The structural requirements for assembly of functional class II transfer RNA core regions have been examined by sequence analysis and tested by reconstruction of alternative folds into the tertiary domain of Escherichia coli tRNA2Gln. At least four distinct designs have been identified that permit stable folding and efficient synthetase recognition, as assessed by thermal melting profiles and glutaminylation kinetics. Although most large variable-arm tRNAs found in nature possess an enlarged D-loop, lack of this feature can be compensated for by insertion of nucleotides either 3′ to the variable loop or within the short acceptor/D-stem connector region. Rare pyrimidines at nt 9 in the core region can be accommodated in the class II framework, but only if specific nucleotides are present either in the D-loop or 3′ to the variable arm. Glutaminyl-tRNA synthetase requires one or two unpaired uridines 3′ to the variable arm to efficiently aminoacylate several of the class II frameworks. Because there are no specific enzyme contacts in the tRNAGln core region, these data suggest that tRNA discrimination by GlnRS depends in part on indirect readout of RNA sequence information.

The monoclonal anti-dsRNA antibody J2 binds double-stranded RNAs (dsRNA) in an apparently sequence-nonspecific way. The mAb only recognizes antigens with double-stranded regions of at least 40 bp and its affinity to poly(A) poly(U) and to dsRNAs with mixed base pair composition is about tenfold higher than to poly(I) poly(C). Because no specific binding site could be determined, the number, the exact dimensions, and other distinct features of the binding sites on a given antigen are difficult to evaluate by biochemical methods. We therefore employed scanning force microscopy (SFM) as a method to analyze antibody–dsRNA interaction and protein–RNA binding in general. Several in vitro-synthesized dsRNA substrates, generated from the Dictyostelium PSV-A gene, were used. In addition to the expected sequence-nonspecific binding, imaging of the complexes indicated preferential binding of antibodies to the ends of dsRNA molecules as well as to certain internal sites. Analysis of 2,000 bound antibodies suggested that the consensus sequence of a preferential internal binding site is A2N9A3N9A2, thus presenting A residues on one face of the helix. The site was verified by site-directed mutagenesis, which abolished preferential binding to this region. The data demonstrate that SFM can be efficiently used to identify and characterize binding sites for proteins with no or incomplete sequence specificity. This is especially the case for many proteins involved in RNA metabolism.

Rpb5-H147R is an AT-GC transition replacing CAC(His) by CGC(Arg) at a conserved and critical position of ABC27 (Rpb5p), one of the five common and essential subunits shared by all three eukaryotic RNA polymerases. This mutation is viable at 25 °C, but has a lethal phenotype at 34 °C. A search for dosage-dependent suppressors identified five distinct clones that all bear a copy of the tRNAHisGUG gene. Suppression was also observed with a small genomic insert bearing this tRNA gene and no other coding sequences, under conditions where there is a sevenfold increase in the cellular concentration of tRNAHisGUG. Overexpressing tRNAArgICG, which normally decodes the suppressed CGC codon, counteracted suppression. Suppression is codon specific because it was abolished when replacing CGC by its synonymous codons CGA, CGU, or AGA, but was not detectably affected by several nucleotide substitutions modifying the surrounding sequence and is thus largely insensitive to the nucleotide context. It is proposed that overexpressing tRNAHisGUG extends its decoding properties from CAC(His) to the noncognate CGC(Arg) codon through an illegitimate U[bull ]G pairing at the middle base of the anticodon. Accordingly, tRNAHisGUG would compete with tRNAArgICG for chain elongation and generate a significant level of misreading errors under normal growth conditions.

