Essential and nonessential elements in the 3' nontranslated region of Bovine viral diarrhea virus

The 3' nontranslated region (NTR) of the pestivirus Bovine viral diarrhea virus (BVDV), a close relative of human Hepatitis C virus, consists of three stem-loops which are separated by two single-stranded regions. As in other positive-stranded RNA viruses, the 3' NTR of pestiviruses is involved in crucial processes of the viral life cycle. While several studies characterized cis-acting elements within the 3' NTR of a BVDV replicon, there are no studies addressing the significance of these elements in the context of a replicating virus. To examine the functional importance of 3' NTR elements, a set of 4-base deletions and deletions of each of the three stem-loops were introduced into an infectious BVDV cDNA clone. Emerging mutant viruses were characterized with regard to plaque phenotype, growth kinetics, and synthesis of viral RNA. The results indicated that presence of stem-loop (SL) I and the 3'-terminal part of the single-stranded region between stem-loops I and II are indispensable for pestiviral replication. In contrast, deletions within SL II and SL III as well as absence of either SL II or SL III still allowed efficient viral replication; however, a mutant RNA lacking both SL II and SL III was not infectious. The results of this study provide a detailed map of the essential and nonessential elements within the 3' NTR of BVDV and contribute to our understanding of sequence and structural elements important for efficient viral replication of pestiviruses in natural host cells.     

A stem-loop motif formed by the immediate 5' terminus of the bovine viral diarrhea virus genome modulates translation as well as replication of the viral RNA

Bovine viral diarrhea virus (BVDV), a Pestivirus member of the Flaviviridae family, has a positive-stranded RNA genome which consists of a single open reading frame (ORF) and untranslated regions (UTRs) at the 5' and 3' ends. The 5' UTR harbors extensive RNA structure motifs; most of them were shown to contribute to an internal ribosomal entry site (IRES), which mediates cap-independent translation of the ORF. The extreme 5'-terminal region of the BVDV genome had so far been believed not to be required for IRES function. By structure probing techniques, we initially verified the existence of a computer-predicted stem-loop motif at the 5' end of the viral genome (hairpin Ia) as well as at the 3' end of the complementary negative-strand replication intermediate [termed hairpin Ia (-)]. While the stem of this structure is mainly constituted of nucleotides that are conserved among pestiviruses, the loop region is predominantly composed of variable residues. Taking a reverse genetics approach to a subgenomic BVDV replicon RNA (DI9c) which could be equally employed in a translation as well as replication assay system based on BHK-21 cells, we obtained the following results. (i) Proper folding of the Ia stem was found to be crucial for efficient translation. Thus, in the context of an authentic replication-competent viral RNA, the 5'-terminal motif operates apparently as an integral functional part of the ribosome entry. (ii) An intact loop structure and a stretch of nucleotide residues that constitute a portion of the stem of the Ia or the Ia (-) motif, respectively, were defined to represent important determinants of the RNA replication pathway. (iii) Formation of the stem structure of the Ia (-) motif was determined to be not critical for RNA replication. In summary, our findings affirmed that the 5'-terminal region of the BVDV genome encodes a bifunctional secondary structure motif which may enable the viral RNA to switch from the translation to the replicative cycle and vice versa.     

Stem-loop IV in the 5' untranslated region is a cis-acting element in bovine coronavirus defective interfering RNA replication

