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

Abstract

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. 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. 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. 1Coffin J.M. Hughes S.H. Varmus H.E. Coffin J.M. Retroviruses. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1997Google Scholar, 4Wodrich H. Krausslich H.G. Results Probl. Cell Differ. 2001; 34: 197-217Crossref PubMed Scopus (27) Google Scholar). The genomic organization of all retroviruses is similar (Fig. 1A). The gag-pol open reading frame (ORF) 3The 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. 3The 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. 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 (1Coffin J.M. Hughes S.H. Varmus H.E. Coffin J.M. Retroviruses. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1997Google Scholar, 3Pollard V.W. Malim M.H. Annu. Rev. Microbiol. 1998; 52: 491-532Crossref PubMed Scopus (578) Google Scholar, 4Wodrich H. Krausslich H.G. Results Probl. Cell Differ. 2001; 34: 197-217Crossref PubMed Scopus (27) Google Scholar). 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 (5Chang D.D. Sharp P.A. Cell. 1989; 59: 789-795Abstract Full Text PDF PubMed Scopus (390) Google Scholar, 6Legrain P. Rosbash M. Cell. 1989; 57: 573-583Abstract Full Text PDF PubMed Scopus (321) Google Scholar). 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 (7Hammarskjold M.L. Curr. Top. Microbiol. Immunol. 2001; 259: 77-93PubMed Google Scholar, 9Cullen B.R. Trends Biochem. Sci. 2003; 28: 419-424Abstract Full Text Full Text PDF PubMed Scopus (244) Google Scholar). 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 (10Trubetskoy A.M. Okenquist S.A. Lenz J. J. Virol. 1999; 73: 3477-3483Crossref PubMed Google Scholar). 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 (11Bouck J. Fu X.D. Skalka A.M. Katz R.A. Mol. Cell. Biol. 1995; 15: 2663-2671Crossref PubMed Scopus (25) Google Scholar, 12Zhang L. Stoltzfus C.M. Virology. 1995; 206: 1099-1107Crossref PubMed Scopus (42) Google Scholar, 16Kraunus J. Schaumann D.H. Meyer J. Modlich U. Fehse B. Brandenburg G. Von Laer D. Klump H. Schambach A. Bohne J. Baum C. Gene. Ther. 2004; 11: 1568-1578Crossref PubMed Scopus (74) Google Scholar, 25Marquet R. Isel C. Ehresmann C. Ehresmann B. Biochimie (Paris). 1995; 77: 113-124Crossref PubMed Scopus (196) Google Scholar, 34Hoshi S. Odawara T. Oshima M. Kitamura Y. Takizawa H. Yoshikura H. Biochem. Biophys. Res. Commun. 2002; 290: 1139-1144Crossref PubMed Scopus (6) Google Scholar, 35Roscigno R.F. Weiner M. Garcia-Blanco M.A. J. Biol. Chem. 1993; 268: 11222-11229Abstract Full Text PDF PubMed Google Scholar, 41Buratti E. Baralle F.E. Mol. Cell. Biol. 2004; 24: 10505-10514Crossref PubMed Scopus (311) Google Scholar). 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 (13Lazo P.A. Prasad V. Tsichlis P.N. J. Virol. 1987; 61: 2038-2041Crossref PubMed Google Scholar) (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 (14Purcell D.F. Martin M.A. J. Virol. 1993; 67: 6365-6378Crossref PubMed Google Scholar). a showed that in the of retroviral the 3′ss be by from the human gene E. S. Gene. Ther. 2004; 11: PubMed Scopus Google Scholar). the was by of unspliced genomic RNA was for the of splice inhibitory we that complete splicing of the retroviral intron when the was in the primer binding site (PBS) and the 5′ss J. Schaumann D.H. Meyer J. Modlich U. Fehse B. Brandenburg G. Von Laer D. Klump H. Schambach A. Bohne J. Baum C. Gene. Ther. 2004; 11: 1568-1578Crossref PubMed Scopus (74) Google Scholar). a of we that distinct sequence elements upstream of the 5′ss either or splicing. this regulation with the of the RNA secondary proposed by M. J. Ehresmann B. Ehresmann C. Res. 1993; PubMed Scopus Google Scholar). suggest a mechanism of splice regulation in retroviruses, directly or indirectly the of the 5′ss through upstream retroviral were from M. Baum C. J. Virol. 1999; 73: PubMed Google Scholar). the region of the 5′ long located downstream of the site or the PBS located downstream of the site were in a are in the as a two were For the of the primer was used in combination with a reverse primer upstream of The 3′ primer of the 3′ is located downstream of an site This primer was used in combination with a primer downstream of the The two were with and a end in the The two were into strategy was used for the of the and mutants were from by was used as The 5′ and 3′ are The inner the the of is the or the mutations the of is mutations are The was with and and into the For the or mutations and a 3′ primer was an The was and used are in the and To the PBS and the to a complete proviral an of from B. was into this the to the site as was to the sequence and to that of and and yielding and or were by and were in with and were