A naturally arising mutation of a potential silencer of exon splicing in human immunodeficiency virus type 1 induces dominant aberrant splicing and arrests virus production.

We have isolated a naturally arising human immunodeficiency type 1 (HIV-1) mutant containing a point mutation within the env gene. The point mutation resulted in complete loss of balanced splicing, with dominant production of aberrant mRNAs. The aberrant RNAs arose via activation of normally cryptic splice sites flanking the mutation within the env terminal exon to create exon 6D, which was subsequently incorporated in aberrant env, tat, rev, and nef mRNAs. Aberrant multiply spliced messages contributed to reduced virus replication as a result of a reduction in wild-type Rev protein. The point mutation within exon 6D activated exon 6D inclusion when the exon and its flanking splice sites were transferred to a heterologous minigene. Introduction of the point mutation into an otherwise wild-type HIV-1 proviral clone resulted in virus that was severely inhibited for replication in T cells and displayed elevated usage of exon 6D. Exon 6D contains a bipartite element similar to that seen in tat exon 3 of HIV-1, consisting of a potential exon splicing silencer (ESS) juxtaposed to a purine-rich sequence similar to known exon splicing enhancers. In the absence of a flanking 5' splice site, the point mutation within the exon 6D ESS-like element strongly activated env splicing, suggesting that the putative ESS plays a natural role in limiting the level of env splicing. We propose, therefore, that exon silencers may be a common element in the HIV-1 genome used to create balanced splicing of multiple products from a single precursor RNA.     

A tripartite HIV-1 tat-env-rev fusion protein.

A 26 kd protein reactive with antiserum to the transactivator tat of the Human Immunodeficiency Virus Type 1 (HIV-1) has been detected in virus producing cells. The 26 kd protein is shown to be a tripartite fusion protein including coding sequences of the tat, envelope (env) and regulator of virion expression (rev) genes. Fusion of these coding sequences occurs by use of a previously undescribed exon within env. This 26 kd protein, designated tnv, has tat but no rev activity detectable with the assay used. The existence of other less abundant tat and rev related proteins in HIV-1 producing cells is also noted.     

Human cell lines stably expressing HIV env and tat gene products

A DNA fragment containing the tat, rev and env genes of the human immunodeficiency virus type 1 was inserted into the retroviral vector pZIPneoAU3. The resulting plasmid penvAU3 was transfected into HeLa and psi CRIP cells. Resulting recombinant retroviruses were used to infect HeLa and Jurkat cells. Immunoprecipitation analysis of stable transformants showed the expression of HIV env glycoproteins gp160, gp120 and gp41. Transactivation assays with a plasmid containing the gene for chloramphenicol acetyltransferase linked to HIV promoter-enhancer sequences demonstrated the expression of functional tat. These cells constitute virus-free tools for functional and structural studies of native env and tat.     

Splicing regulatory elements within tat exon 2 of human immunodeficiency virus type 1 (HIV-1) are characteristic of group M but not group O HIV-1 strains

In the NL4-3 strain of human immunodeficiency virus type 1 (HIV-1), regulatory elements responsible for the relative efficiencies of alternative splicing at the tat, rev, and the env/nef 3' splice sites (A3 through A5) are contained within the region of tat exon 2 and its flanking sequences. Two elements affecting splicing of tat, rev, and env/nef mRNAs have been localized to this region. First, an exon splicing silencer (ESS2) in NL4-3, located approximately 70 nucleotides downstream from the 3' splice site used to generate tat mRNA, acts specifically to inhibit splicing at this splice site. Second, the A4b 3' splice site, which is the most downstream of the three rev 3' splice sites, also serves as an element inhibiting splicing at the env/nef 3' splice site A5. These elements are conserved in some but not all HIV-1 strains, and the effects of these sequence changes on splicing have been investigated in cell transfection and in vitro splicing assays. SF2, another clade B virus and member of the major (group M) viruses, has several sequence changes within ESS2 and uses a different rev 3' splice site. However, splicing is inhibited by the two elements similarly to NL4-3. As with the NL4-3 strain, the SF2 A4b AG dinucleotide overlaps an A5 branchpoint, and thus the inhibitory effect may result from competition of the same site for two different splicing factors. The sequence changes in ANT70C, a member of the highly divergent outlier (group O) viruses, are more extensive, and ESS2 activity in tat exon 2 is not present. Group O viruses also lack the rev 3' splice site A4b, which is conserved in all group M viruses. Mutagenesis of the most downstream rev 3' splice site of ANT70C does not increase splicing at A5, and all of the branchpoints are upstream of the two rev 3' splice sites. Thus, splicing regulatory elements in tat exon 2 which are characteristic of most group M HIV-1 strains are not present in group O HIV-1 strains.     