The 210-nucleotide (nt) 5' untranslated region (UTR) in the positive-strand bovine coronavirus (BCoV) genome is predicted to contain four higher-order structures identified as stem-loops I to IV, which may function as cis-acting elements in genomic RNA replication. Here, we describe evidence that stem-loop IV, a bulged stem-loop mapping at nt 186 through 215, (i) is phylogenetically conserved among group 2 coronaviruses and may have a homolog in groups 1 and 3, (ii) exists as a higher-order structure on the basis of enzyme probing, (iii) is required as a higher-order element for replication of a BCoV defective interfering (DI) RNA in the positive but not the negative strand, and (iv) as a higher-order structure in wild-type (wt) and mutant molecules that replicate, specifically binds six cellular proteins in the molecular mass range of 25 to 58 kDa as determined by electrophoretic mobility shift and UV cross-linking assays; binding to viral proteins was not detected. Interestingly, the predicted stem-loop IV homolog in the severe acute respiratory syndrome (SARS) coronavirus appears to be group 1-like in that it is in part duplicated with a group 1-like conserved loop sequence and is not group 2-like, as would be expected by the SARS coronavirus group 2-like 3' UTR structure. These results together indicate that stem-loop IV in the BCoV 5' UTR is a cis-acting element for DI RNA replication and that it might function through interactions with cellular proteins. It is postulated that stem-loop IV functions similarly in the virus genome.     

Role of the 5' TAR stem--loop and the U5-AUG duplex in dimerization of HIV-1 genomic RNA

The HIV-1 genome consists of two identical RNAs that are linked together through noncovalent interactions involving nucleotides from the 5' untranslated region (5' UTR) of each RNA strand. The 5' UTR is the most conserved part of the HIV-1 RNA genome, and its 335 nucleotide residues form regulatory motifs that mediate multiple essential steps in the viral replication cycle. Here, studying the effect of selected mutations both singly and together with mutations disabling SL1 (SL1 is a 5' UTR stem-loop containing a palindrome called the dimerization initiation site), we have done a rather systematic survey of the 5' UTR requirements for full genomic RNA dimerization in grown-up (i.e., predominantly >/=10 h old) HIV-1 viruses produced by transfected human and simian cells. We have identified a role for the 5' transactivation response element (5' TAR) and a contribution of a long-distance base pairing between a sequence located at the beginning of the U5 region and nucleotides surrounding the AUG Gag initiation codon. The resulting intra- or intermolecular duplex is called the U5-AUG duplex. The other regions of the 5' UTR have been shown to play no systematic role in genomic RNA dimerization, except for a sequence located around the 3' end of a large stem-loop enclosing the primer binding site, and the well-documented SL1. Our data are consistent with a direct role for the 5' TAR in genomic RNA dimerization (possibly via a palindrome encompassing the apical loop of the 5' TAR).     

A second functional RNA domain in the 5' UTR of the Tomato bushy stunt virus genome: intra- and interdomain interactions mediate viral RNA replication

The 5' untranslated regions (UTRs) of (+)-strand RNA viruses play a variety of roles in the reproductive cycles of these infectious agents. Tomato bushy stunt virus (TBSV) belongs to this class of RNA virus and is the prototype member of the genus Tombusvirus. Previous studies have demonstrated that a T-shaped domain (TSD) forms in the 5' half of the TBSV 5' UTR and that it plays a central role in viral RNA replication. Here we have extended our structure-function analysis to the 3' half of the 5' UTR. Investigation of this region in the context of a model viral replicon (i.e., a TBSV-derived defective interfering [DI] RNA) revealed that this segment contains numerous functionally relevant structural features. In vitro solution structure probing along with comparative and computer-aided RNA secondary structure analyses predicted the presence of a simple stem loop (SL5) followed by a more complex downstream domain (DSD). Both structures were found to be essential for efficient DI RNA accumulation when tested in a plant protoplast system. For SL5, maintenance of the base of its stem was the principal feature required for robust in vivo accumulation. In the DSD, both helical and unpaired regions containing conserved sequences were necessary for efficient DI RNA accumulation. Additionally, optimal DI RNA accumulation required a TSD-DSD interaction mediated by a pseudoknot. Modifications that reduced accumulation did not appreciably affect DI RNA stability in vivo, indicating that the DSD and SL5 act to facilitate viral RNA replication.     

