Sequence Requirements for Viral RNA Replication and VPg Uridylylation Directed by the Internal cis-Acting Replication Element (cre) of Human Rhinovirus Type 14
ABSTRACT
Until recently, the cis-acting signals required for replication of picornaviral RNAs were believed to be restricted to the 5′ and 3′ noncoding regions of the genome. However, an RNA stem-loop in the VP1-coding sequence of human rhinovirus type 14 (HRV-14) is essential for viral minus-strand RNA synthesis (K. L. McKnight and S. M. Lemon, RNA 4:1569-1584, 1998). The nucleotide sequence of the apical loop of this internal cis-acting replication element (cre) was critical for RNA synthesis, while secondary RNA structure, but not primary sequence, was shown to be important within the duplex stem. Similar cres have since been identified in other picornaviral genomes. These RNA segments appear to serve as template for the uridylylation of the genome-linked protein, VPg, providing the VPg-pUpU primer required for viral RNA transcription (A. V. Paul et al., J. Virol. 74:10359-10370, 2000). Here, we show that the minimal functional HRV-14 cre resides within a 33-nucleotide (nt) RNA segment that is predicted to form a simple stem-loop with a 14-nt loop sequence. An extensive mutational analysis involving every possible base substitution at each position within the loop segment defined the sequence that is required within this loop for efficient replication of subgenomic HRV-14 replicon RNAs. These results indicate that three consecutive adenosine residues (nt 2367 to 2369) within the 5′ half of this loop are critically important for cre function and suggest that a common RNNNAARNNNNNNR loop motif exists among the cre sequences of enteroviruses and rhinoviruses. We found a direct, positive correlation between the capacity of mutated cres to support RNA replication and their ability to function as template in an in vitro VPg uridylylation reaction, suggesting that these functions are intimately linked. These data thus define more precisely the sequence and structural requirements of the HRV-14 cre and provide additional support for a model in which the role of the cre in RNA replication is to act as template for VPg uridylylation.
The family Picornaviridae is one of the largest groups of nonenveloped, single-stranded, positive-sense RNA viruses. It comprises nine distinct genera, of which five (the enteroviruses, rhinoviruses, aphthoviruses, cardioviruses, and hepatoviruses) are important human and animal pathogens (23). These genera are distinguished from each other by differences in the proteins they express, but the RNA genomes of each have a similar organization, with a lengthy 5′ nontranslated region (5′ NTR) followed by a single, large open reading frame encoding a single polyprotein, a 3′ NTR, and a genetically templated 3′-terminal poly(A) tail (28). Since it is positive sense, the virion RNA acts directly as message, programming the translation of the polyprotein following its release into the cytoplasm. Virion RNA and synthetic genome-length RNA derived from recombinant cDNA clones are thus infectious when transfected into permissive cells, giving rise to virus particles and subsequent rounds of virus replication (22).
An unusual feature of the genomic RNA of picornaviruses is that it lacks the 5′-terminal cap structure present in most eucaryotic mRNAs and is instead covalently linked to a small, virally encoded protein, VPg (3, 8). An internal ribosomal entry site located within the 5′ NTR directs the cap-independent translation of the polyprotein (6, 19), so that the cellular translational machinery thus bypasses the 5′ end of the genome. The polyprotein contains three major functional segments, defined in part by the order of cleavage events that occur during its processing by one or more viral proteases (26). In the case of the enteroviruses and rhinoviruses, the most N-terminal segment, P1, contains four capsid proteins, VP4, VP2, VP3, and VP1, while the P2 and P3 segments are comprised of nonstructural proteins involved in protein maturation and RNA replication. These include 2Apro, 2B, 2C, 3A, 3B (VPg), 3Cpro, and 3Dpol and their functional precursors, 2BC, 3AB, and 3CDpro (15).
In broad outline, the replication of picornaviral RNAs follows a two-step process. Upon entry into host cells and uncoating of the virus, the incoming positive-strand genome is transcribed into complementary minus-strand RNA by a replicase complex, the catalytic unit of which, 3Dpol, uses a uridylylated form of VPg (VPg-pUpU) as a primer (18). This minus-strand RNA then serves as a template for the production of new, plus-strand, progeny genomes. Although the basic steps of replication are well known, relatively little is understood about the details of these processes. One of the important, yet incompletely answered questions is how the viral replicase specifically selects only viral RNA species for amplification in these reactions, because the 3′-terminal poly(A) sequence of the genomic RNA is indistinguishable from the 3′ termini of cellular mRNAs.
Until recently, it was generally believed that the 5′ and 3′ NTRs of picornaviral RNA contain the cis-acting signals necessary for the initiation of viral RNA replication (28). The specificity of genome amplification was thought to result from interactions of the replicase complex with unique cis-acting signals located in the 5′ and 3′ NTRs (1, 13, 14, 21, 22, 25, 27). However, McKnight and Lemon (11, 12) demonstrated that an internal cis-acting replication element (cre) was necessary for the initiation of RNA synthesis during replication of human rhinovirus type 14 (HRV-14). Although this replication element is located in the long open reading frame, within the segment encoding the VP1 capsid protein, the role of the cre in replication is dependent upon its RNA structure and not its protein-coding capacity (12). Mutational analysis and computer folding algorithms suggested that the cre forms a complex stem-loop structure within the positive-strand RNA that is required for initiation of minus-strand RNA synthesis (Fig. 1) (12).
