Regulation of Rous sarcoma virus RNA splicing and stability

An 'Arms Race' between the Nonsense-mediated mRNA Decay Pathway and Viral Infections

The Nonsense-mediated mRNA Decay (NMD) pathway is an RNA quality control pathway conserved among eukaryotic cells. While historically thought to predominantly recognize transcripts with premature termination codons, it is now known that the NMD pathway plays a variety of roles, from homeostatic events to control of viral pathogens. In this review we highlight the reciprocal interactions between the host NMD pathway and viral pathogens, which have shaped both the host antiviral defense and viral pathogenesis.

HTLV-1 Tax plugs and freezes UPF1 helicase leading to nonsense-mediated mRNA decay inhibition

Up-Frameshift Suppressor 1 Homolog (UPF1) is a key factor for nonsense-mediated mRNA decay (NMD), a cellular process that can actively degrade mRNAs. Here, we study NMD inhibition during infection by human T-cell lymphotropic virus type I (HTLV-1) and characterise the influence of the retroviral Tax factor on UPF1 activity. Tax interacts with the central helicase core domain of UPF1 and might plug the RNA channel of UPF1, reducing its affinity for nucleic acids. Furthermore, using a single-molecule approach, we show that the sequential interaction of Tax with a RNA-bound UPF1 freezes UPF1: this latter is less sensitive to the presence of ATP and shows translocation defects, highlighting the importance of this feature for NMD. These mechanistic insights reveal how HTLV-1 hijacks the central component of NMD to ensure expression of its own genome.

UPF1 is a central protein in nonsense-mediated mRNA decay (NMD), but contribution of its RNA processivity to NMD is unclear. Here, the authors show how the retroviral Tax protein interacts with and inhibits UPF1, and demonstrate that UPF1’s translocase activity contributes to NMD.

Evidence that a threshold of serine/arginine-rich (SR) proteins recruits CFIm to promote rous sarcoma virus mRNA 3' end formation

The weak polyadenylation site (PAS) of Rous sarcoma virus (RSV) is activated by the juxtaposition of SR protein binding sites within the spatially separate negative regulator of splicing (NRS) element and the env RNA splicing enhancer (Env enhancer), which are far upstream of the PAS. Juxtaposition occurs by formation of the NRS - 3' ss splicing regulatory complex and is thought to provide a threshold of SR proteins that facilitate long-range stimulation of the PAS. We provide evidence for the threshold model by showing that greater than three synthetic SR protein binding sites are needed to substitute for the Env enhancer, that either the NRS or Env enhancer alone promotes polyadenylation when the distance to the PAS is decreased, and that SR protein binding sites promote binding of the polyadenylation factor cleavage factor I (CFIm) to the weak PAS. These observations may be relevant for cellular PASs.

Viral Evasion and Manipulation of Host RNA Quality Control Pathways

Viruses have evolved diverse strategies to maximize the functional and coding capacities of their genetic material. Individual viral RNAs are often used as substrates for both replication and translation and can contain multiple, sometimes overlapping open reading frames. Further, viral RNAs engage in a wide variety of interactions with both host and viral proteins to modify the activities of important cellular factors and direct their own trafficking, packaging, localization, stability, and translation. However, adaptations increasing the information density of small viral genomes can have unintended consequences. In particular, viral RNAs have developed features that mark them as potential targets of host RNA quality control pathways. This minireview focuses on ways in which viral RNAs run afoul of the cellular mRNA quality control and decay machinery, as well as on strategies developed by viruses to circumvent or exploit cellular mRNA surveillance.

Viral interactions with components of the splicing machinery

Eukaryotic genes are often interrupted by stretches of sequence with no protein coding potential or obvious function. After transcription, these interrupting sequences must be removed to give rise to the mature messenger RNA. This fundamental process is called RNA splicing and is achieved by complicated machinery made of protein and RNA that assembles around the RNA to be edited. Viruses also use RNA splicing to maximize their coding potential and economize on genetic space, and use clever strategies to manipulate the splicing machinery to their advantage. This article gives an overview of the splicing process and provides examples of viral strategies that make use of various components of the splicing system to promote their replicative cycle. Representative virus families have been selected to illustrate the interaction with various regulatory proteins and ribonucleoproteins. The unifying theme is fine regulation through protein-protein and protein-RNA interactions with the spliceosome components and associated factors to promote or prevent spliceosome assembly on given splice sites, in addition to a strong influence from cis-regulatory sequences on viral transcripts. Because there is an intimate coupling of splicing with the processes that direct mRNA biogenesis, a description of how these viruses couple the regulation of splicing with the retention or stability of mRNAs is also included. It seems that a unique balance of suppression and activation of splicing and nuclear export works optimally for each family of viruses.

Polypyrimidine tract binding protein 1 protects mRNAs from recognition by the nonsense-mediated mRNA decay pathway

The nonsense-mediated mRNA decay (NMD) pathway degrades mRNAs containing long 3'UTRs to perform dual roles in mRNA quality control and gene expression regulation. However, expansion of vertebrate 3'UTR functions has required a physical expansion of 3'UTR lengths, complicating the process of detecting nonsense mutations. We show that the polypyrimidine tract binding protein 1 (PTBP1) shields specific retroviral and cellular transcripts from NMD. When bound near a stop codon, PTBP1 blocks the NMD protein UPF1 from binding 3'UTRs. PTBP1 can thus mark specific stop codons as genuine, preserving both the ability of NMD to accurately detect aberrant mRNAs and the capacity of long 3'UTRs to regulate gene expression. Illustrating the wide scope of this mechanism, we use RNA-seq and transcriptome-wide analysis of PTBP1 binding sites to show that many human mRNAs are protected by PTBP1 and that PTBP1 enrichment near stop codons correlates with 3'UTR length and resistance to NMD.

DOI:http://dx.doi.org/10.7554/eLife.11155.001

Genes are used as templates to create molecules of messenger RNA (mRNA) that contain all the information needed to make a protein. This information begins with a 'start site' and ends with a 'stop site.' The regions of the mRNA outside of the start and stop sites are called untranslated regions.

Not all mRNAs are correctly made, and cells combat this problem by detecting and destroying faulty mRNAs before they are translated into protein. One way cells do this is by recognizing and destroying mRNAs that include long untranslated regions, which can indicate that the mRNA might have a stop site too early in its sequence. A key problem with this mechanism, however, is that long untranslated regions also serve important roles in the cell: for example, by determining where and when mRNA molecules are read to make protein. How then do mRNAs with long but important untranslated regions escape detection and degradation?