Mutagenic analysis of the 3' cis-acting elements of the rubella virus genome

Thermodynamically predicted secondary structure analysis of the 3'-terminal 305 nucleotides (nt) of the rubella virus (RUB) genome, a region conserved in all RUB defective interfering RNAs, revealed four stem-loop (SL) structures; SL1 and SL2 are both located in the E1 coding region, while SL3 and SL4 are within the 59-nt 3' untranslated region (UTR) preceding the poly(A) tract. SL2 is a structure shown to interact with human calreticulin (CAL), an autoantigen potentially involved in RUB RNA replication and pathogenesis. RNase mapping indicated that SL2 and SL3 are in equilibrium between two conformations, in the second of which the previously proposed CAL binding site in SL2, a U-U bulge, is not formed. Site-directed mutagenesis of the 3' UTR with a RUB infectious clone, Robo302, revealed that most of the 3' UTR is required for viral viability except for the 3'-terminal 5 nt and the poly(A) tract, although poly(A) was rapidly regenerated during subsequent replication. Maintenance of the overall SL3 structure, the 11-nt single-stranded sequence between SL3 and SL4, and the sequences forming SL4 were all important for viral viability. Studies on the interaction between host factors and the 3' UTR showed the formation of three RNA-protein complexes by gel mobility shift assay, and UV-induced cross-linking detected six host protein species, with molecular masses of 120, 80, 66, 55, 48, and 36 kDa, interacting with the 3' UTR. Site-directed mutagenesis of SL2 by nucleotide substitutions showed that maintenance of SL2 stem rather than the U-U bulge was critical in CAL binding since mutants having the U-U bulge base paired had a similar binding activity for CAL as the native structure whereas mutants having the SL2 stem destabilized had much lower binding activity. However, all of these mutations gave rise to viable viruses when introduced into Robo302, indicating that binding of CAL to SL2 is independent of viral viability.     

Pestivirus NS3 (p80) protein possesses RNA helicase activity.

The pestivirus bovine viral diarrhea virus (BVDV) p80 protein (referred to here as the NS3 protein) contains amino acid sequence motifs predictive of three enzymatic activities: serine proteinase, nucleoside triphosphatase, and RNA helicase. We have previously demonstrated that the former two enzymatic activities are associated with this protein. Here, we show that a purified recombinant BVDV NS3 protein derived from baculovirus-infected insect cells possesses RNA helicase activity. BVDV NS3 RNA helicase activity was specifically inhibited by monoclonal antibodies to the p80 protein. The activity was dependent on the presence of nucleoside triphosphate and divalent cation, with a preference for ATP and Mn2+. Hydrolysis of the nucleoside triphosphate was necessary for strand displacement. The helicase activity required substrates with an un-base-paired region on the template strand 3' of the duplex region. As few as three un-base-paired nucleotides were sufficient for efficient oligonucleotide displacement. However, the enzyme did not act on substrates having a single-stranded region only to the 5' end of the duplex or on substrates lacking single-stranded regions altogether (blunt-ended duplex substrates), suggesting that the directionality of the BVDV RNA helicase was 3' to 5' with respect to the template strand. The BVDV helicase activity was able to displace both RNA and DNA oligonucleotides from RNA template strands but was unable to release oligonucleotides from DNA templates. The possible role of this activity in pestivirus replication is discussed.     

Sequences at the 3' untranslated region of bamboo mosaic potexvirus RNA interact with the viral RNA-dependent RNA polymerase

The 3' untranslated region (UTR) of bamboo mosaic potexvirus (BaMV) genomic RNA was found to fold into a series of stem-loop structures including a pseudoknot structure. These structures were demonstrated to be important for viral RNA replication and were believed to be recognized by the replicase (C.-P. Cheng and C.-H. Tsai, J. Mol. Biol. 288:555-565, 1999). Electrophoretic mobility shift and competition assays have now been used to demonstrate that the Escherichia coli-expressed RNA-dependent RNA polymerase domain (Delta 893) derived from BaMV open reading frame 1 could specifically bind to the 3' UTR of BaMV RNA. No competition was observed when bovine liver tRNAs or poly(I)(C) double-stranded homopolymers were used as competitors, and the cucumber mosaic virus 3' UTR was a less efficient competitor. Competition analysis with different regions of the BaMV 3' UTR showed that Delta 893 binds to at least two independent RNA binding sites, stem-loop D and the poly(A) tail. Footprinting analysis revealed that Delta 893 could protect the sequences at loop D containing the potexviral conserved hexamer motif and part of the stem of domain D from chemical cleavage.     

