the active site of reverse transcriptase, 74% identity), whereas the packaging signal shows no homology. We have shown that the PBS plays an exceptional role in MLV splice regulation by counteracting adjacent strong splice inhibitory structures, without the need for primer binding. This points to a previously unknown mechanism of splice regulation in simple retroviruses, possibly also relevant for eukaryotic cellular genes. This mechanism involves sequences located upstream of the 5'ss to regulate alternative splicing, without the need for a virus-encoded trans-acting factor.

Interestingly, deleting the putative export element of MLV (RSL) slightly enhanced splicing. However, in the case of SF91, the export efficiency of the unspliced transcript is rather low so that most of the unspliced RNA stays in the nucleus. Therefore, it is not very likely that complete splicing of other mutants can be attributed to a prolonged nuclear retention. In addition, all of these mutants still contain the putative export element.

Regulation of alternative splicing is an essential step in the life cycle of all retroviruses and thus subject to tight evolutionary control. HIV, as a complex retrovirus, and RSV, as an example for a simple retrovirus, are the best studied viruses in this respect. In HIV, all 5'ss match the cellular consensus sequence (29), in contrast to the 3'ss, which are weakened by several means, including short and interrupted PPTs (14). Besides, the exons and introns of HIV contain a well balanced assembly of splicing silencers and enhancers (30–32). RSV explores a different mechanism in addition to weak 3'ss for env and src (32). The gag gene contains a sequence known as negative regulator of splicing, which acts as a decoy 5'ss to generate a non-productive spliceosome, thereby reducing the efficiency of the actual upstream 5'ss (33).

The mechanisms of splice regulation in the otherwise well investigated MLV have not been subject to detailed studies (13, 34). Several lines of evidence led us to hypothesize that MLV negatively regulates the 5'ss instead of the 3'ss, in contrast to many other retroviruses. Although the MLV 3'ss does not fully match the consensus (Fig. 1B), the preceding PPT is close to the cellular average in length and, more importantly, not interrupted by attenuating purines. Such a relatively strong PPT can substitute for either a weak branch point or poor sequence conservation of the actual 3'ss (35). Moreover, introducing a very efficient cellular 3'ss into MLV vectors did not fully prevent the formation of genomic RNA (15). Finally, when constructing a new generation of retroviral self-inactivating vectors, we placed the internal promoter 18 bp downstream of the PBS and thus 24 bp upstream of the 5'ss. In this configuration, we observed complete splicing of the retroviral intron, strongly suggesting that sequences located upstream of the promoter insertion site (comprising the second half of R, U5, and 18 bp 3' of the PBS) negatively regulate gamma-retroviral splicing (16).

These observations and the opposite, splice-promoting effect of the PBS in MLV splice regulation can be explained by the secondary structure of the first untranslated exon. Mougel et al. (17) determined the structure of the leader region of MLV by chemical probing. Their stem-loop model is schematically shown in Fig. 7A. The 5'ss is located at the bottom of the stem, where the first G residue of the intron is still paired (Fig. 7A). Within the stem structure, the PBS is the only sequence element that loops out. This is likely a prerequisite for its main function, binding of the tRNA in the producer cell (17), and thus a consequence of a strong evolutionary pressure. Interestingly, deleting the PBS leads to an even more stable structure as determined by the MFOLD program (36), whereas deletion of U5 and sequences 3' of the PBS destabilize the structure. Thus, the stability of this region correlates with the degree of 5'ss attenuation. In addition, fine mapping of the 3'PBS region revealed that in particular, the lower part of the stem is important to form a splice inhibitory structure (Fig. 7). Compensatory mutations strongly argue for the correctness of the structural model.

The stability may well affect the frequency of stem-loop formation. A similar phenomenon has also been reported for alternative splicing of the fibroblast growth factor receptor mRNA, where two regulating sequences are juxtaposed by formation of a stem (37, 38). If the stability of the upstream region is important for negative regulation of the 5'ss, one could envision that recognition of the 5'ss by U1snRNP would be impaired, thus leading to balanced splicing. Examples for this kind of splice regulation have been found in two cellular genes: adenosine deaminase (39) and tau (Ref. 40 and reviewed in Ref. 41). In these two cases, the complete 5'ss is part of the stem, whereas we favor a model for MLV of an inhibitory secondary structure formed by sequences located upstream of the 5'ss. This structure may determine the accessibility for cellular splice regulators, whose identity and recognition motifs remain to be determined. Interestingly, a secondary structure is also implicated in the function of the negative regulator of splicing of RSV (42). In this case, a single nucleotide deletion leads to an increase in stability of the stem-loop structure in analogy to the PBS deletion. As a result, the mutated negative regulator of splicing displays a reduced binding of U1snRNP, and the authors conclude that for balanced splicing, a moderately destabilized structure is necessary (42).