The crystal structure of r(GCCACCCUG)[bull ]r(CAGGGUCGGC), helix II of the Xenopus laevis 5S rRNA with a cytosine bulge (underlined), has been determined in two forms at 2.2 Å (Form I, space group P42212, a = b = 57.15 Å and c = 43.54 Å) and 1.7 Å (Form II, space group P43212, a = b = 32.78 Å and c = 102.5 Å). The helical regions of the nonamers are found in the standard A-RNA conformations and the two forms have an RMS deviation of 0.75 Å. However, the cytosine bulge adopts two significantly different conformations with an RMS deviation of 3.9 Å. In Form I, the cytosine bulge forms an intermolecular C+∗G[bull ]C triple in the major groove of a symmetry-related duplex with intermolecular hydrogen bonds between N4C and O6G, and between protonated N3+C and N7G. In contrast, a minor groove C∗G[bull ]C triple is formed in Form II with intermolecular hydrogen bonds between O2C and N2G, and between N3C and N3G with a water bridge. A partial major groove opening was observed in Form I structure at the bulge site. Two Ca2+ ions were found in Form I helix whereas there were none in Form II. The structural comparison of these two forms indicates that bulged residues can adopt a variety of conformations with little perturbation to the global helix structure. This suggests that bulged residues could function as flexible latches in bridging double helical motifs and facilitate the folding of large RNA molecules.

The replicase activity of rotavirus open cores has been used to study the synthesis of (−) strand RNA from viral (+) strand RNA in a cell-free replication system. The last 7 nt of the (+) strand RNA, 5′-UGUGACC-3′, are highly conserved and are necessary for efficient (−) strand synthesis in vitro. Characterization of the cell-free replication system revealed that the addition of NaCl inhibited (−) strand synthesis. By preincubating open cores with (+) strand RNA and ATP, CTP, and GTP prior to the addition of NaCl and UTP, the salt-sensitive step was overcome. Thus, (−) strand initiation, but not elongation, was a salt-sensitive process in the cell-free system. Further analysis of the requirements for initiation showed that preincubating open cores and the (+) strand RNA with GTP or UTP, but not with ATP or CTP, allowed (−) strand synthesis to occur in the presence of NaCl. Mutagenesis suggested that in the presence of GTP, (−) strand synthesis initiated at the 3′-terminal C residue of the (+) strand template, whereas in the absence of GTP, an aberrant initiation event occurred at the third residue upstream from the 3′ end of the (+) strand RNA. During preincubation with GTP, formation of the dinucleotides pGpG and ppGpG was detected; however, no such products were made during preincubation with ATP, CTP, or UTP. Replication assays showed that pGpG, but not GpG, pApG, or ApG, served as a specific primer for (−) strand synthesis and that the synthesis of pGpG may occur by a template-independent process. From these data, we conclude that initiation of rotavirus (−) strand synthesis involves the formation of a ternary complex consisting of the viral RNA-dependent RNA polymerase, viral (+) strand RNA, and possibly a 5′-phosphorylated dinucleotide, that is, pGpG or ppGpG.

Human TAP and Saccharomyces cerevisiae Mex67p belong to a family of proteins that mediate mRNA export. Computer searches identified previously two Caenorhabditis elegans genes, C15H11.3 and C15H11.6, that encode putative homologs of hTAP and Mex67p (Segref et al., EMBO J, 1997, 16:3256–3271). Using RNA interference experiments in C. elegans, we found that functional knockout of C15H11.3 resulted in nuclear accumulation of poly(A)-containing RNAs and was lethal for both embryos and adult nematodes. No embryonic or progeny abnormality was observed in functional knockout of C15H11.6. Taken together, these data established that the C15H11.3 gene product is an ortholog of hTAP and Mex67p; thus, it was named Ce-NXF-1. Ce-NXF-1 binds RNA directly and is a nucleocytoplasmic shuttle protein accumulating in the nucleoplasm and at the nuclear rim. The rim association is mediated via unique signals present in the C-terminal portion of all TAP/NXF and Mex67p proteins. This region was shown to interact with the FG-repeat domains of nucleoporins Nup98, Nup153, and Nup214, indicating that the rim association occurs through components of the nuclear pore complex. In summary, Ce-NXF-1 belongs together with hTAP and Mex67p to a family of proteins that participate in mRNA export and can provide a direct molecular link between mRNAs and components of the nuclear pore complex. Therefore, despite differences in mRNA metabolism between these species, they utilize a conserved mRNA transport mechanism.