Complex signals in the genomic 3' nontranslated region of bovine viral diarrhea virus coordinate translation and replication of the viral RNA

The genomes of positive-strand RNA viruses strongly resemble cellular mRNAs. However, besides operating as a messenger to generate the virus-encoded proteins, the viral RNA serves also as a template during replication. A central issue of the viral life cycle, the coordination of protein and RNA synthesis, is yet poorly understood. Examining bovine viral diarrhea virus (BVDV), we report here on the role of the variable 3'V portion of the viral 3' nontranslated region (3'NTR). Genetic studies and structure probing revealed that 3'V represents a complex RNA motif that is composed of synergistically acting sequence and structure elements. Correct formation of the 3'V motif was shown to be an important determinant of the viral RNA replication process. Most interestingly, we found that a proper conformation of 3'V is required for accurate termination of translation at the stop-codon of the viral open reading frame and that efficient termination of translation is essential for efficient replication of the viral RNA. Within the viral 3'NTR, the complex 3'V motif constitutes also the binding site of recently characterized cellular host factors, the so-called NFAR proteins. Considering that the NFAR proteins associate also with the 5'NTR of the BVDV genome, we propose a model where the viral 3'NTR has a bipartite functional organization: The conserved 3' portion (3'C) is part of the nascent replication complex; the variable 5' portion (3'V) is involved in the coordination of the viral translation and replication. Our data suggest the accuracy of translation termination as a sophisticated device determining viral adaptation to the host.     

Interaction between the cellular protein eEF1A and the 3'-terminal stem-loop of West Nile virus genomic RNA facilitates viral minus-strand RNA synthesis

RNase footprinting and nitrocellulose filter binding assays were previously used to map one major and two minor binding sites for the cell protein eEF1A on the 3'(+) stem-loop (SL) RNA of West Nile virus (WNV) (3). Base substitutions in the major eEF1A binding site or adjacent areas of the 3'(+) SL were engineered into a WNV infectious clone. Mutations that decreased, as well as ones that increased, eEF1A binding in in vitro assays had a negative effect on viral growth. None of these mutations affected the efficiency of translation of the viral polyprotein from the genomic RNA, but all of the mutations that decreased in vitro eEF1A binding to the 3' SL RNA also decreased viral minus-strand RNA synthesis in transfected cells. Also, a mutation that increased the efficiency of eEF1A binding to the 3' SL RNA increased minus-strand RNA synthesis in transfected cells, which resulted in decreased synthesis of genomic RNA. These results strongly suggest that the interaction between eEF1A and the WNV 3' SL facilitates viral minus-strand synthesis. eEF1A colocalized with viral replication complexes (RC) in infected cells and antibody to eEF1A coimmunoprecipitated viral RC proteins, suggesting that eEF1A facilitates an interaction between the 3' end of the genome and the RC. eEF1A bound with similar efficiencies to the 3'-terminal SL RNAs of four divergent flaviviruses, including a tick-borne flavivirus, and colocalized with dengue virus RC in infected cells. These results suggest that eEF1A plays a similar role in RNA replication for all flaviviruses.     