in a For the was and was DNA was the C. H. Mol. Cell. Biol. 1987; PubMed Scopus Google Scholar). was after and the were after and and expression were by in a particles were by of of retroviral with expression for gag-pol and into In the of for M. J. P. J. Virol. PubMed Google were the viral particles were after through a and used to in for was by and for and were for the of enhanced green fluorescent was by analysis was to of were RNA and of nuclear and RNA, were after and to the of D. S. A. RNA Spring Scholar). the were in of for for the the was The nuclear was with RNA was from or nuclear and the RNA to the For of RNA were in after RNA with and were to by and used for to the of eGFP, GAPDH, cytochrome c oxidase II, and the of were the The to an of and from nucleotides on DNA used for the in transcription to RNA for the intron was by on genomic DNA reverse primer the RNA polymerase sequence in and primer were with of in a of in transcription and of and of RNA polymerase The was by of and of and for nucleotides were removed as a of were and to or by analysis The The M. Baum C. J. Virol. 1999; 73: PubMed Google Scholar) (Fig. was used to the splice role of sequences located upstream of the 5′ss, PBS, and a region downstream of PBS to as The is from by the gag-pol ORF and the 3′ss its and sequences 3′ of the by the which of a gene such as Importantly, the a intron in the 5′-untranslated which the 5′ss and 3′ss as the proviral gag-pol For a analysis of splicing, were into human were after to nuclear and RNA. or RNA were by The showed balanced splicing with a of the unspliced RNA in the (Fig. whereas the revealed an of spliced most by export of RNA (Fig. to this in the export of spliced and unspliced RNAs, we RNA that mutants with inhibited splicing be in RNA and in the To the RNA two were The first intron of (Fig. be to the nuclear the intron we two and 13 in the to the and a which was C. M. Y. S. G. L. P. J.M. P. J. Biochem. PubMed Scopus Google these transcripts were or in the (Fig. was in nuclear and (Fig. The to the RNA and this was also with we as The second was a of cytochrome c oxidase a gene transcribed in the as a to of from the the nuclear was fully of RNA (Fig. In we a to retroviral splice regulation and nuclear export of RNA. in the and RNA we on the region upstream of the 5′ss and its role in splice Following J. Schaumann D.H. Meyer J. Modlich U. Fehse B. Brandenburg G. Von Laer D. Klump H. Schambach A. Bohne J. Baum C. Gene. Ther. 2004; 11: 1568-1578Crossref PubMed Scopus (74) Google we bases to the site upstream of the 5′ss, in the (Fig. in with showed an splicing efficiency in the nuclear (Fig. the almost spliced RNA, whereas unspliced RNA be after not This revealed that the located upstream of the 5′ss, play a role in retroviral RNA This rather also to the RNA level (Fig. and In that also the efficiency of splicing not RNA To the role of the RSL, to the of unspliced RNA in the (10Trubetskoy A.M. Okenquist S.A. Lenz J. J. Virol. 1999; 73: 3477-3483Crossref PubMed Google we the bases of the R region and the compared with of the enhanced nuclear splicing by by In the inhibited the of unspliced RNA in the (Fig. and The of RNA in the was for and to for that the is involved in RNA export (10Trubetskoy A.M. Okenquist S.A. Lenz J. J. Virol. 1999; 73: 3477-3483Crossref PubMed Google suggest that the in the retroviral splice in the the role of the sequences in the untranslated first exon in we of this region The as in were 3′ of the of of the and of bases to downstream of the Following of and of nuclear and RNA, were as of or similar of RNA but enhanced the of spliced in the (Fig. and The and revealed that unspliced RNA was with these In splicing was when the PBS (Fig. with the of PBS sequences and downstream sequences to a similar as in which the bases to were (Fig. and the was the of the For all these mutants the one with the PBS, we spliced RNA in the (Fig. with and This the of the nuclear splicing As of the showed unspliced RNA in these not a effect of the sequences on RNA export was The of the transcripts was by and subsequent sequence analysis not The in the was to sequence analysis to the first untranslated exon in a splice site from the revealed a 5′ss nucleotides to the 5′ss, with the splicing this site was not inhibited by of the PBS (Fig. In these revealed an role of the PBS the within the first untranslated exon that retroviral splicing from the 5′ss, whereas the surrounding sequences this process. of and of encoded by the we also the of the 5′-untranslated region on the level in the of and almost spliced RNA in the the was (Fig. and the by when compared with the in these the of spliced RNA in the not with translational with the PBS expression by which with the of spliced RNA. these to which The presence of not the translational regulation not were in not suggest that the sequences involved in splice regulation also to translational of the retroviral The of the PBS of and the role of the PBS in splice regulation is on the of the and is on the PBS sequence or primer binding regulates splicing, we of PBS typically a PBS with a for the a PBS with a for the as this PBS not transcription in and stem M. M. J. M. J. Virol. PubMed Google Scholar, T. J. Virol. 