Skip to main content
NCBI home page
Elsevier - PMC COVID-19 Collection logoLink to Elsevier - PMC COVID-19 Collection
. 2003 Jul 17;313(2):481–491. doi: 10.1016/S0042-6822(03)00387-8

Frameshift mutations in infectious cDNA clones of Citrus tristeza virus: a strategy to minimize the toxicity of viral sequences to Escherichia coli

Tatineni Satyanarayana a, Siddarame Gowda a, María A Ayllón a, William O Dawson a,*
PMCID: PMC7125997  PMID: 12954215

Abstract

The advent of reverse genetics revolutionized the study of positive-stranded RNA viruses that were amenable for cloning as cDNAs into high-copy-number plasmids of Escherichia coli. However, some viruses are inherently refractory to cloning in high-copy-number plasmids due to toxicity of viral sequences to E. coli. We report a strategy that is a compromise between infectivity of the RNA transcripts and toxicity to E. coli effected by introducing frameshift mutations into “slippery sequences” near the viral “toxicity sequences” in the viral cDNA. Citrus tristeza virus (CTV) has cDNA sequences that are toxic to E. coli. The original full-length infectious cDNA of CTV and a derivative replicon, CTV-ΔCla, cloned into pUC119, resulted in unusually limited E. coli growth. However, upon sequencing of these cDNAs, an additional uridinylate (U) was found in a stretch of U’s between nts 3726 and 3731 that resulted in a change to a reading frame with a stop codon at nt 3734. Yet, in vitro produced RNA transcripts from these clones infected protoplasts, and the resulting progeny virus was repaired. Correction of the frameshift mutation in the CTV cDNA constructs resulted in increased infectivity of in vitro produced RNA transcripts, but also caused a substantial increase of toxicity to E. coli, now requiring 3 days to develop visible colonies. Frameshift mutations created in sequences not suspected to facilitate reading frame shifting and silent mutations introduced into oligo(U) regions resulted in complete loss of infectivity, suggesting that the oligo(U) region facilitated the repair of the frameshift mutation. Additional frameshift mutations introduced into other oligo(U) regions also resulted in transcripts with reduced infectivity similarly to the original clones with the +1 insertion. However, only the frameshift mutations introduced into oligo(U) regions that were near and before the toxicity region improved growth and stability in E. coli. These data demonstrate that, when hosts are sufficiently susceptible for infection by transcripts of reduced specific infectivity, introduction of frameshift mutations at “slippery sequences” near toxic regions of viral cDNAs can be used as an additional strategy to clone recalcitrant viral sequences in high-copy-number plasmids for reverse genetics.

Introduction

Reverse genetics, the ability to produce specific mutations followed by examination of phenotype, revolutionized the study of RNA viruses. The first successful attempts initiated infection of cells by inoculation with bacterial plasmids containing the viral cDNA inserts Taniguchi et al 1978, Racaniello and Baltimore 1981. However, production of in vitro RNA transcripts that were identical to the virion RNA (except for designed mutations and perhaps a few extra nucleotides at either terminus) allowed genetic manipulation using the standard inoculation procedures previously in use for virion RNAs (Ahlquist et al., 1984). The current ideal is to clone a full-length viral cDNA in a high-copy-number bacterial plasmid behind a promoter for a commercially available RNA polymerase. Most viruses with relatively small RNA genomes have been amenable for such a genetic system. However, some viruses are inherently difficult to clone due to the instability of the viral cDNA, often because the viral sequences are toxic to the bacterium Boyer and Haenni 1994, Lai 2000.