Assignment of the Multifunctional NS3 Protein of Bovine Viral Diarrhea Virus during RNA Replication: an In Vivo and In Vitro Study

Studies on the replication of the pestivirus bovine viral diarrhea virus (BVDV) were considerably facilitated by the recent discovery of an autonomous subgenomic BVDV RNA replicon (DI9c). DI9c comprises mainly the untranslated regions of the viral genome and the coding region of the nonstructural proteins NS3, NS4A, NS4B, NS5A, and NS5B. To assess the significance of the NS3-associated nucleoside triphosphatase/helicase activity during RNA replication and to explore other functional features of NS3, we generated a repertoire of DI9c derivatives bearing in-frame mutations in different parts of the NS3 coding unit. Most alterations resulted in deficient replicons, several of which encoded an NS3 protein with an inhibited protease function. Three lesions permitted replication, though at a lower level than that of the wild-type RNA, i.e., replacement of the third position of the DEYH helicase motif II by either T or F and an insertion of four amino acid residues in the C-terminal part of NS3. While polyprotein proteolysis was found to be almost unaffected in these latter replicons, in vitro studies with the purified mutant NS3 proteins revealed a significantly impaired helicase activity for the motif II substitutions. NS3 with a DEFH motif, moreover, showed a significantly lower ATPase activity. In contrast, the C-terminal insertion had no negative impact on the ATPase/RNA helicase activity of NS3. All three mutations affected the synthesis of both replication products-negative-strand intermediate and progeny positive-strand RNA-in a symmetric manner. Unexpectedly, various attempts to rescue or enhance the replication capability of nonfunctional or less functional DI9c NS3 derivatives, respectively, by providing intact NS3 in trans failed. Our experimental data thus demonstrate that the diverse enzymatic activities of the NS3 protein-in particular the ATPase/RNA helicase-play a pivotal role even during early steps of the viral replication pathway. They may further indicate the C-terminal part of NS3 to be an important functional determinant of the RNA replication process.     

An internally located RNA hairpin enhances replication of Tomato bushy stunt virus RNAs

Defective interfering (DI) RNAs of Tomato bushy stunt virus (TBSV), a plus-sense RNA virus, comprise four conserved noncontiguous regions (I through IV) derived from the viral genome. Region III, a 70-nucleotide-long sequence corresponding to a genomic segment located 378 nucleotides upstream of the 3' terminus of the genome, has been found to enhance DI RNA accumulation by approximately 10-fold in an orientation-independent manner (D. Ray and K. A. White, Virology 256:162-171, 1999). In this study, a more detailed structure-function analysis of region III was conducted. RNA secondary-structure analyses indicated that region III contains stem-loop structures in both plus and minus strands. Through deletion analyses of a DI RNA, a primary determinant of region III activity was mapped to the 5'-proximal 35-nucleotide segment. Compensatory-type mutational analyses showed that a stem-loop structure in the minus strand of this subregion was required for enhanced DI RNA replication. The same stem-loop structure was also found to function in a position-independent manner in a DI RNA (albeit at reduced levels) and to be important for efficient accumulation within the context of the TBSV genome. Taken together, these observations suggest that the 5'-proximal segment of region III is a modular RNA replication element that functions primarily through the formation of an RNA hairpin structure in the minus strand.     

Genetic analysis of the pestivirus nonstructural coding region: defects in the NS5A unit can be complemented in trans