Mutational analysis of the human immunodeficiency virus type 2 (HIV-2) genome in relation to HIV-1 and simian immunodeficiency virus SIV (AGM).

We constructed an infectious molecular clone of the human immunodeficiency virus type 2 (HIV-2) and generated nine frameshift mutants corresponding to nine open reading frames identified so far. Three structural (gag, pol, env) and two regulative (tat, rev) gene mutants were not infectious, whereas vif, vpx, vpr, and nef genes were dispensable for infectivity. All of the mutants except env and rev were cytopathic in CD4+ human leukemia cells. In transfection assays, the expression of HIV-2 long terminal repeat was activated by infectious clones of HIV-1, HIV-2, and simian immunodeficiency virus from African green monkey but not by the tat mutants. However, an HIV-2 tat mutant could produce small amounts of virus proteins and particles in contrast to a rev mutant, which directed no detectable synthesis of virus proteins and virions.     

The rev (trs/art) protein of human immunodeficiency virus type 1 affects viral mRNA and protein expression via a cis-acting sequence in the env region.

The study of expression of several human immunodeficiency virus type 1 proviral mutants in human cells in the presence or absence of rev (trs/art) protein reveals that rev increases the levels of unspliced and env mRNA and the accumulated structural viral proteins. rev protein produced from appropriate expression vectors fully complements the rev-defective mutants. rev requires the presence of a specific cis-acting sequence for its function. This rev-responsive element sequence has been localized within a 520 base-pair fragment in the env region of human immunodeficiency virus type 1. gag and env expression is coordinately regulated by rev. Two independent cis-acting elements localized in the gag and env regions are responsible for the low levels of gag and env mRNA in the absence of rev. These elements are different than the rev-responsive element and act independent of each other.     

Simian immunodeficiency virus displays complex patterns of RNA splicing.

The human and simian immunodeficiency viruses encode at least six gene products that apparently serve regulatory functions. To evaluate the regulation of simian immunodeficiency virus gene expression at the level of RNA splicing, we used the polymerase chain reaction to amplify and clone cDNAs corresponding to a large array of mRNAs from infected cells. We identified mRNAs that used splice acceptor sites upstream of the initiator codons for tat, rev, vpr, nef, vif, and vpx, suggesting that these proteins may be expressed from different mRNAs. We also provide hybridization data suggesting that the same splice acceptor site may be used for both rev and env mRNAs. Furthermore, we isolated both tat and rev cDNAs that utilized three alternative splice acceptor sites at the start of coding exon 2, indicating that different versions of these proteins may be encoded. Finally, approximately 10 to 20% of simian immunodeficiency virus mRNAs spliced an intron from their untranslated 5' ends, and sequences contained within this intron constituted a portion of the tat-responsive TAR element. Thus, alternative pre-mRNA splicing adds a level of complexity to simian immunodeficiency virus expression, which may affect several levels of gene regulation.     

Equine infectious anemia virus tat: insights into the structure, function, and evolution of lentivirus trans-activator proteins.