Following the identification of the HRV-14 cre, similar internal replication signals were identified within the open reading frames of Theiler's virus, a cardiovirus (9), poliovirus type 1 (PV-1) (5), and, more recently, HRV-2 (4). The latter observations suggest that an internally located cre may be a common feature of the RNA replication schemes of all picornaviruses. The RNA segments comprising these putative cres can be folded into relatively simple stem-loop structures, but the elements differ in terms of their primary nucleotide sequence as well as their location within the open reading frame. In cardioviruses, the cre is located in the VP2 region (9). In contrast, the PV-1 cre is not located in the P1 capsid region, but in the 2C (P2) region (5), while the HRV-2 cre is located within the 2A coding sequence (4). Studies by Paul et al. (17) indicate that the cre acts as the primary template for uridylylation of VPg by the 3Dpol polymerase in vitro. In addition, a recent mutational analysis of the poliovirus polymerase indicates that amino acid residues on the surface of the protein that are essential for uridylylation of VPg are also involved in the interaction of the polymerase with the membrane-bound 3AB precursor protein (10). Taken together, these data suggest that the cre plays a critical role in bringing viral RNA into the replication complex and in initiating VPg uridylylation, the first step in the process of viral RNA replication.
Here, we describe experiments aimed at better defining both the sequence and structural requirements for cre function during the replication of HRV-14 RNA. We show that the fully functional cre resides within a 33-nucleotide (nt) RNA segment that is predicted to form a simple stem-loop structure. By introducing single-base substitutions at each position within the loop sequence, we have determined which nucleotides are essential for replication and which nucleotide substitutions are tolerated without significant degradation of cre function. We further show that the ability of individual mutant cre sequences to support RNA replication is closely correlated with their ability to serve as template for the uridylylation of VPg in an in vitro reaction.
MATERIALS AND METHODS
Cells.
HeLa cells were obtained from the American Type Culture Collection and maintained in Dulbecco's modified Eagle's medium (DMEM; Gibco/BRL) with 5% fetal bovine serum (FBS).
HRV-14 RNA replication assay.
The replication of HRV-14 RNA was monitored by measurement of luciferase expression in HeLa cells transfected with ΔP1LucCRE RNA (12). ΔP1LucCRE is a subgenomic RNA replicon in which most of the P1 segment of the HRV-14 genome has been replaced by an in-frame insertion of the firefly luciferase coding sequence (Fig. 1). Replicon transcripts were obtained by T7 polymerase-mediated transcription (T7 MEGAscript; Ambion) of pΔP1LucCRE and related plasmids after linearization at the unique MluI site downstream of the poly(A) tail. The integrity and abundance of transcript RNAs were determined by agarose gel analysis. These transcripts were electroporated into HeLa cells as previously described (12). Briefly, 5 × 106 cells were electroporated with 20 μg of RNA by using a GenePulser II electroporation apparatus (Bio-Rad) with the pulse controller unit set at 980 V and 25 μF and with maximum resistance. The cells were subsequently seeded into a six-well plate and cultured in DMEM with 5% FBS at 34°C until processed for luciferase assays. Luciferase activity was measured in cell lysates with the Promega luciferase assay kit. Briefly, 125 μl of lysis buffer (Promega) was added to each well of a six-well plate, and the lysate was stored at −70°C until assay. Luciferase quantitation was carried out as described by the supplier with a TD-20/20 luminometer (Turner Designs).
Site-directed mutagenesis of the cre loop segment.
Site-directed mutagenesis was facilitated by subcloning a DNA fragment containing the cre sequence from the plasmid pΔP1LucCRE (12) into the vector pLITMUS 29 (New England Biolabs) at XhoI and NcoI restriction endonuclease sites. Mutations were introduced with the QuickChange site-directed mutagenesis kit (Stratagene) with a series of degenerate oligonucleotide primers containing a variety of single-base substitutions within the region representing the cre loop segment. The products of these mutagenesis reactions were characterized by DNA sequencing. Selected mutants were digested with XhoI and NheI, and a small DNA fragment was inserted into pΔP1LucCRE at the XhoI and NheI sites flanking the cre region (Fig. 1). All substitutions were confirmed by DNA sequencing.
Construction of cre mutants with in-frame deletions.
To define the minimal functional cre sequence, PCR mutagenesis was used to create in-frame deletions upstream and downstream of the previously defined apical stem-loop of the cre, as shown in Fig. 1. The oligonucleotide primers for PCRs leading to the construction of the deletion mutant D1 were D1+ (CCGCTCGAGGGCTTAGGTGATG) and D1− (CTAGCTAGCTGGACCAGATGAGATTG); for mutant D2, the primers were D2+ (CCGCTCGAGGATGAATTAGAAGAAG) and D2− (CTAGCTAGCGATTGAGGCCACCG); and for mutant D3, the primers were D3+ (CCGCTCGAGGTCATCGTTGAG) and D3− (CTAGCTAGCGGCCACCGTCTGTTTCG). The resulting PCR products were digested with XhoI and NheI and ligated into pΔP1LucCRE as described above. All regions subjected to PCR amplification were sequenced to verify the deletions and ensure that no undesired mutations were introduced.
Computer-based prediction of RNA structures.
RNA secondary structure was predicted by using the MFOLD program with the Zuker energy minimization algorithm (Genetics Computer Group, University of Wisconsin).
In vitro VPg uridylylation assay.