Ge et al. have now investigated this question using an approach that allows a 'handle' to be attached to particular RNA molecules. This allows the RNA and any proteins bound to it to be purified away from all other RNAs and proteins in the cell, and the proteins can then be identified by a technique called mass spectrometry.

Ge at al. found that mRNAs can recruit a protein called PTBP1 to part of the RNA sequence near the stop site. This prevents an RNA decay protein recognizing and triggering the degradation of the mRNA, even if the mRNA has a long untranslated region.

Thus, PTBP1 plays a crucial role in protecting human RNAs with long untranslated regions from destruction by the nonsense-mediated decay pathway. Some viral RNAs are also able to evade decay, and so Ge et al. hypothesize that the virus stole this method for maintaining its RNAs from host cells. A future goal is to understand whether this system works the same way in all cell types or protects different RNAs in different cells.

DOI:http://dx.doi.org/10.7554/eLife.11155.002

Interplay between the alpharetroviral Gag protein and SR proteins SF2 and SC35 in the nucleus

Retroviruses are positive-sense, single-stranded RNA viruses that reverse transcribe their RNA genomes into double-stranded DNA for integration into the host cell chromosome. The integrated provirus is used as a template for the transcription of viral RNA. The full-length viral RNA can be used for the translation of the Gag and Gag-Pol structural proteins or as the genomic RNA (gRNA) for encapsidation into new virions by the Gag protein. The mechanism by which Gag selectively incorporates unspliced gRNA into virus particles is poorly understood. Although Gag was previously thought to localize exclusively to the cytoplasm and plasma membrane where particles are released, we found that the Gag protein of Rous sarcoma virus, an alpharetrovirus, undergoes transient nuclear trafficking. When the nuclear export signal of RSV Gag is mutated (Gag.L219A), the protein accumulates in discrete subnuclear foci reminiscent of nuclear bodies such as splicing speckles, paraspeckles, and PML bodies. In this report, we observed that RSV Gag.L219A foci appeared to be tethered in the nucleus, partially co-localizing with the splicing speckle components SC35 and SF2. Overexpression of SC35 increased the number of Gag.L219A nucleoplasmic foci, suggesting that SC35 may facilitate the formation of Gag foci. We previously reported that RSV Gag nuclear trafficking is required for efficient gRNA packaging. Together with the data presented herein, our findings raise the intriguing hypothesis that RSV Gag may co-opt splicing factors to localize near transcription sites. Because splicing occurs co-transcriptionally, we speculate that this mechanism could allow Gag to associate with unspliced viral RNA shortly after its transcription initiation in the nucleus, before the viral RNA can be spliced or exported from the nucleus as an mRNA template.

A 38 nt region and its flanking sequences within gag of Friend murine leukemia virus are crucial for splicing at the correct 5' and 3' splice sites

The genome of the Friend murine leukemia virus (Fr-MLV) contains a 5' splice site (5'ss) located at 205 nt and a 3'ss located at 5489 nt. In our previous studies, it was shown that if the HindIII-BglII (879-1904 bp) fragment within gag is deleted from the proA8m1 vector, which carries the entire Fr-MLV sequence, then cryptic splicing of env-mRNA occurs. Here, attempts were made to identify the genomic segment(s) in this region that is/are essential to correct splicing. First, vectors with a serially truncated HindIII-BglII fragment were constructed. The vector, in which a 38 bp fragment (1612-1649 bp) is deleted or reversed in proA8m1, only produced splice variants. It was found that a 38 nt region within gag contains important elements that positively regulate splicing at the correct splice sites. Further analyses of a series of vectors carrying the 38 bp fragment and its flanking sequences showed that a region (1183-1611 nt) upstream of the 38 nt fragment also contains sequences that positively or negatively influence splicing at the correct splice sites. The SphI-NdeI (5140-5400 bp) fragment just upstream of the 3'ss was deleted from vectors that carried the 38 bp fragment and its flanking sequences, which yielded correctly spliced mRNA; interestingly, these deleted vectors showed cryptic splicing. These findings suggest that the 5140-5400 nt region located just upstream of the 3'ss is required for the splicing function of the 38 nt fragment and its flanking sequences.

Regulation of retroviral polyadenylation

Cellular and viral preRNAs are extensively cotranscriptionally modified. These modifications include the processing of the 3' end. Most preRNAs are polyadenylated, which is required for nuclear export, RNA stability, and efficient translation. Integrated retroviral genomes are flanked by 3' and 5' long terminal repeats (LTRs). Both LTRs are identical on the nucleotide level, but 3' processing has to be limited to the 3'LTR. Otherwise, polyadenylation at the 5'LTR would result in prematurely terminated, noncoding viral RNAs. Retroviruses have developed a variety of different mechanisms to restrict polyadenylation to the 3'LTR, although the overall structure of the LTRs is similar among all retroviruses. In general, these mechanisms can be divided into three main groups: (1) activation of polyadenylation only at the 3' end by encoding the essential polyadenylation signal in the unique 3 region; (2) suppression of polyadenylation at the 5'LTR by downstream elements such as the major splice donor; and (3) the usage of weak polyadenylation sites, which results in some premature polyadenylated noncoding RNAs and in read-through transcripts at the 3'LTR. All these mechanisms exhibit intrinsic problems, and retroviruses have evolved additional regulatory elements to promote polyadenylation at the 3'LTR only. In this review, we describe the molecular regulation of retroviral polyadenylation and highlight the different mechanisms used for polyadenylation control.

The structure and function of the rous sarcoma virus RNA stability element

For simple retroviruses, such as the Rous sarcoma virus (RSV), post-transcriptional control elements regulate viral RNA splicing, export, stability, and packaging into virions. These RNA sequences interact with cellular host proteins to regulate and facilitate productive viral infections. One such element, known as the RSV stability element (RSE), is required for maintaining stability of the full-length unspliced RNA. This viral RNA serves as the mRNA for the Gag and Pol proteins and also as the genome packaged in progeny virions. When the RSE is deleted from the viral RNA, the unspliced RNA becomes unstable and is degraded in a Upf1-dependent manner. Current evidence suggests that the RSE inhibits recognition of the viral gag termination codon by the nonsense-mediated mRNA decay (NMD) pathway. We believe that the RSE acts as an insulator to NMD, thereby preventing at least one of the required functional steps that target an mRNA for degradation. Here, we discuss the history of the RSE and the current model of how the RSE is interacting with cellular NMD factors.