The cloning difficulties of some recalcitrant viral genomes have been surmounted by a variety of methods. In vitro ligation of individually cloned cDNA fragments followed by in vitro transcription and transfection was used to overcome the cloning difficulties of Yellow fever virus, Mouse hepatitis virus, Transmissible gastroenteritis virus (TGEV), and Beet necrotic yellow vein virus RNA2 Quillet et al 1989, Rice et al 1989, Yount et al 2000, Yount et al 2002. The disadvantages of this approach, though, are relatively low levels of full-length cDNA and reduced levels of RNA transcripts compared to that of in vitro transcripts from a complete clone in one plasmid. Another approach is to reduce the toxicity of viral sequences in high-copy-number plasmids by inserting introns into the viral sequences. Recently, cDNA clones of TGEV, Japanese encephalitis flavivirus, Pea seedborne mosaic virus, and Plum pox virus were stabilized by this method Johansen 1996, López-Moya and Garcia 2000, Yamshchikov et al 2001, González et al 2002. This strategy, however, requires cytoplasmic viruses to have a nuclear phase and also requires that the viral sequences not have endogenous intron sequences, which has been suggested as a possible problem of large cytoplasmic RNA viruses (Lai, 2000). In fact, Tobacco mosaic virus (TMV) appears to be an example of such a virus with an endogenous intron, based on the fact that the infectivity of 35S:cDNA constructs are several orders of magnitude less than RNA transcripts and often are infectious in only certain plant species (W.O. Dawson, unpublished data). Moreover, most cells of tobacco plants with transgenes of the infectious cDNA of TMV behind the 35S promoter do not become infected from the transgene, although they do become infected via virus movement from other cells (Turpen, 1992). Yet another approach is to avoid the use of high-copy-number plasmids. The reticent RNA genome of TGEV could be propagated in Escherichia coli when cloned into a bacterial artificial chromosome, a low-copy-number plasmid (Almazán et al., 2000). Other approaches avoid the use of bacteria. Vaccinia virus was used as a vector to clone the coronaviruses, Avian infectious bronchitis virus (IBV) and Human coronavirus Casais et al 2001, Thiel et al 2001. Although this approach allowed the cloning of the full-length coronavirus cDNA, the viral cDNA in the vaccinia virus vector is not very amenable for mutational analysis, requiring multiple cycles of ligation and recombination. Here we report an additional strategy to allow cloning of recalcitrant viral cDNAs in high-copy-number plasmids in E. coli that can be added to the repertoire of strategies available.

Citrus tristeza virus (CTV), a member of the Closterovirus genus of the Closteroviridae family, has a single-stranded positive-sense RNA genome of ∼20.0 kb that is organized into 12 open reading frames (ORFs) (Fig. 1A) Pappu et al 1994, Karasev et al 1995. ORF 1a encodes a polyprotein containing two papain-like protease domains, plus methyltransferase-like (MT) and helicase-like (HEL) domains, and a large interdomain region (IDR) between the MT and HEL domains (Fig. 1A). ORF 1b encodes an RNA-dependent RNA polymerase-like domain that is thought to be translated by a +1 frameshift from ORF 1a (Karasev et al., 1995). The ten 3′ ORFs are expressed by 3′-coterminal sgRNAs (Hilf et al., 1995) that encode an assembly gene block of four proteins (Satyanarayana et al., 2000), a protein that controls asymmetry of CTV RNAs (p23) (Satyanarayana et al., 2002a), and five proteins of unknown function.

Fig. 1.

Fig. 1

Open in a new tab

Replication of wild-type CTV and replicon with and without a +1 insertion at nt 3732 in Nicotiana benthamiana mesophyll protoplasts. (A) Schematic diagram of the genome organization of wild-type CTV and derivative replicon showing the putative domains of papain-like proteases (PRO), methyltransferase (MT), helicase (HEL), a large interdomain region (IDR), and an RNA-dependent RNA polymerase (RdRp). The ORFs (boxes) with their numbers and corresponding translation products are also indicated. HSP70h, HSP70 homolog; CPm, minor coat protein; CP, major coat protein; (2/11), fusion of ORFs 2 and 11 with the deletion of internal sequence. The fortuitous frameshift found at nt 3732 (nt in bold) in CTV9 (a) and CTV-ΔCla (c) was corrected in CTV9R (b) and CTV-ΔClaR (d). (B) Northern blot hybridization analysis of total RNA isolated from protoplasts transfected with CTV9 (a), CTV9R (b), CTV-ΔCla (c), and CTV-ΔClaR (d) at 3 and 4 days postinoculation (dpi). The position of genomic and subgenomic (sg) RNAs corresponding to ORFs 2–11 are indicated by arrowheads and arrows, respectively. The blots were hybridized with 3′ 900-nt positive-stranded RNA-specific digoxigenin-labeled probe.