The functional analysis of molecular determinants which control the replication of pestiviruses was considerably facilitated by the finding that subgenomic forms of the positive-strand RNA genome of BVDV (bovine viral diarrhea virus) are capable of autonomous replication in transfected host cells. The prototype replicon, BVDV DI9c, consists of the genomic 5' and 3' untranslated regions and a truncated open reading frame (ORF) encoding mainly the nonstructural proteins NS3, NS4A, NS4B, NS5A, and NS5B. To gain insight into which of these proteins are essential for viral replication and whether they act in cis or in trans, we introduced a large spectrum of in-frame mutations into the DI9c ORF. Tests of the mutant RNAs in terms of their replication capacity and their ability to support translation and cleavage of the nonstructural polyprotein, and whether defects could be rescued in trans, yielded the following results. (i) RNA replication was found to be dependent on the expression of each of the DI9c-encoded mature proteins NS3 to NS5B (and the known associated enzymatic activities). In the same context, a finely balanced molar ratio of the diverse proteolytic processing products was indicated to be crucial for the formation of an active catalytic replication complex. (ii) Synthesis of negative-strand intermediate and progeny positive-strand RNA was observed to be strictly coupled with all functional DI9c ORF derivatives. NS3 to NS5B were hence suggested to play a pivotal role even during early steps of the viral replication pathway. (iii) Mutations in the NS3 and NS4B units which generated nonfunctional or less functional RNAs were determined to be cis dominant. Likewise, lethal alterations in the NS4A and NS5B regions were invariably noncomplementable. (iv) In surprising contrast, replication of functional and nonfunctional NS5A mutants could be clearly enhanced and restored, respectively. In summary, our data provide initial insights into the organization of the pestivirus replication machinery.     

Structure-function analysis of the 3' stem-loop of hepatitis C virus genomic RNA and its role in viral RNA replication

Previous studies indicate that the 3' terminal 46 nt of the RNA genome of hepatitis C virus (HCV) are highly conserved among different viral strains and essential for RNA replication. Here, we describe a mutational analysis of the 3' terminal hairpin (stem-loop I) that is putatively formed by this sequence and demonstrate its role in replication of the viral RNA. We show that single base substitutions within the 6-nt loop at positions adjacent to the stem abrogate replication of a subgenomic RNA, whereas substitutions in the three apical nucleotides were well tolerated without loss of replication competence. Single point mutations were also well tolerated within the middle section of the duplex, but not at the penultimate nucleotide positions near either end of the stem. However, complementary substitutions at the -19 and -28 positions (from the 3' end) restored replication competence, providing strong evidence for the existence of the structure and its involvement in RNA replication. This was confirmed by rescue of replicating RNAs from mutants containing complementary 10-nt block substitutions at the base of the stem. Each of these RNAs contained an additional U at the 3' terminus. Further experiments indicated a strong preference for U at the 3' terminal position (followed in order by C, A, and G), and a G at the -2 position. These features of stem-loop I are likely to facilitate recognition of the 3' end of the viral RNA by the viral RNA replicase.     

trans regulation of cap-independent translation by a viral subgenomic RNA

Many positive-strand RNA viruses generate 3'-coterminal subgenomic mRNAs to allow translation of 5'-distal open reading frames. It is unclear how viral genomic and subgenomic mRNAs compete with each other for the cellular translation machinery. Translation of the uncapped Barley yellow dwarf virus genomic RNA (gRNA) and subgenomic RNA1 (sgRNA1) is driven by the powerful cap-independent translation element (BTE) in their 3' untranslated regions (UTRs). The BTE forms a kissing stem-loop interaction with the 5' UTR to mediate translation initiation at the 5' end. Here, using reporter mRNAs that mimic gRNA and sgRNA1, we show that the abundant sgRNA2 inhibits translation of gRNA, but not sgRNA1, in vitro and in vivo. This trans inhibition requires the functional BTE in the 5' UTR of sgRNA2, but no translation of sgRNA2 itself is detectable. The efficiency of translation of the viral mRNAs in the presence of sgRNA2 is determined by proximity to the mRNA 5' end of the stem-loop that kisses the 3' BTE. Thus, the gRNA and sgRNA1 have "tuned" their expression efficiencies via the site in the 5' UTR to which the 3' BTE base pairs. We conclude that sgRNA2 is a riboregulator that switches off translation of replication genes from gRNA while permitting translation of structural genes from sgRNA1. These results reveal (i) a new level of control of subgenomic-RNA gene expression, (ii) a new role for a viral subgenomic RNA, and (iii) a new mechanism for RNA-mediated regulation of translation.     