The structure of the viral RNA (vRNA) inside intact nucleocapsids of vesicular stomatitis virus was studied by chemical probing experiments. Most of the Watson–Crick positions of the nucleotide bases of vRNA in intact virus and in nucleoprotein (N)-RNA template were accessible to the chemical probes and the phosphates were protected. This suggests that the nucleoprotein binds to the sugar–phosphate backbone of the RNA and leaves the Watson–Crick positions free for the transcription and replication activities of the viral RNA-dependent RNA polymerase. The same architecture has been proposed for the influenza virus nucleocapsids. However, about 5% of the nucleotide bases were found to be relatively nonreactive towards the chemical probes and some bases were hyperreactive. The pattern of reactivities was the same for RNA inside virus and for RNA in N-RNA template that was purified over a CsCl gradient and which had more than 94% of the polymerase and phosphoprotein molecules removed. All reactivities were more or less equal on naked vRNA. This suggests that the variations in reactivity towards the chemical probes are caused by the presence of the nucleoprotein.

The translation machinery recognizes codons that enter the ribosomal A site with remarkable accuracy to ensure that polypeptide synthesis proceeds with a minimum of errors. When a termination codon enters the A site of a eukaryotic ribosome, it is recognized by the release factor eRF1. It has been suggested that the recognition of translation termination signals in these organisms is not limited to a simple trinucleotide codon, but is instead recognized by an extended tetranucleotide termination signal comprised of the stop codon and the first nucleotide that follows. Interestingly, pharmacological agents such as aminoglycoside antibiotics can reduce the efficiency of translation termination by a mechanism that alters this ribosomal proofreading process. This leads to the misincorporation of an amino acid through the pairing of a near-cognate aminoacyl tRNA with the stop codon. To determine whether the sequence context surrounding a stop codon can influence aminoglycoside-mediated suppression of translation termination signals, we developed a series of readthrough constructs that contained different tetranucleotide termination signals, as well as differences in the three bases upstream and downstream of the stop codon. Our results demonstrate that the sequences surrounding a stop codon can play an important role in determining its susceptibility to suppression by aminoglycosides. Furthermore, these distal sequences were found to influence the level of suppression in remarkably distinct ways. These results suggest that the mRNA context influences the suppression of stop codons in response to subtle differences in the conformation of the ribosomal decoding site that result from aminoglycoside binding.

Using random mutagenesis and a genetic screening in yeast, we isolated 26 mutations that inactivate Saccharomyces cerevisiae arginyl–tRNA synthetase (ArgRS). The mutations were identified and the kinetic parameters of the corresponding proteins were tested after purification of the expression products in Escherichia coli. The effects were interpreted in the light of the crystal structure of ArgRS. Eighteen functional residues were found around the arginine-binding pocket and eight others in the carboxy-terminal domain of the enzyme. Mutations of these residues all act by strongly impairing the rates of tRNA charging and arginine activation. Thus, ArgRS and tRNAArg can be considered as a kind of ribonucleoprotein, where the tRNA, before being charged, is acting as a cofactor that activates the enzyme. Furthermore, by using different tRNAArg isoacceptors and heterologous tRNAAsp, we highlighted the crucial role of several residues of the carboxy-terminal domain in tRNA recognition and discrimination.