Wild-type and mutant HRV-14 cre templates (representing HRV14 nt 2318 to 2413) were prepared by T7 transcription of DNA fragments amplified from the cognate plasmid DNAs by PCR with the oligonucleotide primers E-T7-5′CRE (CCGGAATTCTAATACGACTCACTATAGGGGCACTCACTGAAGGC) and N-3′CRE (CTAGCTAGCTGTGTGTTTTG). Transcription reactions were carried out with the T7 Megashortscript kit (Ambion), and transcript RNAs were purified by phenol-chloroform extraction and ethanol precipitation prior to use in the uridylylation reactions.
The uridylylation of a synthetic PV-1 VPg oligopepetide with wild-type and mutant HRV-14 cre RNA templates was assayed as previously described (17). Briefly, reaction mixtures (20 μl) containing 50 mM HEPES buffer (pH 7.5), 8% glycerol, 3.5 mM magnesium acetate, 0.5 μg of HRV-14 cre transcript RNA, 2 μg of synthetic PV-1 VPg, 1 μg of purified PV-1 3D polymerase, 0.75 μCi of [α-32P]UTP (3,000 Ci/mmol; Dupont-NEN), 10 μM unlabeled UTP, and 0.5 μg of PV-1 3CDpro[3Cpro(H40A)] were incubated at 34°C for 60 min. Reaction products were analyzed by Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (Bio-Rad) with 13.5% polyacrylamide. Gels were dried and subjected to autoradiography. Reaction products were quantitated by PhosphorImager (Molecular Dynamics) analysis.
RESULTS
RNA structure requirement for HRV-14 cre function.
To analyze the effects of nucleotide deletions and substitutions within the previously identified 96-nt HRV-14 cre segment (Fig. 1), we utilized a replication-competent, subgenomic RNA replicon, ΔP1LucCRE (Fig. 1). This replicon was constructed from an HRV-14 infectious clone (7, 12) by replacing the P1, capsid protein-coding segment of the genome with the firefly luciferase coding sequence. Retention of the 21-nt sequence encoding the carboxy-terminal 7 amino acids of VP1 allowed for recognition and processing of the polyprotein at the VP1/2Apro cleavage site by the 2Apro proteinase. The 96-nt cre sequence (nt 2318 to 2413) was inserted in frame at unique restriction sites engineered between the 3′ end of the luciferase sequence and the residual VP1 sequence (12) (Fig. 1).
Replication of the subgenomic RNA transcribed from this plasmid was monitored by measuring luciferase expression following transfection of the RNA into cultured HeLa cells. As shown in Fig. 2A, transfection of ΔP1LucCRE RNA gave rise to an increase in luciferase activity between 3 and 24 h. In contrast, transfection of a similar quantity of subgenomic RNA transcribed from ΔP1LucXN, which is identical to ΔP1LucCRE, but lacks any cre sequence (Fig. 1), resulted in no increase in luciferase activity beyond the minimal amount expressed at 3 h following transfection (Fig. 2A). In the presence of 2 mM guanidine, a potent inhibitor of HRV-14 RNA replication, neither ΔP1LucCRE nor ΔP1LucXN gave rise to an increase of luciferase activity beyond that present 3 h after transfection (Fig. 2A). Since there were no differences in luciferase activity between ΔP1LucCRE and ΔP1LucXN at 3 h following transfection in either the presence or absence of guanidine, this early low level of luciferase activity can be considered to be due to the translation of the input RNAs in the absence of any RNA amplification. In contrast, the subsequent increase in luciferase activity between 3 and 24 h, which was observed only with ΔP1LucCRE and only in the absence of guanidine, reflects viral RNA replication. In support of this interpretation, the level of luciferase expression following expression of ΔP1LucCRE RNA was shown previously to closely follow its RNA replication kinetics as assessed by RNase protection assay (12).
As indicated above, previous studies by McKnight and Lemon mapped the cre to a 96-nt-long segment within the VP1 coding region of HRV-14 that is capable of forming a complex stem-loop structure within the positive-strand HRV-14 RNA (12) (Fig. 1). A mutational analysis confirmed the necessity for duplex formation within the apical stem of this structure. However, the requirement for the lower stem was less clearly documented, because only one of two mutant HRV-14 replicons with a 4-nt substitution in the lower stem was significantly impaired in replication (12). These data suggest that the lower duplex structure may not be essential for cre function. Thus, to better define the 5′ and 3′ boundaries of the minimal HRV-14 cre, we created a series of in-frame deletions within the previously identified 96-nt cre segment in the background of the replication-competent ΔP1LucCRE subgenomic replicon. As shown in Fig. 1, a series of double-deletion mutants were constructed (D1, D2, and D3), in which increasing lengths of sequences flanking the predicted apical stem-loop structure of the cre were removed from pΔP1LucCRE. In D1, the single-stranded regions 5′ and 3′ of the predicted complex stem-loop structure (nt 2318 to 2329 and 2405 to 2413) were deleted. In D2, the 5′ deletion was extended inward to include nt 2338 and the 3′ deletion to nt 2393, effectively removing most of the predicted distal duplex structure. In D3, the 5′ deletion was further extended to include nt 2353 and the 3′ deletion to nt 2387, removing the internal bulge-loops and leaving only a simple stem-loop structure with a 9-bp stem (Fig. 1).