Equine infectious anemia virus (EIAV) contains a tat gene which is closely related to the trans-activator genes of the human and simian immunodeficiency viruses. Nucleotide sequence analysis of EIAV cDNA clones revealed that the tat mRNA is composed of three exons; the first two encode Tat and the third may encode a Rev protein. Interestingly, EIAV Tat translation is initiated at a non-AUG codon in exon 1 of the mRNA, perhaps allowing an additional level of gene regulation. The deduced amino acid sequence of EIAV tat, combined with functional analyses of tat cDNAs in transfected cells, has provided some unique insights into the domain structure of Tat. EIAV Tat has a C-terminal basic domain and a highly conserved 16-amino-acid core domain, but not the cysteine-rich region, that are present in the primate immunodeficiency virus Tat proteins. Thus, EIAV encodes a relatively simple version of this kind of trans activator.     

Mutational analysis of simian immunodeficiency virus from African green monkeys and human immunodeficiency virus type 2

We constructed ten mutants of simian immunodeficiency virus isolated from African green monkey (SIVAGM), and nine mutants of human immunodeficiency virus type 2 (HIV-2) in vitro. Their infectivity, cytopathogenicity, transactivation potential, virus RNA, and protein synthesis were examined by transfection and infection experiments. Mutations in three structural (gag, pol, env) and two regulator (tat, rev) genes abolished the infectivity of both viruses, but vpx, vpr (HIV-2), and nef were dispensable and mutant viruses were indistinguishable phenotypically from wild type virus. A vif mutant of HIV-2 showed poor infectivity in cell-free condition, whereas SIVAGM mutants grew equally well with wild type virus. In transient transfection assays, rev mutants derived from both viruses produced mainly small mRNA species and no detectable virus proteins and particles. Transactivation potential of tat mutants originated from both viruses was about three- to ten-fold less than that of respective wild type DNAs, generating small amounts of virus.     

Mechanism of translation of monocistronic and multicistronic human immunodeficiency virus type 1 mRNAs.

We have used a panel of cDNA clones expressing wild-type and mutant human immunodeficiency virus type 1 (HIV-1) mRNAs to study translation of these mRNAs in eucaryotic cells. The tat open reading frame (ORF) has a strong signal for translation initiation, while rev and vpu ORFs have weaker signals. The expression of downstream ORFs is inhibited in mRNAs that contain the tat ORF as the first ORF. In contrast, downstream ORFs are expressed efficiently from mRNAs that have rev or vpu as the first ORF. All env mRNAs contain the upstream vpu ORF. Expression of HIV-1 Env protein requires a weak vpu AUG, which allows leaky scanning to occur, thereby allowing ribosomes access to the downstream env ORF. We concluded that HIV-1 mRNAs are translated by the scanning mechanism and that expression of more than one protein from each mRNA was caused by leaky scanning at the first AUG of the mRNA.     

Cloning and characterization of cDNAs encoding equine infectious anemia virus tat and putative Rev proteins.

We isolated and characterized six cDNA clones from an equine infectious anemia virus-infected cell line that displays a Rev-defective phenotype. With the exception of one splice site in one of the clones, all six cDNAs exhibited the same splicing pattern and consisted of four exons. Exon 1 contained the 5' end of the genome; exon 2 contained the tat gene from mid-genome; exon 3 consisted of a small section of env, near the 5' end of the env gene; and exon 4 contained the putative rev open reading frame from the 3' end of the genome. The structures of the cDNAs predict a bicistronic message in which Tat is encoded by exons 1 and 2 and the presumptive Rev protein is encoded by exons 3 and 4. tat translation appears to be initiated at a non-AUG codon within the first 15 codons of exon 1. Equine infectious anemia virus-specific tat activity was expressed in transient transfections with cDNA expression plasmids. The predicted wild-type Rev protein contains 30 env-derived amino acids and 135 rev open reading frame residues. All of the cDNAs had a frameshift in exon 4, leading to a truncated protein and thus providing a plausible explanation for the Rev-defective phenotype of the original cells. We used peptide antisera to detect the faulty protein, thus confirming the cDNA sequence, and to detect the normal protein in productively infected cells.     

Characterization of multiple mRNA species of simian immunodeficiency virus from macaques in a CD4+ lymphoid cell line.