Translation of MMTV Gag requires nuclear events involving splicing motifs in addition to the viral Rem protein and RmRE

Background

Retroviral Gag proteins are encoded in introns and, because of this localization, they are subject to the default pathways of pre-mRNA splicing. Retroviruses regulate splicing and translation through a variety of intertwined mechanisms, including 5'- post-transcriptional control elements, 3'- constitutive transport elements, and viral protein RNA interactions that couple unspliced and singly spliced mRNAs to transport machinery. Sequences within the gag gene termed inhibitory or instability sequences also appear to affect viral mRNA stability and translation, and the action of these sequences can be countered by silent mutation or the presence of RNA interaction proteins like HIV-1 Rev. Here, we explored the requirements for mouse mammary tumor virus (MMTV) Gag expression using a combination of in vivo and in vitro expression systems.

Results

We show that MMTV gag alleles are inhibited for translation despite possessing a functional open reading frame (ORF). The block to expression was post-transcriptional and targeted the mRNA but was not a function of mRNA transport or stability. Using bicistronic reporters, we show that inhibition of gag expression imparted a block to both cap-dependent and cap-independent translation onto the mRNA. Direct introduction of in vitro synthesized gag mRNA resulted in translation, implying a nuclear role in inhibition of expression. The inhibition of expression was overcome by intact proviral expression or by flanking gag with splice sites combined with a functional Rem-Rem response element (RmRE) interaction.

Conclusions

Expression of MMTV Gag requires nuclear interactions involving the viral Rem protein, its cognate binding target the RmRE, and surprisingly, both a splice donor and acceptor sequence to achieve appropriate signals for translation of the mRNA in the cytoplasm.

Epigenetic regulation of transcription and splicing of syncytins, fusogenic glycoproteins of retroviral origin

Syncytin-1 and -2, human fusogenic glycoproteins encoded by the env genes of the endogenous retroviral loci ERVWE1 and ERVFRDE1, respectively, contribute to the differentiation of multinucleated syncytiotrophoblast in chorionic villi. In non-trophoblastic cells, however, the expression of syncytins has to be suppressed to avoid potential pathogenic effects. We studied the epigenetic suppression of ERVWE1 and ERVFRDE1 5′-long terminal repeats by DNA methylation and chromatin modifications. Immunoprecipitation of the provirus-associated chromatin revealed the H3K9 trimethylation at transcriptionally inactivated syncytins in HeLa cells. qRT-PCR analysis of non-spliced ERVWE1 and ERVFRDE1 mRNAs and respective env mRNAs detected efficient splicing of endogenously expressed RNAs in trophoblastic but not in non-placental cells. Pointing to the pathogenic potential of aberrantly expressed syncytin-1, we have found deregulation of transcription and splicing of the ERVWE1 in biopsies of testicular seminomas. Finally, ectopic expression experiments suggest the importance of proper chromatin context for the ERVWE1 splicing. Our results thus demonstrate that cell-specific retroviral splicing represents an additional epigenetic level controling the expression of endogenous retroviruses.

Structural and functional analysis of the Rous Sarcoma virus negative regulator of splicing and demonstration of its activation by the 9G8 SR protein

Retroviruses require both spliced and unspliced RNAs for replication. Accumulation of Rous Sarcoma virus (RSV) unspliced RNA depends upon the negative regulator of splicing (NRS). Its 5′-part is considered as an ESE binding SR proteins. Its 3′-part contains a decoy 5′-splice site (ss), which inhibits splicing at the bona fide 5′-ss. Only the 3D structure of a small NRS fragment had been experimentally studied. Here, by chemical and enzymatic probing, we determine the 2D structure of the entire RSV NRS. Structural analysis of other avian NRSs and comparison with all sequenced avian NRSs is in favour of a phylogenetic conservation of the NRS 2D structure. By combination of approaches: (i) in vitro and in cellulo splicing assays, (ii) footprinting assays and (iii) purification and analysis of reconstituted RNP complex, we define a small NRS element retaining splicing inhibitory property. We also demonstrate the capability of the SR protein 9G8 to increase NRS activity in vitro and in cellulo. Altogether these data bring new insights on how NRS fine tune splicing activity.

Structural features in the Rous sarcoma virus RNA stability element are necessary for sensing the correct termination codon

Background

Nonsense-mediated mRNA decay (NMD) is an mRNA quality control mechanism that selectively recognizes and targets for degradation mRNAs containing premature termination codons. Retroviral full-length RNA is presented to the host translation machinery with characteristics rarely observed among host cell mRNAs: a long 3' UTR, retained introns, and multiple open reading frames. As a result, the viral RNA is predicted to be recognized by the host NMD machinery and degraded. In the case of the Rous sarcoma virus (RSV), we identified a stability element (RSE), which resides immediately downstream of the gag termination codon and facilitates NMD evasion.

Results

We defined key RNA features of the RSE through directed mutagenesis of the virus. These data suggest that the minimal RSE is 155 nucleotides (nts) and functions independently of the nucleotide sequence of the stop codon or the first nucleotide following the stop codon. Further data suggested that the 3'UTRs of the RSV pol and src may also function as stability elements.

Conclusions

We propose that these stability elements in RSV may be acting as NMD insulators to mask the preceding stop codon from the NMD machinery.

Adaptive evolution of a tagged chimeric gammaretrovirus: identification of novel cis-acting elements that modulate splicing

Retroviruses are well known for their ability to incorporate envelope (Env) proteins from other retroviral strains and genera, and even from other virus families. This characteristic has been widely exploited for the generation of replication-defective retroviral vectors, including those derived from murine leukemia virus (MLV), bearing heterologous Env proteins. We investigated the possibility of "genetically pseudotyping" replication-competent MLV by replacing the native env gene in a full-length viral genome with that of another gammaretrovirus. Earlier, we developed replication-competent versions of MLV that stably transmit and express transgenes inserted into the 3' untranslated region of the viral genome. In one such tagged MLV expressing green fluorescent protein, we replaced the native env sequence with that of gibbon ape leukemia virus (GALV). Although the GALV Env protein is commonly used to make high-titer pseudotypes of MLV vectors, we found that the env replacement greatly attenuated viral replication. However, extended cultivation of cells exposed to the chimeric virus resulted in selection of mutants exhibiting rapid replication kinetics and different variants arose in different infections. Two of these variants had acquired mutations at or adjacent to the splice acceptor site, and three others had acquired dual mutations within the long terminal repeat. Analysis of the levels of unspliced and spliced viral RNA produced by the parental and adapted viruses showed that the mutations gained by each of these variants functioned to reverse an imbalance in splicing caused by the env gene substitution. Our results reveal the presence of previously unknown cis-acting sequences in MLV that modulate splicing of the viral transcript and demonstrate that tagging of the retroviral genome with an easily assayed transgene can be combined with in vitro evolution as an approach to efficiently generating and screening for replicating mutants of replication-impaired recombinant viruses.