In summary, the complex structure of the first untranslated exon of MLV is important for several steps in the retroviral life cycle and represents a compromise of several evolutionary needs. The RSL is required for the export of the unspliced RNA, in line with previous results (10). The U5, in particular the stem structure upstream of the PBS and the pairing with the region 3' of the PBS, is important for the initiation of reverse transcription (43). In addition, both deletions affect the translational utilization of the respective RNAs (Fig. 3C), reminiscent of the role of the R/U5 region of spleen necrosis virus in translational regulation (44, 45). The main finding of the current study is a novel function for the sequences downstream of the RSL, and especially for the PBS, in retroviral splice regulation.


   FOOTNOTES
 
* The work was supported by Deutsche Forschungsgemeinschaft Grant 1837/Ba4 and the European Union (INHERINET and CONSERT grants). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Formula The on-line version of this article (available at http://www.jbc.org) contains a supplemental table. Back

1 To whom correspondence may be addressed: Dept. of Experimental Hematology, Hannover Medical School, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany. Tel.: 49-511-532-6067; Fax: 49-511-532-6068; E-mail: bohne.jens{at}mh-hannover.de. 2 To whom correspondence may be addressed: Dept. of Experimental Hematology, Hannover Medical School, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany. Tel.: 49-511-532-6067; Fax: 49-511-532-6068; E-mail: baum.christopher{at}mh-hannover.de.

3 The abbreviations used are: ORF, open reading frame; PBS, primer binding site; MLV, murine leukemia virus; ss, splice site(s); HIV, human immunodeficiency virus; RSL, R region stem-loop; PPT, polypyrimidine tract; GFP, green fluorescent protein; eGFP, enhanced GFP; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; CytC, cytochrome c oxidase II; pol, polymerase. Back