Cytoplasmic poly(A)+ RNA was isolated from CEMX721.174 cells 5 to 10 days after infection with molecularly cloned simian immunodeficiency virus SIVmac239. Expression of SIV RNA was analyzed by Northern (RNA) blot hybridization and by sequencing of cDNA clones. As expected, a splice donor site was demonstrated in the untranslated leader sequence outside the left long terminal repeat. The region between pol and env was found to contain at least two splice donor and six splice acceptor sites. Splice acceptor and donor sites in the intergenic region were suitably positioned for expression of vpx, vpr, tat, and rev. Splice acceptor sites at nucleotides 8802 and 8805 were demonstrated in singly and doubly spliced RNAs with the potential of expressing nef and the second exons of tat and rev. Our results demonstrate a complex pattern of alternative splicing of SIV mRNAs. The results are very similar to what has been observed in human immunodeficiency virus type 1-infected cells, suggesting that both human and simian immunodeficiency viruses are subject to multiple levels of regulation.     

Truncation of the human immunodeficiency virus type 1 transmembrane glycoprotein cytoplasmic domain blocks virus infectivity.

Human immunodeficiency virus type 1 contains a transmembrane glycoprotein with an unusually long cytoplasmic domain. To determine the role of this domain in virus replication, a series of single nucleotide changes that result in the insertion of premature termination codons throughout the cytoplasmic domain has been constructed. These mutations delete from 6 to 192 amino acids from the carboxy terminus of gp41 and do not affect the amino acid sequence of the regulatory proteins encoded by rev and tat. The effects of these mutations on glycoprotein biosynthesis and function as well as on virus infectivity have been examined in the context of a glycoprotein expression vector and the viral genome. All of the mutant glycoproteins were synthesized, processed, and transported to the cell surface in a manner similar to that of the wild-type glycoprotein. With the exception of mutants that remove the membrane anchor domain, all of the mutant glycoproteins retained the ability to cause fusion of CD4-bearing cells. However, deletion of more than 19 amino acids from the C terminus of gp41 blocked the ability of mutant virions to infect cells. This defect in virus infectivity appeared to be due at least in part to a failure of the virus to efficiently incorporate the truncated glycoprotein. Similar data were obtained for mutations in two different env genes and two different target cell lines. These results indicate that the cytoplasmic domain of gp41 plays a critical role during virus assembly and entry in the life cycle of human immunodeficiency virus type 1.     

Generation of a chimeric human and simian immunodeficiency virus infectious to monkey peripheral blood mononuclear cells.

We constructed five chimeric clones between human immunodeficiency virus type 1 (HIV-1) and simian immunodeficiency virus (SIVMAC) and four SIVMAC mutants by recombinant DNA techniques. Three chimeric clones and all mutants with an alteration in either the vif, vpx, vpr, or nef gene were infectious to human CD4-positive cell lines. The susceptibility of macaque monkey peripheral blood mononuclear cells (PBMC) to infection by these mutants and chimeras was examined in vitro. Macaque PBMC supported the replication of wild-type and vpx, vpr, and nef mutant SIVMAC strains. A chimera carrying the long terminal repeats (LTRs), gag, pol, vif, and vpx of SIVMAC and tat, rev, vpu, and env of HIV-1 was also replication competent in PBMC. In contrast, HIV-1, the vif mutant of SIVMAC, a chimera containing rev and env of SIVMAC, and a chimera containing vpx, vpr, tat, rev, and env of SIVMAC did not grow in PBMC. Western immunoblotting analysis of the replicating chimera in PBMC confirmed the hybrid nature of the virus. These data strongly suggested that the sequence important for macaque cell tropism lies within the LTR, gag, pol, and/or vif sequences of the SIVMAC genome.     

Synthesis of functional human immunodeficiency virus tat protein in baculovirus as determined by a cell-cell fusion assay.