Heterogeneous nuclear ribonucleoprotein H is required for optimal U11 small nuclear ribonucleoprotein binding to a retroviral RNA-processing control element: implications for U12-dependent RNA splicing

An RNA-processing element from Rous sarcoma virus, the negative regulator of splicing (NRS), represses splicing to generate unspliced RNA that serves as mRNA and as genomic RNA for progeny virions and also promotes polyadenylation of the unspliced RNA. Integral to NRS function is the binding of U1 small nuclear ribonucleoprotein (snRNP), but its binding is controlled by U11 snRNP that binds to an overlapping site. U11 snRNP, the U1 counterpart for splicing of U12-dependent introns, binds the NRS remarkably well and requires G-rich elements just downstream of the consensus U11 binding site. We present evidence that heterogeneous nuclear ribonucleoprotein (hnRNP) H binds to the NRS G-rich elements and that hnRNP H is required for optimal U11 binding in vitro. It is further shown that hnRNP H (but not hnRNP F) can promote U11 binding and splicing from the NRS in vivo when tethered to the RNA as an MS2 fusion protein. Interestingly, 17% of the naturally occurring U12-dependent introns have at least two potential hnRNP H binding sites positioned similarly to the NRS. For two such introns from the SCN4A and P120 genes, we show that hnRNP H binds to each in a G-tract-dependent manner, that G-tract mutations strongly reduce splicing of minigene RNA, and that tethered hnRNP H restores splicing to mutant RNA. In support of a role for hnRNP H in both splicing pathways, hnRNP H antibodies co-precipitate U1 and U11 small nuclear ribonucleoproteins. These results indicate that hnRNP H is an auxiliary factor for U11 binding to the NRS and that, more generally, hnRNP H is a splicing factor for a subset of U12-dependent introns that harbor G-rich elements.

Solution structure of the pseudo-5' splice site of a retroviral splicing suppressor

Control of Rous sarcoma virus RNA splicing depends in part on the interaction of U1 and U11 snRNPs with an intronic RNA element called the negative regulator of splicing (NRS). A 23mer RNA hairpin (NRS23) of the NRS directly binds U1 and U11 snRNPs. Mutations that disrupt base-pairing between the loop of NRS23 and U1 snRNA abolish its negative control of splicing. We have determined the solution structure of NRS23 using NOEs, torsion angles, and residual dipolar couplings that were extracted from multidimensional heteronuclear NMR spectra. Our structure showed that the 6-bp stem of NRS23 adopts a nearly A-form duplex conformation. The loop, which consists of 11 residues according to secondary structure probing, was in a closed conformation. U913, the first residue in the loop, was bulged out or dynamic, and loop residues G914-C923, G915-U922, and U916-A921 were base-paired. The remaining UUGU tetraloop sequence did not adopt a stable structure and appears flexible in solution. This tetraloop differs from the well-known classes of tetraloops (GNRA, CUYG, UNCG) in terms of its stability, structure, and function. Deletion of the bulged U913, which is not complementary to U1 snRNA, increased the melting temperature of the RNA hairpin. This hyperstable hairpin exhibited a significant decrease in binding to U1 snRNP. Thus, the structure of the NRS RNA, as well as its sequence, is important for interaction with U1 snRNP and for splicing suppression.

Two regions promote U11 small nuclear ribonucleoprotein particle binding to a retroviral splicing inhibitor element (negative regulator of splicing)

The Rous sarcoma virus (RSV) negative regulator of splicing (NRS) is an RNA element that represses splicing and promotes polyadenylation of viral RNA. The NRS acts as a pseudo 5' splice site (ss), and serine-arginine (SR) proteins, U1snRNP, and U6 small nuclear ribonucleoproteins (snRNPs) are implicated in its function. The NRS also efficiently binds U11 snRNP of the U12-dependent splicing pathway, which is interesting, because U11 binds only poorly to authentic substrates that lack a U12-type 3' splice site. It is of considerable interest to understand how the low abundance U11 snRNP binds the NRS so well. Here we show that U11 can bind the NRS as a mono-snRNP in vitro and that a G-rich element located downstream of the U11 site is required for efficient binding. Mutational analyses indicated that two of four G tracts in this region were important for optimal U11 binding and that the G-rich region did not function indirectly by promoting U1 snRNP binding to an overlapping site. Surprisingly, inactivation of U2 snRNP also decreased U11 binding to the NRS. The NRS harbors a branch point-like/pyrimidine tract sequence (BP/Py) just upstream of the U1/U11 site that is characteristic of 3' splice sites. Deletion of this region decreased U2 and U11 binding, and deletion of the G-rich region also reduced U2 binding. The G element, but not the BP/Py sequence, was also required for U11 binding to the NRS in vivo as assessed by minor class splicing from the NRS to a minor class 3'ss from the P120 gene. These results indicate that efficient U11 binding to the isolated NRS involves at least two elements in addition to the U11 consensus sequence and may have implications for U11 binding to authentic splicing substrates.

Packaging and reverse transcription of snRNAs by retroviruses may generate pseudogenes

Retroviruses specifically package two copies of their RNA genome in each viral particle, along with some small cellular RNAs, including tRNAs and 7S L RNA. We show here that Rous sarcoma virus (RSV) also packages U6 snRNA at approximately one copy per virion. In addition, trace amounts of U1 and U2 snRNAs were detected in purified virus by Northern blotting. U6 snRNA comigrated with the RSV 70S genomic RNA dimer on sucrose gradients. We observed reverse transcription of U6 snRNA in an endogenous reaction in which RSV particles were the source of both reverse transcriptase and RNA substrates. This finding led us to examine mammalian genomic sequences for the presence of snRNA pseudogenes. A survey of the human, mouse, and rat genomes revealed a high number of spliceosomal snRNA pseudogenes. U6 pseudogenes were the most abundant, with approximately 200 copies in each genome. In the human genome, 67% of U6 snRNA pseudogenes, and a significant number of the other snRNA pseudogenes, were associated with LINE, SINE, or retroviral LTR repeat sequences. We propose that the packaging of snRNAs in retroviral particles leads to their reverse transcription in an infected cell and the integration of snRNA/viral recombinants into the host genome.