   ACKNOWLEDGMENTS
 
We thank B. Schnierle for the plasmid MOVGFP and C. Stocking, K. Harbers, H. Will, and W. Ostertag for support of J. K.'s thesis at the Heinrich-Pette-Institute.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Coffin, J. M., Hughes, S. H., and Varmus, H. E. (1997) Retroviruses (Coffin, J. M., ed) Cold Spring Harbor Laboratory, Cold Spring Harbor, NY
  2. Frankel, A. D., and Young, J. A. (1998) Annu. Rev. Biochem. 67, 1–25[CrossRef][Medline] [Order article via Infotrieve]
  3. Pollard, V. W., and Malim, M. H. (1998) Annu. Rev. Microbiol. 52, 491–532[CrossRef][Medline] [Order article via Infotrieve]
  4. Wodrich, H., and Krausslich, H. G. (2001) Results Probl. Cell Differ. 34, 197–217[Medline] [Order article via Infotrieve]
  5. Chang, D. D., and Sharp, P. A. (1989) Cell 59, 789–795[CrossRef][Medline] [Order article via Infotrieve]
  6. Legrain, P., and Rosbash, M. (1989) Cell 57, 573–583[CrossRef][Medline] [Order article via Infotrieve]
  7. Hammarskjold, M. L. (2001) Curr. Top. Microbiol. Immunol. 259, 77–93[Medline] [Order article via Infotrieve]
  8. Bray, M., Prasad, S., Dubay, J. W., Hunter, E., Jeang, K. T., Rekosh, D., and Hammarskjold, M. L. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 1256–1260[Abstract/Free Full Text]
  9. Cullen, B. R. (2003) Trends Biochem. Sci. 28, 419–424[CrossRef][Medline] [Order article via Infotrieve]
  10. Trubetskoy, A. M., Okenquist, S. A., and Lenz, J. (1999) J. Virol. 73, 3477–3483[Abstract/Free Full Text]
  11. Bouck, J., Fu, X. D., Skalka, A. M., and Katz, R. A. (1995) Mol. Cell. Biol. 15, 2663–2671[Abstract]
  12. Zhang, L., and Stoltzfus, C. M. (1995) Virology 206, 1099–1107[CrossRef][Medline] [Order article via Infotrieve]
  13. Lazo, P. A., Prasad, V., and Tsichlis, P. N. (1987) J. Virol. 61, 2038–2041[Abstract/Free Full Text]
  14. Purcell, D. F., and Martin, M. A. (1993) J. Virol. 67, 6365–6378[Abstract/Free Full Text]
  15. Lee, J. T., Yu, S. S., Han, E., and Kim, S. (2004) Gene. Ther. 11, 94–99[CrossRef][Medline] [Order article via Infotrieve]
  16. Kraunus, J., Schaumann, D. H., Meyer, J., Modlich, U., Fehse, B., Brandenburg, G., Von Laer, D., Klump, H., Schambach, A., Bohne, J., and Baum, C. (2004) Gene. Ther. 11, 1568–1578[CrossRef][Medline] [Order article via Infotrieve]
  17. Mougel, M., Tounekti, N., Darlix, J. L., Paoletti, J., Ehresmann, B., and Ehresmann, C. (1993) Nucleic Acids Res. 21, 4677–4684[Abstract/Free Full Text]
  18. Hildinger, M., Abel, K. L., Ostertag, W., and Baum, C. (1999) J. Virol. 73, 4083–4089[Abstract/Free Full Text]
  19. Chen, C., and Okayama, H. (1987) Mol. Cell. Biol. 7, 2745–2752[Abstract/Free Full Text]
  20. Lund, A. H., Duch, M., Lovmand, J., Jorgensen, P., and Pedersen, F. S. (1997) J. Virol. 71, 1191–1195[Abstract]
  21. Weil, D., Boutain, S., Audibert, A., and Dautry, F. (2000) RNA (Cold Spring Harbor) 6, 962–975
  22. Dani, C., Piechaczyk, M., Audigier, Y., El Sabouty, S., Cathala, G., Marty, L., Fort, P., Blanchard, J. M., and Jeanteur, P. (1984) Eur. J. Biochem. 145, 299–304[Medline] [Order article via Infotrieve]
  23. Grez, M., Zornig, M., Nowock, J., and Ziegler, M. (1991) J. Virol. 65, 4691–4698[Abstract/Free Full Text]
  24. Logan, A. C., Haas, D. L., Kafri, T., and Kohn, D. B. (2004) J. Virol. 78, 8421–8436[Abstract/Free Full Text]
  25. Marquet, R., Isel, C., Ehresmann, C., and Ehresmann, B. (1995) Biochimie (Paris) 77, 113–124
  26. Iborra, F. J., Jackson, D. A., and Cook, P. R. (2001) Science 293, 1139–1142[Abstract/Free Full Text]
  27. Galla, M., Will, E., Kraunus, J., Chen, L., and Baum, C. (2004) Mol. Cell 16, 309–315[CrossRef][Medline] [Order article via Infotrieve]
  28. Erlwein, O., Buchholz, C. J., and Schnierle, B. S. (2003) J. Gen. Virol. 84, 369–373[Abstract/Free Full Text]
  29. O'Reilly, M. M., McNally, M. T., and Beemon, K. L. (1995) Virology 213, 373–385[CrossRef][Medline] [Order article via Infotrieve]
  30. Amendt, B. A., Si, Z. H., and Stoltzfus, C. M. (1995) Mol. Cell. Biol. 15, 4606–4615[Abstract]
  31. Bilodeau, P. S., Domsic, J. K., Mayeda, A., Krainer, A. R., and Stoltzfus, C. M. (2001) J. Virol. 75, 8487–8497[Abstract/Free Full Text]
  32. Katz, R. A., Skalka, A. M., and Kotler, M. (1990) Mol. Cell. Biol. 10, 696–704[Abstract/Free Full Text]
  33. Paca, R. E., Hibbert, C. S., O'Sullivan, C. T., and Beemon, K. L. (2001) J. Virol. 75, 7763–7768[Abstract/Free Full Text]
  34. Hoshi, S., Odawara, T., Oshima, M., Kitamura, Y., Takizawa, H., and Yoshikura, H. (2002) Biochem. Biophys. Res. Commun. 290, 1139–1144[CrossRef][Medline] [Order article via Infotrieve]
  35. Roscigno, R. F., Weiner, M., and Garcia-Blanco, M. A. (1993) J. Biol. Chem. 268, 11222–11229[Abstract/Free Full Text]
  36. Zuker, M. (1989) Methods Enzymol. 180, 262–288[Medline] [Order article via Infotrieve]
  37. Muh, S. J., Hovhannisyan, R. H., and Carstens, R. P. (2002) J. Biol. Chem. 277, 50143–50154[Abstract/Free Full Text]
  38. Baraniak, A. P., Lasda, E. L., Wagner, E. J., and Garcia-Blanco, M. A. (2003) Mol. Cell. Biol. 23, 9327–9337[Abstract/Free Full Text]
  39. Bratt, E., and Ohman, M. (2003) RNA (Cold Spring Harbor) 9, 309–318
  40. Grover, A., Houlden, H., Baker, M., Adamson, J., Lewis, J., Prihar, G., Pickering-Brown, S., Duff, K., and Hutton, M. (1999) J. Biol. Chem. 274, 15134–15143[Abstract/Free Full Text]
  41. Buratti, E., and Baralle, F. E. (2004) Mol. Cell. Biol. 24, 10505–10514[Free Full Text]
  42. Cabello-Villegas, J., Giles, K. E., Soto, A. M., Yu, P., Mougin, A., Beemon, K. L., and Wang, Y. X. (2004) RNA (Cold Spring Harbor) 10, 1388–1398
  43. Miller, J. T., Ge, Z., Morris, S., Das, K., and Leis, J. (1997) J. Virol. 71, 7648–7656[Abstract]
  44. Roberts, T. M., and Boris-Lawrie, K. (2000) J. Virol. 74, 8111–8118[Abstract/Free Full Text]
  45. Roberts, T. M., and Boris-Lawrie, K. (2003) J. Virol. 77, 11973–11984[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Virol.Home page
M. Galla, A. Schambach, G. J. Towers, and C. Baum
Cellular Restriction of Retrovirus Particle-Mediated mRNA Transfer
J. Virol., March 15, 2008; 82(6): 3069 - 3077.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
281/49/37381    most recent
M601537200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
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 arrow reprints & permissions
Citing Articles