2004; PubMed Scopus Google Scholar). the PBS for used in to the PBS for found in mutations within the bases of the (Fig. In the and showed similar splicing as in and and The of the of PBS and in splice regulation was also by a the by a of to compared with its a similar splice was in (Fig. and and showed that the splice inhibitory effect of the region 3′ of the PBS is independent of the of the PBS and that the of the to the PBS R. Isel C. Ehresmann C. Ehresmann B. Biochimie (Paris). 1995; 77: 113-124Crossref PubMed Scopus (196) Google Scholar). is that which are in the 2001; PubMed Scopus Google to the PBS of the To of primer binding in splice we an PBS which not cellular the by and M. J. P. J. Virol. PubMed Google as in M. E. J. L. Baum C. Mol. Cell. 2004; Full Text Full Text PDF PubMed Scopus Google this is to splicing in the or presence of a found that the of a not splicing that binding of a primer is not important for splice regulation not for a of the PBS in that the role of the PBS in balanced splicing is not on or primer To the role of the PBS in splice regulation is we a that the PBS in orientation this showed an splice as the (Fig. and This was by of the RNA (Fig. the of the PBS a in splicing, was when the PBS in (Fig. and and of the splicing from were The of spliced and unspliced RNA are the of independent In contrast to the of the PBS, the bases downstream of the PBS enhanced splicing (Fig. and in and and in To the role of the sequence downstream of the PBS in splice regulation is on its we a that this region in orientation In this the and orientation similar splicing (Fig. and in results were when murine that the are not not that the role of the PBS in splice regulation is independent of its suggest that the PBS splicing by forming a within a structural The by the PBS and to a the results to an proviral genome, we the two mutants with the most the PBS and the into a J. Virol. 2003; PubMed Scopus Google Scholar). the proviral genome with the gene into the region of the also a of the into the proviral yielding a and the respective and on analysis of RNA, the into splicing of the RNA (Fig. and the PBS to of unspliced RNA, as by the (Fig. In the region 3′ of the PBS the of genomic RNA and of spliced RNA (Fig. As an an was used (Fig. the in the of the were for RNA processing of a complete The also that the to the splice is not a of the splice inhibitory of the sequences located upstream of the To the splice is in this we this the preceding PPT, and the these splicing to a splice located in the In this not as well as in the the splice (Fig. or the proviral (Fig. sequences 3′ of the PBS enhanced splicing. Together, these that the splice regulation by sequences upstream of the 5′ss is independent of the of the intron and the of the splice This the 5′ss as the most of the splice sequences located in the first untranslated exon. Detailed of the of the PBS a for splicing in on sequence or of the first untranslated we on the validated secondary M. J. Ehresmann B. Ehresmann C. Res. 1993; PubMed Scopus Google Scholar) and mutations into the region 3′ to the PBS (Fig. the was into and mutants and RNA by The was compared with and (Fig. and the on splicing from with RNA. The which the first half of the toward the spliced RNA (Fig. and The of the second half of the as the complete (Fig. and a binding for a splice with we by to This showed a similar as and (Fig. and that binding this site is not involved in splice To the sequence or the secondary is we compensatory mutations into the bases on the of the stem Importantly, these mutations the splicing (Fig. and A RNA for was in this particular and not in (Fig. two analysis that the splice inhibitory effect in the of the stem and and that the of the (Fig. and all mutants were to splice to the 3′ss not findings suggest that alternative splicing in is regulated a secondary upstream of the The the of sequences located upstream of the 5′ss of MLV, a paradigmatic on retroviral splice that sequences surrounding and the PBS are of alternative splicing. sequences a of with as leukemia virus, when compared with as the site of reverse whereas the no that the PBS plays an role in splice regulation by splice inhibitory the for primer This to a unknown mechanism of splice regulation in simple retroviruses, also for cellular genes. This mechanism sequences located upstream of the 5′ss to alternative splicing, the for a trans-acting the export of (RSL) enhanced splicing. in the of the export efficiency of the unspliced transcript is rather that most of the unspliced RNA in the is not that complete splicing of mutants be to a nuclear In all of these mutants the export of alternative splicing is an in the of all retroviruses and to HIV, as a and as an for a simple are the in this In HIV, all 5′ss the cellular consensus sequence Virology. 