Position dependence of the Rous sarcoma virus negative regulator of splicing element reflects proximity to a 5' splice site

Rous sarcoma virus (RSV) requires incomplete splicing of its viral transcripts to maintain efficient replication. A splicing inhibitor element, the negative regulator of splicing (NRS), is located near the 5' end of the RNA but the significance of this positioning is not known. In a heterologous intron the NRS functions optimally when positioned close to the authentic 5' splice site. This observation led us to investigate the basis of the position dependence. Four explanations were put forth and stressed the role of three major elements involved in splicing, the 3' splice site, the 5' splice site, and the 5' end cap structure. NRS function was unrelated to its position relative to the 3' splice site or the cap structure and appeared to depend on its position relative to the authentic 5' splice site. We conclude that position dependence may reflect distance constraints necessary for competition of the NRS with the authentic 5' splice site for pairing with the 3' splice sites.

Rous sarcoma virus negative regulator of splicing selectively suppresses SRC mRNA splicing and promotes polyadenylation

Retroviruses require a balance of spliced and unspliced RNA for efficient replication. Here, we examined the effect of mutations in a splicing suppressor sequence called the negative regulator of splicing (NRS), located within the gag gene of Rous sarcoma virus. While the NRS mutant viruses showed only small changes in the levels of spliced env mRNAs, they had significant increases in src mRNA levels and transformed cells more efficiently than wild-type virus. None of these mutations prevented viral replication; however, some of the mutant viruses replicated more slowly than wild-type virus. In addition, increased transcriptional readthrough of the poly(A) site in the 3' LTR was observed with the NRS mutant viruses, suggesting that the wild-type NRS sequence promotes polyadenylation.

cis-Elements involved in expression of unspliced RNA in Moloney murine leukemia virus

Murine leukemia virus (MLV) produces the unspliced RNA and the singly spliced RNA at a proper ratio at a time. To identify cis-elements involved in the production of the unspliced RNA, we examined the level of unspliced RNA in a series of mutants derived from a prototype Moloney MLV mutant gag-encoding G3.6. Our present data indicated that nt 1560-1906 region in the CA-encoding region in gag was the negative cis-element and nt 5119-5355 region, which was immediately upstream of the splice acceptor site, was the positive cis-element for expression of the unspliced RNA. It was found that the former element made expression of the unspliced RNA dependent upon the latter. These two elements were functional as discrete elements and their activities were relatively position-independent.

A cellular protein, hnRNP H, binds to the negative regulator of splicing element from Rous sarcoma virus

Incomplete RNA splicing is a key feature of the retroviral life cycle. This is in contrast to the processing of most cellular pre-mRNAs, which are usually spliced to completion. In Rous sarcoma virus, splicing control is achieved in part through a cis-acting RNA element termed the negative regulator of splicing (NRS). The NRS is functionally divided into two parts termed NRS5' and NRS3', which bind a number of splicing factors. The U1 and U11 small nuclear ribonucleoproteins interact with sequences in NRS3', whereas NRS5' binds several proteins including members of the SR [corrected] family of proteins. Among the proteins that specifically bind NRS5' is a previously unidentified 55-kDa protein (p55). In this report we describe the isolation and identification of p55. The p55 binding site was localized by UV cross-linking to a 31-nucleotide segment, and a protein that binds specifically to it was isolated by RNA affinity selection and identified by mass spectrometry as hnRNP H. Antibodies against hnRNP H immunoprecipitated cross-linked p55 and induced a supershift of a p55-containing complex formed in HeLa nuclear extract. Furthermore, UV cross-linking and electrophoretic mobility shift assays indicated that recombinant hnRNP H specifically interacts with the p55 binding site, confirming that hnRNP H is p55. The possible roles of hnRNP H in NRS function are discussed.

Possible regulatory function of the Saccharomyces cerevisiae Ty1 retrotransposon core protein

The yeast Ty1 retrotransposon encodes proteins and RNA that assemble into virus-like particles (VLPs) as part of the life cycle of the retro-element. The Tya protein, which is equivalent to the retroviral Gag, is the major structural component of these particles. In this work, we demonstrate that Tya proteins fulfil other functions apart from their structural role. We show that Tya interacts in vitro with the Ty1 RNA domain required for RNA packaging, suggesting that this RNA-protein interaction may direct the packaging process. Furthermore, the overexpression of both Tya proteins, i.e. p1, the primary translation product, and p2, the mature form, increases endogenous Ty1 RNA levels in trans without increasing translation significantly. These observations suggest that Tya may exert a regulatory function during transposition. Interestingly, however, only p2, the mature form of Tya, trans-activates transposition of a marked genomic Ty element. This confirms that processing is required for transposition.

The role of overlapping U1 and U11 5' splice site sequences in a negative regulator of splicing

Splicing of Rous sarcoma virus RNA is regulated in part by a cis-acting intronic RNA element called the negative regulator of splicing (NRS). An NRS mutant affecting nt 916-923 disrupts U11 snRNP binding and reduces NRS activity (Gontarek et al., 1993, Genes & Dev 7:1926-1936). However, we observed that a U15' splice site-like sequence, which overlapped the U11 site, was also disrupted by this mutation. To determine whether the U1 or the U11 site was essential for NRS activity, we analyzed twelve additional mutants involving nt 915-926. All mutations that disrupted the potential base pairing between U1 snRNA and the NRS reduced NRS activity, including single point mutations at nt 915, 916, and 919. The point mutation at nt 919 was partially suppressed by a compensatory base change mutation in U1 snRNA. In contrast, a mutation which strengthened the potential base pairing between the U1 site and the NRS increased NRS activity. Surprisingly, mutations that specifically targeted the U115' splice site consensus sequence increased the levels of unspliced RNA, suggesting U11 binding plays an antagonistic role to NRS activity. We propose that U1 snRNP binding to the NRS inhibits splicing and is regulated by U11 snRNP binding to the overlapping sequence. Competition between U1 and U11 snRNPs would result in the appropriate balance of spliced to unspliced RNAs for optimal viral replication. Further, a virus mutated in the U1/U11 region of the NRS was found to have delayed replication.