The development of infectious cDNA clones of CTV was difficult due to the large size of the genomic RNA and its toxicity to E. coli (Satyanarayana et al., 1999). We were able to maintain the plasmids containing the viral cDNA in only one strain of E. coli, which after transformation grew very slowly into minute colonies. However, we found that this level of growth was due to a fortuitous frameshift in the CTV cDNA that allowed its cloning by reducing the toxicity to E. coli. The corrected cDNA exhibited substantially greater toxicity to E. coli, now requiring 3 days to see visible colonies. Examination of other frameshift mutations demonstrated that 6–7 U’s functioned as “slippery” frameshift sites, reducing toxicity to E. coli if positioned appropriately near the toxicity region in the viral cDNA. Thus, purposely introduced frameshift mutations into oligo(U) regions in a toxic viral cDNA can provide a compromise between infectivity and toxicity to E. coli. If hosts cells are susceptible enough to allow a reduction in infectivity, frameshift mutations can be inserted in the viral cDNA to allow cloning in high-copy-number plasmids of E. coli.

Results

Characteristics of the original infectious cDNA clones of CTV

The full-length infectious cDNA clones of CTV, pCTV9, and its derivatives were constructed in pUC119 behind the SP6 RNA polymerase promoter and grown in E. coli strain JM109 (Satyanarayana et al., 1999). CTV-ΔCla contains the 5′-nontranslated region (NTR) plus ORFs 1a and 1b and the first 105 nts of ORF 2 (nts 1–11,005) fused to the 3′ 768 nts of the genomic RNA that includes part of the p23 ORF and the 3′-NTR (nts 18,526–19,293). The cDNA of this small replicon with the ten 3′ genes (7.5 kb) deleted is much easier to manipulate than the full-length virus (Satyanarayana et al., 2002b). The in vitro generated RNA transcripts were biologically active as evidenced by their replication in Nicotiana benthamiana protoplasts and synthesis of sgRNAs (Fig. 1, CTV9 and CTV-ΔCla), and production of viable CTV9 virions that passaged with a specific infectivity similar to that of the parental wild-type virus Gowda et al 2001, Satyanarayana et al 1999, Satyanarayana et al 2000, Satyanarayana et al 2001, Satyanarayana et al 2002a, Satyanarayana et al 2002b. The RNA transcripts of full-length CTV and the replicon replicated similarly in citrus protoplasts (data not shown). The recombinant virus had a phenotype identical to that of the parental wild-type virus in citrus trees (Satyanarayana et al., 2001).

pCTV9 and pCTV-ΔCla were equally toxic to E. coli, but grew enough on LB agar-amp plates for minute colonies to be picked for propagation after 24 h, compared to 1.8-mm-diameter colonies of the vector pUC119 after overnight incubation at 37°C (Fig. 2). The cDNA clones grew marginally in 2xYT liquid medium, but failed to grow in LB liquid medium. From numerous E. coli strains examined, the original cDNA clones of CTV in pUC119 were stable only in JM109.

Fig. 2.

Fig. 2

Open in a new tab

E. coli colonies on LB-agar plates transformed with pCTV9 (with a +1 insertion at nt 3732) and pCTV9R (without a frameshift at nt 3732) at 1, 2, and 3 days posttransformation. The parental plasmid pUC119 and the cDNA clone of TMV in pUC19 (pTMV) (Dawson et al., 1988) were used as controls for comparison of the size of E. coli colonies to evaluate the toxicity of CTV sequences.

Sequence analysis of pCTV9

To compare the sequence of the full-length infectious cDNA clone with that of the consensus sequence from the native quasispecies of parental CTV T36 isolate (GenBank Accession No. NC_001661), both strands of pCTV9 were sequenced completely (GenBank Accession No. AY170468). Although we found numerous nucleotide differences, we observed an additional “U” inserted at nt 3732 into a homopolymeric run of 6 U’s that resulted in a shift to a reading frame with a stop codon at nt 3734. A similar frameshift mutation was also found in the derivative replicon pCTV-ΔCla. However, the sequence of the progeny virion RNA of CTV9 from infected protoplasts or citrus plants did not contain the additional nt at position 3732 that occurred in the cDNA clones, suggesting that the frameshift mutation present in the original cDNA clone was repaired either during the in vitro transcription or during replication in the protoplasts.