RNA transcripts derived from pΔP1LucCRE and each of the three deletion mutants were electroporated into HeLa cells. Cell lysates were analyzed for luciferase activity at 3 h and 24 h following transfection, since, as shown above (Fig. 2A), an increase in luciferase expression during this period of time is indicative of RNA replication. Transfection of the three double-deletion mutants resulted in similar levels of luciferase expressed as a result of translation of the input RNA at 3 h after transfection (Fig. 2B). However, each also led to a significant increase in luciferase activity between 3 and 24 h following transfection (Fig. 2B), indicating that each RNA had undergone replication and that the minimal functional HRV-14 cre is located within the 33 nt remaining in mutant D3 (nt 2354 to 2386). In contrast, in cells transfected with ΔP1LucXN, which lacks any cre sequence (Fig. 1), luciferase activity decreased significantly between 3 and 24 h posttransfection (Fig. 2B), indicating a failure of viral RNA replication. These data indicate that the minimal functional HRV-14 cre is considerably smaller than the 96-nt structure proposed previously (12).
Nucleotide sequence required within the cre for RNA replication.
Previous studies have indicated that extensive substitutions can be made within the duplex stem of the cre without compromising replication, provided that compensatory changes are introduced to maintain base pairing within the stem (12). Several nucleotide substitutions within the loop segment, however, were found to be lethal to replication. To better define the requirements for specific nucleotide sequence within the loop, we carried out an extensive mutational analysis of this region (nt 2363 to 2376). Using site-specific mutagenesis with degenerate oligonucleotide primers, we constructed a series of pΔP1LucCRE mutants containing single-base substitutions in which every possible nucleotide substitution was created at each of the nucleotide positions within the loop (Fig. 3A). (For convenience, we refer to these nucleotide positions according to their unique last two digits [e.g., “26” rather than “2326”]in Fig. 3 and the following sections.) Four of the intended substitutions, 63U, 66U, 70U, and 75U, introduced stop codons into the reading frame of ΔP1LucCRE, and hence could not be evaluated for their effect on replication as single-base substitutions. Thus, a total of 38 mutants were evaluated, each containing a point mutation at one of the 14 bases in the loop (Fig. 3A). These mutated cre sequences were analyzed by the MFOLD program to predict whether the single-nucleotide substitutions would lead to potential secondary structure alterations. Only three of them—69U, 76U, and 76C—were suggested by MFOLD to have a significant change in secondary structure from that of the wild-type cre (Fig. 3B).
Runoff transcripts representing the wild-type replicon and replicons with single-base substitutions within the cre were electroporated into HeLa cells. Replication was monitored by measuring luciferase activity at 3 and 24 h following electroporation. The efficiency of RNA replication was expressed as the fold increase in luciferase activity at 24 h over the luciferase activity at 3 h relative to that observed with the parental HRV-14 cre (ΔP1LucCRE), which was considered to be 100%. An increase in luciferase activity that was ≥15% of the parental level was considered indicative of significant RNA replication, while a fold increase in luciferase activity <15% that observed with the parent was considered to represent failure of significant replication and only the accumulation of translation products derived from input RNA. This 15% cutoff level represents approximately 3 standard deviations below the mean fold increase in luciferase expression observed between 3 and 24 h after transfection of the parental RNA, ΔP1LucCRE, in a series of four independent experiments, each done in triplicate (average fold increase, 32.9 ± 8.7). As shown in Fig. 4, most mutants replicated at levels either significantly above or below this breakpoint.
Surprisingly, at most positions within the loop (positions 64 to 66 and 70 to 75), any base substitution resulted in a replication-competent phenotype (Fig. 4). In contrast, any substitution of the three adenosines spanning positions 67 to 69, except for 69G, a purine transition, resulted in a failure to replicate, as evidenced by a lack of increase in the level of luciferase expression following electroporation of the RNA (Fig. 4). Since the 69U substitution causes a significant perturbation in the predicted secondary structure of the cre (Fig. 3B), it is not possible to discriminate whether the lethal effect of this mutation is related to an alteration in structure or primary sequence, or both. However the failure of 69C to replicate suggests the latter, since this substitution was not predicted to have altered the structure of the loop. These results indicate that A67, A68, and (to a lesser extent) A69 are essential for cre function in supporting RNA replication.
We created additional mutants to assess the effects of base substitutions that would result in the formation of stop codons within the cre. Since the 72U substitution creates a stop codon, we introduced another substitution at the second position of the codon, 73C. 73C by itself is replication competent, as was the multiple-substitution mutant MS2, which contains both 72U and 73C substitutions (Fig. 3C and 4). We could not use this approach to evaluate the effect of 66U, which creates a UAA stop codon, since any substitution at the relevant second base position, 67, was lethal (Fig. 4).
63U and 75U also introduce stop codons into the cre sequence. The further evaluation of these substitutions was complicated by the results of other substitutions at positions 63G and 76A, which are located at the junction of the stem and the loop in the predicted structure of the wild-type cre (Fig. 3A). Some substitutions at these bases, 63A and 76G, resulted in a replication-competent phenotype (Fig. 4). In contrast, other substitutions (63C, 76C, and 76U) were lethal or resulted in mutants with marginal replication capacity (Fig. 4). Although 63G and 76A may form a noncanonical base pair in the wild-type structure (Fig. 3A), the two latter substitutions at position 76 have the potential to result in the formation of stronger base pair interactions at the base of the cre loop, 63G-76C or 63G-76U. Such base pairing could reduce the size of the loop from 14 nt to 12 nt or otherwise change its conformation (Fig. 3B). Thus, we suspected that the inhibition of replication by the 76C and 76U substitutions might be caused by alteration in the secondary or tertiary structure of the loop. To test this hypothesis, and at the same time to evaluate the effect of the remaining two substitutions that result in stop codons, 63U and 75U, we created the multiple-substitution mutant MS1 shown in Fig. 3C. This mutant contains 63U, 64U, 75U, and 76U substitutions and is both free of stop codons and not capable of base pairing between positions 63 and 76. Although the 4 base substitutions in MS1 do not change the predicted secondary structure, this mutated cre failed to support RNA replication (Fig. 4). Thus, it is not clear whether the negative effects of some single-nucleotide substitutions at positions 63 and 76 result from alterations of the secondary structure alteration, rather than changes in the primary sequence of the RNA (see Discussion). Similar to the situation at position 69A, it seems that only a purine transition is well tolerated at positions 63G and 76A.