SR protein and snRNP requirements for assembly of the Rous sarcoma virus negative regulator of splicing complex in vitro

Retroviruses use unspliced RNA as mRNA for expression of virion structural proteins and as genomic RNA; the full-length RNA often constitutes the majority of the viral RNA in an infected cell. Maintenance of this large pool of unspliced RNA is crucial since even a modest increase in splicing efficiency can lead to impaired replication. In Rous sarcoma virus, the negative regulator of splicing (NRS) was identified as a cis element that negatively impacts splicing of viral RNA. Components of the splicing apparatus appear to be involved in splicing inhibition since binding of a number of splicing factors (snRNPs and SR proteins) and assembly of a large complex (NRS-C) in nuclear extracts correlate with NRS-mediated splicing inhibition. In determining the requirements for NRS complex assembly, we show that NRS-C assembly can be reconstituted by addition of total SR proteins to an S100 extract that lacks these factors. Of the purified SR proteins tested, SF2/ASF was functional in NRS-C assembly, whereas SC35 and SRp40 were not. The participation of snRNPs in NRS-C assembly was addressed by selectively depleting individual snRNPs with oligonucleotides and RNase H or by sequestering critical snRNA domains with 2'-O-methyl RNA oligonucleotides. The results indicate that in addition to U11 snRNP, U1 snRNP and SR proteins, but not U2 snRNP, are involved in NRS-C assembly.

Characterization of an RNP complex that assembles on the Rous sarcoma virus negative regulator of splicing element

We have characterized an RNP complex that assembles in nuclear extracts on the negative regulator of splicing (NRS) element from Rous sarcoma virus. While no complex was detected by native gel electrophoresis under conditions that supported spliceosome assembly, gel filtration revealed a specific ATP-independent complex that rapidly assembled on NRS RNA. No complexes were formed on non-specific RNA. Unlike the non-specific H complex, factors required for NRS complex assembly are limiting in nuclear extract. The NRS complex was not detected in reactions containing ATP and pre-formed complexes were dissociated in the presence of ATP. In addition, the assembly process was sensitive to high salt but NRS complexes were salt stable once formed. Assembly of the NRS complex appears functionally significant since mutated NRS RNAs that fail to inhibit splicing in vivo are defective for NRS complex assembly in nuclear extract. The probable relationship of the NRS complex to spliceosomal complexes is discussed.

Localization of human immunodeficiency virus Rev in transfected and virus-infected cells

The rev gene product of human immunodeficiency virus (HIV) is obligatory for viral replication. Rev interacts specifically with a structured RNA sequence within the viral genome termed the REV response element (RRE). Although the importance of Rev for the expression of viral proteins is well documented, its functional mechanism remains unresolved. Previous studies identified Rev in the absence of RRE to be a nuclear protein localized primarily within the nucleoli. To extend our understanding of the role of Rev in viral replication, immunolocalization studies of Rev and other nuclear components were carried out in transfected cells expressing both the Rev protein and RRE-containing mRNA and in cells infected with HIV. In both types of cells, Rev-like immunoreactivity was distributed both in the nucleoplasm and cytoplasm. Within the nucleus, Rev immunoreactivity was not evenly distributed but was present within focal concentrations. In transfected cells that were double labeled for Rev and SC-35, which labels a known component of spliceosomes, the foci of Rev labeling were distinct from the "speckles" labeled by SC-35, although Rev foci and speckles were often juxtaposed. In addition, morphological changes in the three-dimensional network of speckles were observed in both transfected cells expressing both the Rev protein and RRE-containing mRNA and in cells infected with HIV-1 and HIV-2. Our observations are consistent with the proposed dual role of Rev in mRNA transport and splicing.

Identification of a Rev-related protein by analysis of spliced transcripts of the human endogenous retroviruses HTDV/HERV-K

The human endogenous retrovirus family HTDV/HERV-K codes for the viral particles observed in teratocarcinoma cell lines. Two types of proviral genomes exist; these differ in the presence or absence of a stretch of 292 nucleotides. This sequence comprises the amino-terminal part of the env gene, the putative signal peptide, which overlaps in part with the carboxy terminus of the pol gene. Type 2 genomes containing this sequence presumably more closely reflect the structure of the infectious, replication-competent retrovirus ancestors of the HERV-K family than do type 1 genomes that lack the sequence. In human teratocarcinoma cell lines, both variants are expressed. Type 1 genomes, in which pol and env genes are fused, are deficient in splicing. Type 2 transcripts are spliced to subgenomic env mRNA and smaller messages. A doubly spliced transcript encodes a short open reading frame, preliminarily designated cORF (R. Löwer, K. Boller, B. Hasenmeier, C. Korbmacher, N. Mueller-Lantzsch, J. Löwer, and R. Kurth, Proc. Natl. Acad. Sci. USA 90:4480-4484). The genomic organization of cORF resembles that of nonprimate lentivirus rev genes: the first exon comprises nearly the entire signal peptide of env, and the second exon is derived from a different reading frame in the 3' part of the genome. A nucleolar localization signal, which is also a putative RNA binding domain, as well as a sequence with similarities to the Rev effector domain consensus sequence is present in the first exon. Secondary structure analysis reveals similarities to basic helix-loop-helix proteins. cORF is a small protein with an apparent molecular mass of 14 kDa which accumulates in the nucleolus as has been described for Rev proteins.

Alternative splicing of pre-mRNAs encoding the nonstructural proteins of minute virus of mice is facilitated by sequences within the downstream intron

mRNAs R1 and R2 of the parvovirus minute virus of mice encode the two essential viral regulatory proteins NS1 and NS2. Both RNAs are spliced between map units 44 and 46 (nucleotides 2280 and 2399); R2 RNAs are additionally spliced upstream between map units 10 and 39 (nucleotides 514 and 1989), using a nonconsensus donor and poor 3' splice site. The relative accumulation of R1 and R2 is determined by alternative splicing: there is twice the steady-state accumulation of R2 relative to that of R1 throughout viral infection, though they are generated from the same promoter and have indistinguishable stabilities. Here we demonstrate that efficient excision of the large intron to generate R2 is dependent on at least the initial presence, in P4-generated pre-mRNAs, of sequences within the downstream small intron. This effect is orientation dependent and related to the size of the intervening exon. Prior splicing of the small intron is unnecessary. Excision of the large intron is enhanced by changing its donor site to consensus, but only in the presence of the small intron sequences. Excision of the large intron is also enhanced by improving the polypyrimidine tract within its 3' splice site; however, in contrast, this change renders excision of the large intron independent of the downstream small intron. We suggest that sequences within the small intron play a primary role in efficient excision of the upstream large intron, perhaps as the initial entry site(s) for an element(s) of the splicesome, which stabilizes the binding of required factors to the polypyrimidine tract within the 3' splice site of the large intron.