An RNA domain within the 5' untranslated region of the tomato bushy stunt virus genome modulates viral RNA replication

The terminal half of the 5' untranslated region (UTR) in the (+)-strand RNA genome of tomato bushy stunt virus was analyzed for possible roles in viral RNA replication. Computer-aided thermodynamic analysis of secondary structure, phylogenetic comparisons for base-pair covariation, and chemical and enzymatic solution structure probing were used to analyze the 78 nucleotide long 5'-terminal sequence. The results indicate that this sequence adopts a branched secondary structure containing a three-helix junction core. The T-shaped domain (TSD) formed by this terminal sequence is closed by a prominent ten base-pair long helix, termed stem 1 (S1). Deletion of either the 5' or 3' segment forming S1 (coordinates 1-10 or 69-78, respectively) in a model subviral RNA replicon, i.e. a prototypical defective interfering (DI) RNA, reduced in vivo accumulation levels of this molecule approximately 20-fold. Compensatory-type mutational analysis of S1 within this replicon revealed a strong correlation between formation of the predicted S1 structure and efficient DI RNA accumulation. RNA decay studies in vivo did not reveal any notable changes in the physical stabilities of DI RNAs containing disrupted S1s, thus implicating RNA replication as the affected process. Further investigation revealed that destabilization of S1 in the (+)-strand was significantly more detrimental to DI RNA accumulation than (-)-strand destabilization, therefore S1-mediated activity likely functions primarily via the (+)-strand. The essential role of S1 in DI RNA accumulation prompted us to examine the 5'-proximal secondary structure of a previously identified mutant DI RNA, RNA B, that lacks the 5' UTR but is still capable of low levels of replication. Mutational analysis of a predicted S1-like element present within a cryptic 5'-terminal TSD confirmed the importance of the former in RNA B accumulation. Collectively, these data support a fundamental role for the TSD, and in particular its S1 subelement, in tombusvirus RNA replication.     

Specific interaction of hepatitis C virus protease/helicase NS3 with the 3'-terminal sequences of viral positive- and negative-strand RNA

The hepatitis C virus (HCV)-encoded protease/helicase NS3 is likely to be involved in viral RNA replication. We have expressed and purified recombinant NS3 (protease and helicase domains) and Delta pNS3 (helicase domain only) and examined their abilities to interact with the 3'-terminal sequence of both positive and negative strands of HCV RNA. These regions of RNA were chosen because initiation of RNA synthesis is likely to occur at or near the 3' untranslated region (UTR). The results presented here demonstrate that NS3 (and Delta pNS3) interacts efficiently and specifically with the 3'-terminal sequences of both positive- and negative-strand RNA but not with the corresponding complementary 5'-terminal RNA sequences. The interaction of NS3 with the 3'-terminal negative strand [called 3'(-) UTR(127)] was specific in that only homologous (and not heterologous) RNA competed efficiently in the binding reaction. A predicted stem-loop structure present at the 3' terminus (nucleotides 5 to 20 from the 3' end) of the negative-strand RNA appears to be important for NS3 binding to the negative-strand UTR. Deletion of the stem-loop structure almost totally impaired NS3 (and Delta pNS3) binding. Additional mutagenesis showed that three G-C pairs within the stem were critical for helicase-RNA interaction. The data presented here also suggested that both a double-stranded structure and the 3'-proximal guanosine residues in the stem were important determinants of protein binding. In contrast to the relatively stringent requirement for 3'(-) UTR binding, specific interaction of NS3 (or Delta pNS3) with the 3'-terminal sequences of the positive-strand RNA [3'(+) UTR] appears to require the entire 3'(+) UTR of HCV. Deletion of either the 98-nucleotide 3'-terminal conserved region or the 5' half sequence containing the variable region and the poly(U) and/or poly(UC) stretch significantly impaired RNA-protein interaction. The implication of NS3 binding to the 3'-terminal sequences of viral positive- and negative-strand RNA in viral replication is discussed.