In summary, no substitution of 67A or 68A was permissible. At the position immediately following these two adenosines, 69A, only a purine transition was acceptable. This was also the case with the 2 nt at the bottom of the loop, 63G and 76A, at which only purine transitions were tolerated. In contrast, any substitution was permitted at other positions in the loop without significantly impairing the cre function in replication. These conclusions are summarized in Fig. 5.
Capacity of mutant cre RNAs to support VPg uridylylation in vitro.
The HRV-14 cre has been shown to be an effective template for the uridylylation of the VPg proteins of either HRV-14 or PV-1 type 1 by the PV-1 3Dpol enzyme (17). This prompted us to investigate whether the capacity of the cre mutants to support replication in vivo would correlate with their ability to function in an in vitro VPg uridylylation reaction. VPg uridylylation reactions were carried out with synthetic PV-1 VPg, and recombinant PV-1 3Dpol and 3CDpro, as described previously (17). RNA transcripts representing each of the HRV-14 cre mutants were added to individual reaction mixtures. The efficiency of the reaction was measured by PhosphorImager quantification of the yield of uridylylated VPg, VPg-pU, and VPg-pUpU, as shown in Fig. 6. In each experiment, the efficiency of uridylylation was compared to that of a reaction mixture containing the wild-type HRV-14 cre RNA.
As shown in Fig. 6A, the PV-1 3Dpol enzyme efficiently directed the uridylylation of the PV-1 VPg protein in the presence of the wild-type HRV-14 cre RNA. Those mutant cre sequences that retained the ability to support RNA replication also remained active in the VPg uridylylation reaction, although with various levels of efficiency producing various amounts of VPg-pU and VPg-pUpU. A few mutants produced yields of uridylylated VPg that were comparable to or higher than that produced with the wild-type cre, such as 64C, 66C, or 66G (Fig. 6A). However, most of the mutant cre RNAs produced lower yields of the uridylylated protein (15 to 80% of the wild-type level), such as 63A, 64G, 64U, 69G, 76G, and all substitutions at positions 65 and 70 to 75 (Fig. 6).
As shown previously, each of the nucleotide substitutions at positions 67 and 68 abolished replicon amplification (Fig. 4). Importantly, these substitutions also failed to produce detectable uridylylation products (Fig. 6). The 63C, 69C, and 76C substitutions, as well as the multiple mutant MS1, each of which was also deficient in RNA replication (Fig. 4), resulted in yields of uridylylated VPg that were no more than the background level (Fig. 6). The marginally replicating mutant 76U and a nonreplicating mutant 69U (Fig. 4) produced yields of uridylylated VPg that were less than 10% of the wild-type level. There was thus a good correlation between the capacity of each mutant cre to support RNA replication and its ability to function in the in vitro uridylylation reaction, as shown schematically in Fig. 7.
DISCUSSION
To date, cres have been identified within the protein-coding sequences of viruses representing three genera of picornaviruses: rhinoviruses, enteroviruses, and cardioviruses (4, 5, 9, 11, 12). These cres are similar in that they are all located within protein-coding sequence, but function at the level of RNA. Each is predicted to form a stem-loop structure about 30 to 65 nt in length, generally involving nucleotide tracts with low P-num values supporting the likelihood that the structure is conserved and thus biologically significant (16). The available evidence suggests that these RNA structures are required within the positive strand of the RNA for efficient initiation of minus-strand RNA synthesis (12) and that they are likely to be common to the RNA replication scheme of most if not all picornaviruses.
However, there are differences in these cres that are also remarkable. First, they are located in different regions (VP1, VP2, 2A, or 2C coding region) of the genome, suggesting that cre function is not dependent on a specific location and that different viruses have evolved cres at different sites within the genome in ways conducive to their dual roles as both replication element and protein-coding sequence. Another surprising difference, considering their common roles in RNA replication, is the extent of the diversity that is evident in their primary sequences as well as predicted secondary structures. The sequence differences present in the HRV-14, PV-1, and HRV-2 cres (Fig. 8A) are even more surprising, since the HRV-14 cre is capable of substituting for the PV-1 and HRV-2 cres in VPg uridylylation reactions with PV-1 and HRV-2 enzymes (4, 17). This suggests that the PV-1 and rhinovirus enzymes involved recognize some common sequence and/or structural features in these cres. Our primary aim in this study was to identify the sequences that are critical for HRV-14 cre function and, by comparing these with other known or predicted cres, to identify those sequences and structural features that are important for RNA replication.
Since previous results suggested that the minimal functional HRV-14 cre (12) was significantly larger (96 nt) than the cres identified subsequently in other picornaviruses (4, 5), we created a series of deletion mutants in which the sequences flanking the apical loop of the structure predicted for the HRV-14 cre were progressively deleted (Fig. 1). The ability of each of these mutants to support the replication of HRV-14 RNA (Fig. 2B) indicated that the minimal functional cre resides within a 33-nt sequence that is predicted to form a simple stem-loop (nt 2354 to 2386) (Fig. 1). The HRV-14 cre is thus no larger than the cre identified in other picornaviruses.