Albumin Rugby Park: a truncated albumin variant caused by a G-->C splice-site mutation in intron 13

Three members of a family were found to be heterozygous for a fast albumin variant (Albumin Rugby Park) that made up only 8% of total serum albumin. Isoelectric focussing indicated an increased negative charge on the C-terminal CNBr peptide and C-terminal sequence analysis of the native protein showed an aberrant sequence of -Ser-Phe. Sequence analysis of PCR-amplified DNA indicated a G-->C mutation at position 1 of the 13th intron and this was confirmed by restriction digestion. The replacement of the obligate GT sequence by CT at the exon/intron boundary prevents splicing of the 13th intron and translation continues for 21 nucleotides until a stop codon is reached. The new protein lacks the 14 amino acids coded for in the 14th exon (GKKLVAASQAALGH), but these are replaced by 7 new residues (LLQFSSF), giving a truncated albumin of 578 residues.

5' avian leukosis virus sequences and osteopetrotic potential

Recombinants of Rous-associated virus-0 and Br21 have been used to localize 5' viral sequences that affect the osteopetrotic potential of avian leukosis viruses. Rous-associated virus-0 is a benign subgroup E virus of endogenous origin that does not cause osteopetrosis. Br21 is a constructed subgroup E virus with high osteopetrotic potential. 5' sequences that affected osteopetrotic potential resided in an 834-bp region near the 5' LTR. Sequence analysis of this region revealed differences between Br21 and RAV-0 in the mRNA leader and codons for MA.

The role of the tnv protein and tnv RNA splicing signals in replication of HIV-1 IIIB isolates

The requirement for tnv, a tat-env-rev fusion protein expressed by the IIIB strain of HIV-1, was tested. The expression of tnv was prevented by altering the 5' splice site that flanks the central coding exon of tnv. Mutants that carry such an altered 5' splice site replicate normally in an established T-cell line and in peripheral blood lymphocytes, demonstrating that tnv has no effect on virus replication. However, two mutants that carry an alteration in the 3' splice site of the same exon are replication defective. The 3' splice site mutations result in significant reduction in the expression of the 16-kDa tat protein and induce the expression of large amounts of a 19-kDa rev-related protein that initiates within the central coding exon of tnv. S1 nuclease analysis reveals that splicing to the central tnv exon occurs with substantially increased efficiency via the use of an alternate 3' splice site six nucleotides 3' from the mutated site. The effect of the 3' splice site mutations on viral protein expression and replication are fully reversed by a second site mutation that eliminates the alternate splice site.

Improved gene expression by a U3-based retroviral vector

To improve the expression of the genes transduced by retroviral vectors, we have constructed a U3-based retroviral vector and evaluated its effect on the expression of an insert from the internal promoter. The unique feature of the vector is that the transduced gene is inserted at the U3 region of the 3' long terminal repeats (LTR). Consequently, in the infected cells the gene is duplicated and transferred to the 5'-LTR. When compared with the conventional retroviral vectors which insert the gene within the retroviral transcriptional unit, the U3-based vectors greatly enhanced the expression of the transduced gene under all three promoters tested, viz. the cytomegalovirus immediately early gene promoter (CMV), the SV40 early gene promoter (SV), and the herpes simplex virus thymidine kinase gene promoter (TK). The SV and TK promoters which were previously shown suppressed by the retroviral promoter in the conventional construction restored their potencies in the U3-based vectors. Our results therefore suggested that the U3-based vectors are more advantageous than the conventional vectors for gene expression.

Characterization of Rous sarcoma virus intronic sequences that negatively regulate splicing

Retroviruses splice only a fraction of their primary RNA transcripts to subgenomic mRNA. The unspliced RNA is transported to the cytoplasm, where it serves as genomic RNA as well as mRNA for the gag and pol genes. Deletion of sequences from the Rous sarcoma virus gag gene, which is part of the intron of the subgenomic mRNAs, was previously observed to result in an increase in the ratio of spliced to unspliced RNA. These sequences, which we termed a negative regulator of splicing (NRS), can be moved to the intron of a heterologous gene resulting in an accumulation of unspliced RNA in the nucleus. We have used such constructs, assayed by transient expression in chicken embryo fibroblasts, to define the minimal sequences necessary to inhibit splicing. Maximal NRS activity was observed with a 300-nt fragment containing RSV nts 707-1006; two noncontiguous domains within this fragment, one of which contains a polypyrimidine tract, were both found to be essential. The NRS element was active exclusively in the sense orientation in two heterologous introns tested and in both avian and mammalian cells. Position dependence was also observed, with highest activity when the NRS was inserted in the intron near the 5' splice site. The NRS element was also active at an exon position 136 nts upstream of the 5' splice site but not at sites further upstream. In addition, it did not affect the splicing of a downstream intron.

Analysis of spliced and unspliced Rous sarcoma virus RNAs early and late after infection of chicken embryo fibroblasts: effect of cell culture conditions

Quantitative determinations of spliced and unspliced viral RNA early after retrovirus infection or after transient transfection with proviral DNA have been difficult because of low intracellular viral RNA levels. In this report we describe conditions for the sensitive and quantitative assay of unspliced viral RNA and spliced mRNAs from avian sarcoma virus-infected and transiently transfected chicken embryo fibroblasts. RNase protection mapping was performed with a tandem 32P-labeled riboprobe consisting of three separate regions complementary to the major 5' splice site and the two major 3' splice sites (env and src). The steady-state levels of viral RNA species were determined as early as 14 hr postinfection; no significant changes were observed in the relative levels of spliced mRNAs or in splice site usage as the infection progressed from 14 to 48 hr. After the cells became morphologically transformed at approximately 72 hr postinfection, the condition of the medium of the transformed cell culture affected the steady-state levels of viral RNAs. Unspliced genomic RNA predominated in the transformed cells at 4.5 to 8 hr after feeding; spliced env mRNA was the major species of RNA at 46 to 103 hr after medium change. In contrast, the steady-state levels of viral RNA in transiently transfected cells or in cells infected with a transformation-defective avian sarcoma virus were not as sensitive to the conditions of medium change. The medium-dependent changes in steady-state levels of viral RNA species may either be caused by a declining transcription rate resulting in the accumulation of the more stable env mRNA relative to the more labile unspliced RNA and src mRNA or to changes in the extent of viral RNA splicing.