1995; PubMed Scopus Google in contrast to the 3′ss, which are by and interrupted PPTs (14Purcell D.F. Martin M.A. J. Virol. 1993; 67: 6365-6378Crossref PubMed Google Scholar). the and of HIV a well balanced of splicing silencers and enhancers Stoltzfus C.M. Mol. Cell. Biol. 1995; 15: PubMed Scopus Google Scholar, R.A. Skalka A.M. M. Mol. Cell. Biol. PubMed Scopus Google Scholar). a mechanism in to 3′ss for and R.A. Skalka A.M. M. Mol. Cell. Biol. PubMed Scopus Google Scholar). The gene a sequence known as of splicing, which as a decoy 5′ss to a the efficiency of the upstream 5′ss J. Virol. 2001; PubMed Scopus Google Scholar). The mechanisms of splice regulation in the well not to (13Lazo P.A. Prasad V. Tsichlis P.N. J. Virol. 1987; 61: 2038-2041Crossref PubMed Google Scholar, 34Hoshi S. Odawara T. Oshima M. Kitamura Y. Takizawa H. Yoshikura H. Biochem. Biophys. Res. Commun. 2002; 290: 1139-1144Crossref PubMed Scopus (6) Google Scholar). of to that regulates the 5′ss of the 3′ss, in contrast to the 3′ss not fully the consensus (Fig. the preceding is to the cellular in length not interrupted by attenuating Such a relatively for either a or sequence of the 3′ss R.F. Weiner M. Garcia-Blanco M.A. J. Biol. Chem. 1993; 268: 11222-11229Abstract Full Text PDF PubMed Google Scholar). a cellular 3′ss into not fully the formation of genomic RNA E. S. Gene. Ther. 2004; 11: PubMed Scopus Google Scholar). when a new of retroviral we the downstream of the PBS and upstream of the In this we complete splicing of the retroviral that sequences located upstream of the site the second half of and 3′ of the splicing J. Schaumann D.H. Meyer J. Modlich U. Fehse B. Brandenburg G. Von Laer D. Klump H. Schambach A. Bohne J. Baum C. Gene. Ther. 2004; 11: 1568-1578Crossref PubMed Scopus (74) Google Scholar). and the effect of the PBS in splice regulation be by the secondary of the first untranslated exon. M. J. Ehresmann B. Ehresmann C. Res. 1993; PubMed Scopus Google Scholar) the of the region of by stem-loop is in The 5′ss is located the of the the first of the intron is (Fig. the stem the PBS is the sequence that This is a for its binding of the in the cell M. J. Ehresmann B. Ehresmann C. Res. 1993; PubMed Scopus Google and a of a the PBS to an as by the M. 1989; PubMed Scopus Google whereas of and sequences 3′ of the PBS the structure. the of this region with the of 5′ss In of the region revealed that in the of the stem is important to a splice inhibitory (Fig. mutations for the of the structural The may well the of stem-loop A similar also for alternative splicing of the two sequences are by formation of a stem J. Biol. Chem. 2002; Full Text Full Text PDF PubMed Scopus Google Scholar, Garcia-Blanco M.A. Mol. Cell. Biol. 2003; PubMed Scopus Google Scholar). the of the upstream region is important for regulation of the 5′ss, one that of the 5′ss by be to balanced splicing. for this of splice regulation found in two cellular E. M. RNA Spring 2003; Scholar) and A. H. M. J. J. G. S. M. J. Biol. Chem. 1999; Full Text Full Text PDF PubMed Scopus Google and in 41Buratti E. Baralle F.E. Mol. Cell. Biol. 2004; 24: 10505-10514Crossref PubMed Scopus (311) Google Scholar). In these two the complete 5′ss is of the whereas we a for of an inhibitory secondary formed by sequences located upstream of the This may the for cellular splice and motifs to be a secondary is also in the function of the of splicing of J. A.M. P. A. RNA Spring 2004; Scholar). In this a single to an increase in of the stem-loop in to the PBS As a the of splicing a binding of and the that for balanced splicing, a destabilized is J. A.M. P. A. RNA Spring 2004; Scholar). In the of the first untranslated exon of is important for in the retroviral and a of The is for the export of the unspliced RNA, in with results (10Trubetskoy A.M. Okenquist S.A. Lenz J. J. Virol. 1999; 73: 3477-3483Crossref PubMed Google Scholar). The in particular the stem upstream of the PBS and the with the region 3′ of the PBS, is important for the of reverse transcription S. J. J. Virol. PubMed Google Scholar). In the translational of the respective (Fig. of the role of the region of virus in translational regulation J. Virol. PubMed Scopus Google Scholar, J. Virol. 2003; 77: PubMed Scopus Google Scholar). The of the is a function for the sequences downstream of the RSL, and for the PBS, in retroviral splice B. for the and C. H. and for of J. the with