Members of the double-stranded RNA- (dsRNA) specific adenosine deaminase gene family convert adenosine residues into inosines in dsRNA and are involved in A-to-I RNA editing of transcripts of glutamate receptor (GluR) subunits and serotonin receptor subtype 2C (5-HT2CR). We have isolated hADAR3, the third member of this class of human enzyme and investigated its editing site selectivity using in vitro RNA editing assay systems. As originally reported for rat ADAR3 or RED2, purified ADAR3 proteins could not edit GluR-B RNA at the “Q/R” site, the “R/G” site, and the intronic “hot spot” site. In addition, ADAR3 did not edit any of five sites discovered recently within the intracellular loop II region of 5-HT2CR RNAs, confirming its total lack of editing activity for currently known substrate RNAs. Filter-binding analyses revealed that ADAR3 is capable of binding not only to dsRNA but also to single-stranded RNA (ssRNA). Deletion mutagenesis identified a region rich in arginine residues located in the N-terminus that is responsible for binding of ADAR3 to ssRNA. The presence of this ssRNA-binding domain as well as its expression in restricted brain regions and postmitotic neurons make ADAR3 distinct from the other two ADAR gene family members, editing competent ADAR1 and ADAR2. ADAR3 inhibited in vitro the activities of RNA editing enzymes of the ADAR gene family, raising the possibility of a regulatory role in RNA editing.

RAE1 is an evolutionarily conserved protein that associates with both mRNPs and nucleoporins, and may bridge the interaction between mRNP export cargoes and the nuclear pore complex (NPC). However, the mechanism by which RAE1 functions in mRNA export is still unknown and the time point at which RAE1 interacts with the exported RNP has not been directly investigated. Here we have addressed this question in the Balbiani ring (BR) system of Chironomus tentans using immunoelectron microscopy. The RAE1 protein of C. tentans, Ct-RAE1, is 70% identical to human RAE1/mrnp41 (hRAE1) and is recognized by antibodies raised against hRAE1. As in vertebrate cells, Ct-RAE1 is concentrated at the nuclear envelope and also dispersed throughout the nuclear interior. Here we show that Ct-RAE1 does not bind to the BR particle either cotranscriptionally or in the nucleoplasm. Instead, the interaction between Ct-RAE1 and the exported BR particle occurs at the NPC. Moreover, the localization of Ct-RAE1 at the NPC is correlated with the presence of an exported RNP in the NPC. Finally, the anti-RAE1 antibody does not label the cytoplasmic side of BR particles in transit through the central channel, which indicates that Ct-RAE1 either remains anchored at the nuclear side of the NPC during translocation of the RNP through the central channel or becomes transiently associated with the RNP but is rapidly released into the cytoplasm.

Using selection-amplification, we have isolated RNAs with affinity for translation termination factors eRF1 and eRF1[bull ]eRF3 complex. Individual RNAs not only bind, but inhibit eRF1-mediated release of a model nascent chain from eukaryotic ribosomes. There is also significant but weaker inhibition of eRF1-stimulated eRF3 GTPase and eRF3 stimulation of eRF1 release activity. These latter selected RNAs therefore hinder eRF1[bull ]eRF3 interactions. Finally, four RNA inhibitors of release suppress a UAG stop codon in mammalian extracts dependent for termination on eRF1 from several metazoan species. These RNAs are therefore new specific inhibitors for the analysis of eukaryotic termination, and potentially a new class of omnipotent termination suppressors with possible therapeutic significance.

The untranslated leader of the RNA genome of the human immunodeficiency virus type 1 (HIV-1) encodes multiple signals that regulate distinct steps of the viral replication cycle. The RNA secondary structure of several replicative signals in the HIV-1 leader is critical for function. Well-known examples include the TAR hairpin that forms the binding site for the viral Tat trans-activator protein and the DIS hairpin that is important for dimerization and subsequent packaging of the viral RNA into virion particles. In this study, we present evidence for the formation of a tertiary structure by the complete HIV-1 leader RNA. This conformer was recognized as a fast-migrating band on nondenaturing polyacrylamide gels, and such a migration effect is generally attributed to differences in compactness. Both the 5′ and 3′ domains of the 335-nt HIV-1 leader RNA are required for the formation of the compact RNA structure, and the presence of several putative interaction domains was revealed by an extensive analysis of the denaturing effect of antisense DNA oligonucleotides. The buffer conditions and sequence requirements for conformer formation are strikingly different from that of the RNA-dimerization reaction. In particular, the conformer was destabilized in the presence of Mg2+ ions and by the viral nucleocapsid (NC) protein. The presence of a stable RNA structure in the HIV-1 leader was also apparent when this RNA was used as template for reverse transcription, which yielded massive stops ahead of the structured leader domain. Formation of the conformer is a reversible event, suggesting that the HIV-1 leader is a dynamic molecule. The putative biological function of this conformational polymorphism as molecular RNA switch in the HIV-1 replication cycle is discussed.