A transformation-competent recombinant between v-src and Rous-associated virus RAV-1

The LTR, v-src, LTR provirus, which arose by the reverse transcription and integration of src mRNA in the H-19 hamster tumor, has been successfully rescued by fusion with chicken fibroblasts infected with Rous-associated virus RAV-1. One rescued virus, E6, acquired 1 kilobase of the 5' end of the gag gene structure. Recombination took place in the region of 15-nucleotide homology exactly between v-src exon (position 7054) and gag (position 1417). This recombination resulted in the alteration of src splice acceptor site sequences, but this site is maintained as a functional splice acceptor site. The nucleotide structure of the long terminal repeat of recombinant E6 virus suggests that it arose by the intermolecular jump of reverse transcription from RAV-1 to src mRNA and then the switch of templates between already depicted regions of homology. The second jump of reverse transcription was apparently an intramolecular event. The acquisition of 1 kilobase of the 5' gag by E6 resulted in maintaining the balance of unspliced and spliced E6 RNAs and assured the replication advantage of rescued E6 virus over rescued F6 virus, the genome of which corresponds to that present in ancestral H-19 cells.

Reversion of thermosensitive splicing defect of Moloney murine sarcoma virus ts110 by oversplicing of viral RNA

Moloney murine sarcoma virus ts110 possesses a thermosensitive splicing defect. By continuously growing nonproducer cells at the nonpermissive temperature, a new class of revertant cells, termed 6m3, that had lost the thermosensitive splicing defect was produced, and six distinct clones were selected. These cell clones were transformed at either permissive or restrictive temperatures. Unlike parental 6m2 cells, which contain two virus-specific RNA species of 4.0 and 3.5 kilobases (kb) at temperatures permissive for transformation, the 3.5-kb RNA was the only virus-specific RNA species detected in 6m3 clones. No new v-mos-containing DNA fragment was observed in Southern blot analysis of these cell clones compared with parental 6m2 cells, indicating that the 3.5-kb RNA was a splicing product rather than a direct transcript. Moreover, these cells expressed P85gag-mos but not P58gag at any temperature. The reversion of the phenotype in 6m3 cell clones appears to be the result of a selective loss of the temperature sensitivity of the splicing reaction, without affecting the thermosensitivity of the protein kinase activity. This change also appears to alter the mechanism regulating the efficiency of the genomic RNA-splicing reaction.

A temperature-sensitive phenotype of avian myeloblastosis virus: determinants that influence the production of viral mRNAs

The oncogene v-myb of avian myeloblastosis virus is expressed from an mRNA that arises by splicing of the viral genome. In previous work, we described a mutant strain of avian myeloblastosis virus (tsAMV) that elicits temperature-sensitive transformation and suggested that the mutation affects production of the mRNA for v-myb. We now report that the principal determinant of the biochemical phenotype of tsAMV is a point mutation located in a crucial region of the splice acceptor site for v-myb mRNA. The mutation reduces v-myb mRNA production but could account for the conditional phenotype only in combination with an independent effect of temperature on the splicing of both wild-type and mutant viral RNAs, which we also describe here. Our findings dramatize the manner in which retroviruses normally control the splicing of their RNAs and implicate the sequence of the splice acceptor site in the control.

Regulation by HIV Rev depends upon recognition of splice sites

The ability of the Rev protein of HIV to regulate the cytoplasmic level of unspliced RNA from a beta-globin gene containing the Rev response element was dependent on the integrity of the 5' and 3' splice sites. A beta-globin pre-mRNA containing the Rev response element is not under regulation by Rev but is made Rev responsive by a mutation at either the 5' or 3' splice site. These mutant RNAs accumulated in the nucleus as unspliced precursors owing to recognition by splicing components. Only in the presence of Rev did these unspliced RNAs appear in the cytoplasm. Thus, regulation by Rev probably involves the dissociation of splicing components and pre-mRNA.

The rev gene product of the human immunodeficiency virus affects envelope-specific RNA localization

By in situ hybridization analysis and immunoprecipitations following transfection of COS cells, we show that the Rev protein of the human immunodeficiency virus is necessary for envelope protein expression, which is correlated with the appearance in the cytoplasm of envelope-specific RNA. In the absence of cotransfection with a plasmid expressing Rev, envelope-specific RNA is retained in the nucleus. Several cis-acting sites in the envelope are involved, one of which is between nucleotides 7330 and 7735 and is required for the response to Rev. Other sequences (nucleotides 5797-7330 and 7735-7989) are involved in the apparent retention of the envelope-specific RNA in the nucleus in the absence of Rev and its response element. Because Rev affects the localization of envelope RNA both in the presence and in the absence of the normal splice sites on the RNA, the mechanism of Rev action is independent of splicing.

The HIV-1 rev trans-activator acts through a structured target sequence to activate nuclear export of unspliced viral mRNA

Human immunodeficiency virus type 1 (HIV-1) replication requires the expression of two classes of viral mRNA. The early class of HIV-1 transcripts is fully spliced and encodes viral regulatory gene products. The functional expression of one of these nuclear regulatory proteins, termed Rev (formerly Art or Trs), induces the cytoplasmic expression of the incompletely spliced, late class of HIV-1 mRNAs that encode the viral structural proteins, including Gag and Env. Here, we provide evidence that this induction reflects the export from the cell nucleus to the cytoplasm of a pool of unspliced viral RNA constitutively expressed in the nucleus. The hypothesis that Rev acts on RNA transport, rather than splicing, is further supported by the observation that the cytoplasmic expression of a non-spliceable HIV-1 env gene sequence is also subject to Rev regulation. Here we show that this Rev response requires a specific target sequence which coincides with a complex RNA secondary structure present in the env gene. The response to Rev is fully maintained when this sequence is relocated to other exonic or intronic locations within env but is ablated by inversion. These results indicate that the HIV-1 rev gene product induces HIV-1 structural gene expression by activating the sequence-specific nuclear export of incompletely spliced HIV-1 RNA species.