The human immunodeficiency virus tat protein is a strong trans-activator of the expression of mRNAs originating from the viral long terminal repeat. We have expressed the first 72 amino acids (coding exon 1) of this protein in eucaryotic Spodoptera frugiperda SF9 cells by using a baculovirus vector, Autographa californica nuclear polyhedrosis virus. We show that the baculovirus vector stably produced the 72-amino-acid form of the tat protein but was unable to stably synthesize a larger 101-amino-acid full-length version of the same polypeptide. The 72-amino-acid tat protein, when introduced into mammalian fibroblasts by using a cell-cell fusion technique, functionally trans-activated the expression of the human immunodeficiency virus long terminal repeat.     

Independent fluctuation of human immunodeficiency virus type 1 rev and gp41 quasispecies in vivo.

The human immunodeficiency virus type 1 (HIV-1) overlapping rev and env coding sequences have been examined from sequential peripheral blood mononuclear cell DNA samples from one individual. These were the same DNA samples from which sequence data for the tat and nef/long terminal repeat loci have been derived and span a 4-year period. The rev/env sequences were established by sequencing cloned polymerase chain reaction products. The structure of the populations of rev protein sequences increased in complexity with disease, while those of the corresponding env sequences remained complex. This suggests that the rev and env populations evolved differently, probably reflecting different selection pressures. No defective rev variants encoded substitutions in residues 76 through 79, indicating that the experimental finding of down regulation of rev activity by competitive inhibition may not necessarily occur in vivo. After having analyzed three HIV loci (15% of the genome) from the same individual over 4 years, it is clear that no two loci evolved similarly, indicating the difficulties in comparing data from different loci.     

Construction and characterization of an infectious DNA clone and of mutants of simian immunodeficiency virus isolated from the African green monkey.

We constructed a full-length molecular clone of simian immunodeficiency virus from an African green monkey. Upon transfection, this clone directed the production of virus particles cytopathic and infectious to human CD4+ leukemia cell lines. Mutations were introduced by recombinant DNA techniques into eight open reading frames of simian immunodeficiency virus from the African green monkey thus far identified. The phenotypes of mutant viruses, i.e., infectivity, cytopathogenicity, transactivation of gene expression controlled by a long terminal repeat, and viral RNA and protein syntheses, were examined by transfection and infection experiments. Three structural (gag, pol, and env) and two regulatory (tat and rev) gene mutants were not infectious, whereas vif, vpx, and nef were dispensable for infectivity and mutant viruses were highly cytopathic. In transient transfection assays, a rev mutant produced mainly small mRNA species and no detectable virus protein and particles. The transactivation potential of a tat mutant was about 10-fold less than that of wild-type DNA, generating small amounts of virus.     

Efficiency of reinitiation of translation on human immunodeficiency virus type 1 mRNAs is determined by the length of the upstream open reading frame and by intercistronic distance.

In this study, we examined the mechanism of translation of the human immunodeficiency virus type 1 tat mRNA in eucaryotic cells. This mRNA contains the tat open reading frame (ORF), followed by rev and nef ORFs, but only the first ORF, encoding tat, is efficiently translated. Introduction of premature stop codons in the tat ORF resulted in efficient translation of the downstream rev ORF. We show that the degree of inhibition of translation of rev is proportional to the length of the upstream tat ORF. An upstream ORF spanning 84 nucleotides was predicted to inhibit 50% of the ribosomes from initiating translation at downstream AUGs. Interestingly, the distance between the upstream ORF and the start codon of the second ORF also played a role in efficiency of downstream translation initiation. It remains to be investigated if these conclusions relate to translation of mRNAs other than human immunodeficiency virus type 1 mRNAs. The strong inhibition of rev translation exerted by the presence of the tat ORF may reflect the different roles of Tat and Rev in the viral life cycle. Tat acts early to induce high production of all viral mRNAs. Rev induces a switch from the early to the late phase of the viral life cycle, resulting in production of viral structural proteins and virions. Premature Rev production may result in entrance into the late phase in the presence of suboptimal levels of viral mRNAs coding for structural proteins, resulting in inefficient virus production.