These results indicate that only the top stem and loop sequences of the originally described 96-nt cre structure (12) are required for replication of HRV-14 RNA and that the large internal loop and lower duplex stem (see ΔP1LucCRE structure in Fig. 1) are not essential. Despite this, McKnight and Lemon (12) found that some mutations in the lower stem were lethal to replication of HRV-14 RNA. Since the sequences forming the lower stem are located outside of the region, we have shown here to contain the minimal functional cre (Fig. 1), these results are best explained by changes in the folded structure of the top loop and stem of the cre mediated by mutations in the lower stem. If this interpretation is correct, the tertiary structure of the cre loop is critically important to its ability to function in RNA replication and probably also uridylylation of VPg. In general, however, it appears that a stem of 8 to 10 bp, similar to what we have determined for the minimal HRV-14 cre, is sufficient to support cre function. Disruption of the lower part of the PV-1 cre stem (leaving 9 bp at the base of the loop) had little effect on cre function (24). In addition, the predicted cardiovirus cre structures have stems formed by 8 to 10 bp (9).
We also carried out an extensive mutational analysis of the loop of the HRV-14 cre, because previous work points to its importance in RNA replication (12). The HRV-14 cre presents a unique opportunity for such a mutational analysis, since unlike the PV-1 or HRV-2 cre, it is located within the P1 region that encodes capsid proteins that are not necessary for RNA replication (12). Thus, mutations in the HRV-14 cre do not alter the amino acid sequence of proteins that contribute to the replicase. This distinguishes the cre mutants we have studied here from many of the mutations that have been studied previously in the PV-1 or HRV-2 cres (4, 17) and has also allowed us to do a more complete analysis. The fact that we found a strong correlation between the effects of these HRV-14 mutations on VPg uridylylation in vitro and on RNA replication in vivo (Fig. 7) provides additional indirect, but nonetheless strong, evidence that the ability of the cre to function as a template for VPg uridylylation is essential for viral RNA synthesis.
We found that an AAA triplet (nt 2367 to 2369, or 67A, 68A, and 69A) located in the 5′ half of the loop and 2 nt, 63G and 76A, at the bottom of the loop, are important for cre function, while there are no requirements for specific bases at other positions (Fig. 5). Any substitution within the AAA triplet abrogated RNA replication, except for the substitution of 69A with 69G. These results are reminiscent of previous studies of the PV-1 cre, in which the mutation of either of the first two A's of the AAA triplet within the 5′ half of the loop in the PV-1 element abolished infectivity (24). Mutation of these adenosines in the HRV-2 cre also abrogated uridylylation of VPg and replication of the virus (4). This AAA triplet is part of the AAACA motif identified by Rieder et al. (24) in the PV-1 cre loop and considered to be a common feature of the cre. The mutational analysis of the HRV-14 cre shown in Fig. 4 and 6, however, demonstrates that the CA dinucleotide within the AAACA motif is not necessary for replication or uridylylation. We also found that 63G and 76A, located at the junction of the cre loop and stem, were critically important for cre activity. As with 69A, only purine transition mutations were permissible at these bases.
Alignments of the sequences around the AAA triplet in the HRV-14, PV-1, and HRV-2 cres indicate that the critical base residues we have identified in the HRV-14 cre are perfectly conserved in these other cres (Fig. 8A, boldface type). This includes the 63G and 76A bases at the junction of the HRV-14 cre loop and stem. However, in contrast to the HRV-14 loop sequence, which is predicted by MFOLD to be 14 nt in length, the homologous sequences in the PV-1 and HRV-2 cres are predicted to contain several base pairs and an internal loop and to fold differently from HRV-14 as well as each other (Fig. 8B). These structure predictions have not been tested by any physical means, however, nor by appropriate mutational analyses, so their validity is unknown. Despite the differences in the computer-predicted secondary structures, we suspect that these sequences are similarly structured in all three cres. Such structural homology would make the observation that these cres are functionally exchangeable with each other in uridylylation reactions understandable (4, 17). Alternatively, it is possible that the binding of proteins such as 3CD, 3D, and VPg to the PV-1 and HRV-2 elements in the earliest steps of the uridylylation reaction might cause a change in the conformation of the RNA, opening up internal base pairs so that the PV-1 and HRV-2 structures resemble that of the HRV-14 cre.
Gerber et al. (4) identified potential cres in two rhinoviruses that are closely related to HRV-2, HRV-16 and HRV-1b. These proposed cre sequences were located at essentially the same position in the genome as the HRV-2 cre. Again, the bases that we found to be essential for HRV-14 cre activity are conserved in these putative cres, including 63G and 76A at the junction of the loop and the stem. It is interesting to note, however, that in the proposed HRV-16 cre, the AAA triplet is replaced with AAG. We found this substitution to be functional in the HRV-14 background, both for replication and for uridylylation (Fig. 4 and 6).
Based on our mutational analysis of the HRV-14 cre and comparisons of the HRV-14 cre sequence with the cres of other viruses, we propose a common motif for the loop segment of rhinovirus and enterovirus cres that likely defines a common structure (Fig. 8B): R1NNNAAR2NNNNNNR3.