Ribonuclease-resistant RNA molecules that bind to infectious human cytomegalovirus (HCMV) were isolated in vitro from a pool of randomized sequences after 16 cycles of selection and amplification. The two ligands (L13 and L19) characterized exhibited high HCMV-binding affinity in vitro and effectively inhibited viral infection in tissue culture. Their antiviral activity was also specific as they only reacted with two different strains of HCMV but not with the related herpes simplex virus 1 and human cells. These two ligands appeared to function as antivirals by blocking viral entry. Ultraviolet (UV) crosslinking studies suggested that L13 and L19 bind to HCMV essential glycoproteins B and H, respectively. Thus, RNA ligands that bind to different surface antigens of HCMV can be simultaneously isolated by the selection procedure. Our study demonstrates the feasibility of using these RNA ligands as a research tool to identify viral proteins required for infectivity and as an antiviral agent to block viral infection.

Compelling in vivo studies suggest a tight functional linkage between RNA polymerase II transcription and pre-messenger RNA splicing. At present, the specific interactions involved in this coupling are poorly understood and deserve investigation. To this end, we developed an in vitro system that permits study of coupled transcription and splicing. Transcripts generated by RNA polymerase II were accurately and efficiently spliced under reaction conditions that permitted both transcription and splicing to occur simultaneously. The splicing of RNA-polymerase-II-driven transcripts was accelerated relative to that of the same transcripts driven by T7 RNA polymerase. Moreover, the product of exon ligation was found associated with the DNA template in reactions driven by RNA polymerase II. These two findings indicate that transcription and splicing were coupled in the in vitro system driven by RNA polymerase II, and suggest that this system will be useful for the biochemical study of this coupling.

We developed a two-step purification of mammalian polyadenylation complexes assembled in vitro. Biotinylated pre-mRNAs containing viral or immunoglobulin poly(A) sites were incubated with nuclear extracts prepared from mouse myeloma cells under conditions permissive for in vitro cleavage and polyadenylation and the mixture was fractionated by gel filtration; complexes containing biotinylated pre-mRNA and bound proteins were affinity purified on avidin-agarose resin. Western analysis of known components of the polyadenylation complex demonstrated copurification of polyadenylation factors with poly(A) site-containing RNA but not with control RNA substrates containing either no polyadenylation signals or a point mutation of the AAUAAA polyadenylation signal. Polyadenylation complexes that were assembled on exogenous RNA eluted from the Sephacryl column in fractions consistent with their size range extending from 2 to 4 × 106Mr. Complexes endogenous to the extract were of approximately the same apparent size, but more heterogeneous in distribution. This method can be used to study polyadenylation/cleavage complexes that may form upon a number of different RNA sequences, an important step towards defining which factors might differentially associate with specific RNAs.

The yeast superkiller (SKI) genes were originally identified from mutations allowing increased production of killer toxin encoded by M “killer” virus, a satellite of the dsRNA virus L-A. XRN1 (SKI1) encodes a cytoplasmic 5′-exoribonuclease responsible for the majority of cytoplasmic RNA turnover, whereas SKI2, SKI3, and SKI8 are required for normal 3′-degradation of mRNA and for repression of translation of poly(A) minus RNA. Ski2p is a putative RNA helicase, Ski3p is a tetratricopeptide repeat (TPR) protein, and Ski8p contains five WD-40 (beta-transducin) repeats. An xrn1 mutation in combination with a ski2, ski3, or ski8 mutation is lethal, suggesting redundancy of function. Using functional epitope-tagged Ski2, Ski3, and Ski8 proteins, we show that Ski2p, Ski3p, and Ski8p can be coimmunoprecipitated as an apparent heterotrimeric complex. With epitope-tagged Ski2p, there was a 1:1:1 stoichiometry of the proteins in the complex. Ski2p did not associate with Ski3p in the absence of Ski8p, nor did Ski2p associate with Ski8p in the absence of Ski3p. However, the Ski3p/Ski8p interaction did not require Ski2p. In addition, ski6-2 or ski4-1 mutations or deletion of SKI7 did not affect complex formation. The identification of a complex composed of Ski2p, Ski3p, and Ski8p explains previous results showing phenotypic similarity between mutations in SKI2, SKI3, and SKI8. Indirect immunofluorescence of Ski3p and subcellular fractionation of Ski2p and Ski3p suggest that Ski2p and Ski3p are cytoplasmic. These data support the idea that Ski2p, Ski3p, and Ski8p function in the cytoplasm in a 3′-mRNA degradation pathway.

The transcript of the Saccharomyces cerevisiae gene, RPL30, is subject to regulated splicing and regulated translation, due to a structure that interacts with its own product, ribosomal protein L30. We have followed the fate of the regulated RPL30 transcripts in vivo. Initially, these transcripts abortively enter the splicing pathway, forming an unusually stable association with U1 snRNP. A large proportion of the unspliced molecules, however, are found in the cytoplasm. Most of these are still bound by L30, as only a small fraction are engaged in translation. Eventually, the unspliced RPL30 transcripts escape the grasp of L30, associate with ribosomes, and fall prey to nonsense mediated decay.

The SRm160/300 splicing coactivator, which consists of the serine/arginine (SR)-related nuclear matrix protein of 160 kDa and a 300-kDa nuclear matrix antigen, functions in splicing by promoting critical interactions between splicing factors bound to pre-mRNA, including snRNPs and SR family proteins. In this article we report the isolation of a cDNA encoding the 300-kDa antigen and investigate the activity of it and SRm160 in splicing. Like SRm160, the 300-kDa antigen contains domains rich in alternating S and R residues but lacks an RNA recognition motif; the protein is accordingly named “SRm300.” SRm300 also contains a novel and highly conserved N-terminal domain, several unique repeated motifs rich in S, R, and proline residues, and two very long polyserine tracts. Surprisingly, specific depletion of SRm300 does not prevent the splicing of pre-mRNAs shown previously to require SRm160/300. Addition of recombinant SRm160 alone to SRm160/300-depleted reactions specifically activates splicing. The results indicate that SRm160 may be the more critical component of the SRm160/300 coactivator in the splicing of SRm160/300-dependent pre-mRNAs.

The human coronavirus 229E replicase gene encodes a protein, p66HEL, that contains a putative zinc finger structure linked to a putative superfamily (SF) 1 helicase. A histidine-tagged form of this protein, HEL, was expressed using baculovirus vectors in insect cells. The purified recombinant protein had in vitro ATPase activity that was strongly stimulated by poly(U), poly(dT), poly(C), and poly(dA), but not by poly(G). The recombinant protein also had both RNA and DNA duplex-unwinding activities with 5′-to-3′ polarity. The DNA helicase activity of the enzyme preferentially unwound 5′-oligopyrimidine-tailed, partial-duplex substrates and required a tail length of at least 10 nucleotides for effective unwinding. The combined data suggest that the coronaviral SF1 helicase functionally differs from the previously characterized RNA virus SF2 helicases.