We predict that an AAR triplet will be present in the 5′ half of the loop sequence in all these replication elements. At the third base position of this triplet (R2), the base may be either A or G, but there is clearly a preference for A. This is also the case for R3, while a G is preferable at R1. R1 and R3 are located at the extreme ends of the loop and seem likely to be involved in a non-Watson-Crick base pair interaction (see below). At the remaining positions, A, C, G, or U appears to be equivalently acceptable for both replication and uridylylation, and the specific base present is likely to be determined more by the requirement for coding specific amino acids than by a requirement for preservation of cre function. However, a mutation involving the substitution of 4 contiguous bases within the loop sequence (69A through 72A) was found previously by McKnight and Lemon (12) to be lethal for replication, even though we have shown here that each of the individual base substitutions in this mutant was permissive for replication and uridylylation. It is likely that the failure of the multiple-substitution mutant to function as a cre was due to perturbation of the structure of the loop following substitution of such a large proportion of the bases within it.
All of the known cardiovirus cre sequences also possess an AAA motif that is predicted to be at least partially single stranded in these viruses. A mutation at the second A of this motif in the Theiler's virus cre was shown to be lethal for RNA replication (9). However, the cardiovirus cre appears to have a very different structure from that of the PV-1 or rhinovirus cre, because the homologous AAA triplet in Theiler's virus appears to be located within a small bulge-loop that is part of a larger hairpin structure (9). This difference in structure is consistent with the failure of the HRV-14 cre to substitute for the cardioviral cre in supporting viral replication (9), as well as the greater phylogenetic distance between the cardioviruses and these other picornaviral genera.
Studies by Paul et al. (17) and Gerber et al. (4) have shown that the PV-1 cre acts as the primary template for VPg uridylylation in vitro, a reaction that requires only synthetic VPg, UTP, purified PV-1 RNA, PV-1 RNA polymerase 3Dpol, and Mg2+. Mutations that abolish the ability of the PV-1 or HRV-2 cre to act as template for VPg uridylylation in vitro also eliminate their ability to support viral RNA replication in vivo (4, 17). Substitution of the AAA triplet in the HRV-2 cre with CAA also led to a change in the specificity of the nucleotidylylation reaction, with the covalent addition of guanine to VPg leading to VPg-pG (4). These observations suggest that the AAA triplet functions as a template for the nucleotidylylation of VPg by using a slide-back mechanism in which the most 5′ adenosine of the AAA triplet templates the nucleotide to be linked to VPg. This indicates that the conserved AAR2 residues within the loop of the rhinovirus and enterovirus cres are likely to be located on the surface of the folded RNA structure. The conserved 63G and 76A residues (Fig. 8B) have the potential to form a non-Watson-Crick closing pair at the base of the loop. Recent studies of synthetic RNA aptomers suggest that the presence of a GA closing pair significantly influences the structure of the adjacent RNA loop and may have a critical role in determining the ability of the loop to form stable loop-loop interactions (2). While there is no evidence that the cre loop is involved in a loop-loop interaction, we speculate that the conserved 63G and 76A residues form such a closing pair and that the presence of this closing pair contributes in an important way to the structure of the cre loop that is required for proper presentation of the AAR2 triplet as the template for uridylylation.
FIG. 1.
FIG. 2.
FIG. 3.
FIG. 4.
FIG. 5.
FIG. 6.
FIG. 7.
FIG. 8.
Acknowledgments
This work was supported by grants from the National Institute of Allergy and Infectious Diseases (RO1-AI140282) (S.M.L.) and the McLaughlin Endowment of the University of Texas Medical Branch (Y.Y.).
REFERENCES
1.
Andino, R., G. E. Rieckhof, P. L. Achacoso, and D. Baltimore. 1993. PV-1 RNA synthesis utilizes an RNP complex formed around the 5′-end of viral RNA. EMBO J. 12 : 3587-3598.
2.
Duconge, F., C. Di Primo, and J. J. Toulme. 2000. Is a closing “GA pair” a rule for stable loop-loop RNA complexes? J. Biol. Chem. 275 : 21287-21294.
3.
Flanegan, J. B., R. F. Petterson, V. Ambros, N. J. Hewlett, and D. Baltimore. 1977. Covalent linkage of a protein to a defined nucleotide sequence at the 5′-terminus of virion and replicative intermediate RNAs of PV-1. Proc. Natl. Acad. Sci. USA 74 : 961-965.
4.
Gerber, K., E. Wimmer, and A. V. Paul. 2001. Biochemical and genetic studies of the initiation of human rhinovirus 2 RNA replication: identification of a cis-replicating element in the coding sequence of 2Apro. J. Virol. 75 : 10979-10990.
5.
Goodfellow, I., Y. Chaudhry, A. Richardson, J. Meredith, J. W. Almond, W. Barclay, and D. J. Evans. 2000. Identification of a cis-acting replication element within the poliovirus coding region. J. Virol. 74 : 4590-4600.
6.
Jang, S. K., H.-G. Krausslich, M. J. H. Nicklin, G. M. Duke, A.C. Palmenberg, and E. Wimmer. 1988. A segment of the 5′ nontranslated region of encephalomyocarditis virus RNA directs internal entry of ribosomes during in vitro translation. J. Virol. 62 : 2636-2643.
7.
Lee, W.-M., S. S. Monroe, and R. R. Rueckert. 1993. Role of maturation cleavage in infectivity of picornaviruses: activation of an infectosome. J. Virol. 67 : 2110-2122.
8.
Lee, Y. F., A. Nomoto, B. M. Detjen, and E. Wimmer. 1977. A protein covalently linked to poliovirus genome RNA. Proc. Natl. Acad. Sci. USA 74 : 59-63.
9.
Lobert, P.E., N. Escriou, J. Ruelle, and T. Michiels. 1999. A coding RNA sequence acts as a replication signal in cardioviruses. Proc. Natl. Acad. Sci. USA 96 : 11560-11565.
10.
Lyle, J. M., A. Clewell, K. Richmond, O. C. Richards, D. A. Hope, S. C. Schultz, and K. Kirkegaard. 2002. Similar structural basis for membrane localization and protein priming by an RNA-dependent RNA polymerase. J. Biol. Chem. 277 : 16324-16331.
11.
McKnight, K. L., and S. M. Lemon. 1996. Capsid coding sequence is required for efficient replication of human rhinovirus 14 RNA. J. Virol. 70 : 1941-1952.
12.
McKnight, K. L., and S.M. Lemon. 1998. The rhinovirus type 14 genome contains an internally located RNA structure that is required for viral replication. RNA 4 : 1569-1584.
13.
Melchers, W. J. G., J. G. J. Hoenderop, H. J. B. Slot, C. W. A. Pleij, E. V. Pilipenko, V. I. Agol, and J. M. D. Galama. 1997. Kissing of the two predominant hairpin loops in the coxsackie B virus 3′ untranslated region is the essential structural feature of the origin of replication required for negative-strand RNA synthesis. J. Virol. 71 : 686-696.
14.
Mirmomeni, M. H., P. J. Hughes, and G. Stanway. 1997. An RNA tertiary structure in the 3′ untranslated region of enteroviruses is necessary for efficient replication. J. Virol. 71 : 2363-2370.
15.
Palmenberg, A. C. 1990. Proteolytic processing of picornaviral polyprotein. Annu. Rev. Microbiol. 44 : 603-623.
16.
Palmenberg, A. C., and J.-Y. Sgro. 1997. Topological organization of picornaviral genomes: statistical prediction of RNA structural signals. Semin. Virol. 8 : 231-241.
17.
Paul, A. V., E. Rieder, D. W. Kim, J. H. Van Boom, and E. Wimmer. 2000. Identification of an RNA hairpin in poliovirus RNA that serves as the primary template in the in vitro uridylylation of VPg. J. Virol. 74 : 10359-10370.
18.
Paul, A. V., J. H. Van Boom, D. Filippov, and E. Wimmer. 1998. Protein-primed RNA synthesis by purified PV-1 RNA polymerase. Nature 393 : 280-284.
19.
Pelletier, J., and N. Sonenberg. 1988. Internal initiation of translation of eukaryotic mRNA directed by a sequence derived from PV-1 RNA. Nature 334 : 320-325.
20.
Pilipenko, E. V., S. V. Maslova, A. N. Sinyakov, and V. I. Agol. 1992. Towards identification of cis-acting elements involved in the replication of enterovirus and rhinovirus RNAs: a proposal for the existence of tRNA-like terminal structures. Nucleic Acids Res. 20 : 1739-1745.
21.
Pilipenko, E. V., V. Poperechny, S. V. Maslova, W. J. G. Melchers, H. J. Bruins Slot, and V. I. Agol. 1996. Cis-element, oriR, involved in the initiation of (−) strand PV-1 RNA: a quasi-globular multi-domain RNA structure maintained by tertiary (′kissing') interactions. EMBO J. 15 : 5428-5436.
22.
Racaniello, V. R., and D. Baltimore. 1981. Cloned PV-1 complementary DNA is infectious in mammalian cells. Science 214 : 916-919.
23.
Regenmortel, M. H. V., C. M. Fauquet, D. L. Bishop, E. B. Carstens, M. K. Estes, S. M. Lemon, J. Maniloff, M. A. Mayo, D. J. McGeogh, R. B. Pringle, and R. B. Wickner. 2000. Virus taxonomy: the classification and nomenclature of viruses. The Seventh Report of the International Committee on Taxonomy of Viruses. Academic Press, San Diego, Calif.
24.
Rieder, E., A. V. Paul, D.W. Kim, J. H. Van Boom, and E. Wimmer. 2000. Genetic and biochemical studies of poliovirus cis-acting replication element cre in relation to VPg uridylylation. J. Virol. 74 : 10371-10380.
25.
Rohll, J. B., D. H. Moon, D. J. Evans, and J. W. Almond. 1995. The 3′ untranslated region of picornavirus RNA: features required for efficient genome replication. J. Virol. 69 : 7835-7844.
26.
Rueckert, R. R., and E. Wimmer. 1984. Systematic nomenclature of picornavirus proteins. J. Virol. 50 : 957-959.
27.
Todd, S., J. H. C. Nguyen, and B. L. Semler. 1995. RNA-protein interactions directed by the 3′ end of human rhinovirus genomic RNA. J. Virol. 69 : 3605-3614.
28.
Wimmer, E., C. U. T. Hellen, and X. Cao. 1993. Genetics of PV-1. Annu. Rev. Genet. 27 : 353-436.
Information & Contributors
Information
Published In
Copyright
© 2002 American Society for Microbiology.
History
Received: 24 January 2002
Accepted: 30 April 2002
Published online: 1 August 2002
Contributors
Metrics & Citations
Metrics
Note:
- For recently published articles, the TOTAL download count will appear as zero until a new month starts.
- There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.
- Citation counts come from the Crossref Cited by service.
Citations
Citation
2002. Sequence Requirements for Viral RNA Replication and VPg Uridylylation Directed by the Internal cis-Acting Replication Element (cre) of Human Rhinovirus Type 14. J Virol 76:.
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. For an editable text file, please select Medlars format which will download as a .txt file. Simply select your manager software from the list below and click Download.