Repair of the +1 insertion in original cDNA clones resulted in increased replication, but increased toxicity to E. coli

The +1 insertion found at nt 3732 of the cDNA clones was corrected in pCTV9 and pCTV-ΔCla by exchanging a RT-PCR amplified product between the unique NcoI (nt 3685) and Bsu36I (nt 4430) restriction endonuclease sites creating pCTV9R and pCTV-ΔClaR (Fig. 1A). Replication of the full-length and replicon transcripts from the original and corrected cDNA clones was compared in N. benthamiana protoplasts. The RNA transcripts from the repaired clones accumulated genomic and sgRNAs about 30 to 40-fold more than those from the original cDNAs (Fig. 1B, compare a and b, and c and d). We then examined the number of protoplasts infected in the presence or absence of a frameshift mutation by introducing green flourescent protein (GFP) ORF into CTV using mutants CTV-GFP-p23 (with a +1 insertion at nt 3732) and CTVR-GFP-p23 (no frameshift at nt 3732) (Fig. 3A). The number of protoplasts infected with CTVR-GFP-p23 that exhibited fluorescence under UV light was approximately 18- to 30-fold more than those inoculated with CTV-GFP-p23 (Fig. 3C), again with the accumulation of approximately 30 to 40-fold more genomic and sgRNAs than those of CTV-GFP-p23 (Fig. 3B).

Fig. 3.

Fig. 3

Open in a new tab

Expression of GFP in N. benthamiana protoplasts transfected with CTV with and without a +1 insertion at nt 3732. (A) Schematic diagram of CTV-GFP-p23. The CP ORF was precisely exchanged with GFP ORF in CTV-CP/p23 (Satyanarayana et al., 2002a). CTV-GFP-p23 contained a +1 insertion at nt 3732 (nt in bold) (a), and the +1 insertion was corrected in CTVR-GFP-p23 (b). (B) Northern blot analysis of accumulation of positive-stranded genomic, and GFP and p23 sgRNAs from N. benthamiana protoplasts transfected with CTV-GFP-p23 and CTVR-GFP-p23 at 3 and 4 dpi using a 3′ positive-stranded RNA-specific riboprobe. The genomic and sgRNAs are indicated with an arrowhead and arrows, respectively. (C) A representative field of protoplasts expressing GFP transfected with CTVR-GFP-p23 or CTV-GFP-p23 at 4 dpi at approximately the same density viewed by a flourescent microscope.

The original CTV clones were toxic to E. coli, resulting in substantially reduced colony sizes (Fig. 2). Bacteria transformed with the repaired clones pCTV9R or pCTV-ΔClaR grew much slower than those containing the original clones, producing visible colonies only 3 days after transformation with ligation mixtures. However, E. coli transformed with purified plasmid DNAs of pCTV9R or pCTV-ΔClaR produced visible colonies slightly earlier. The yields of plasmid DNA from E. coli strain JM109 harboring pCTV9 and pCTV9R after overnight growth in 2xYT liquid medium were 2.5–3.0 and 1.5–2.0 μg of DNA per milliliter of culture, respectively, compared to 8–10 μg DNA per milliliter of culture for pTMV (Dawson et al., 1988). Thus, although the original CTV clones with the frameshift mutation reduced the growth of E. coli and plasmid DNA yields, the correction of the +1 insertion in pCTV9R further reduced the growth of E. coli to a level below the range of expectations for routine laboratory work. Although not quantified here, the degree of difficulty of ligating and isolating intact cDNA clones was proportionally increased with the increased toxicity of the repaired constructs.

Examination of other frameshift mutations in ORF 1a

We expected frameshift mutations in CTV to be lethal. We were surprised that cDNA clones of CTV with a frameshift mutation would produce transcripts with an appreciable amount of infectivity in protoplasts. We further examined this phenomenon by introducing additional frameshift mutations into CTV-ΔClaR. We introduced +1 insertion mutation in other regions of ORF 1a to examine whether the frameshift mutations would result in a viable virus similar to that of original cDNA clones that contained a +1 insertion at 3732 nt. Two mutations were introduced separately at nt 4079 in the IDR (GCC AAG CUU UGU AAG, frameshift nt in bold) and at nt 8390 in the HEL domain (AGG AAG CUU AGC UGC, frameshift mutation in bold), creating ΔClaR-M1 and ΔClaR-M2, respectively (Fig. 4A, e and h). The in vitro RNA transcripts from these mutants failed to replicate to detectable levels in protoplasts (Fig. 4B, e and h).

Fig. 4.

Fig. 4

Open in a new tab

Replication of insertion (+1 and +2) and deletion (−1) mutants in ORF 1a of CTV in N. benthamiana protoplasts. (A) The schematic diagram of CTV replicon, CTV-ΔClaR, containing ORF 1a and 1b plus portions of ORFs 2 and 11 (a). Selected unique restriction sites used to generate mutants in CTV-ΔClaR are indicated above the diagram. The +1 insertion found at nt 3732 (nt in bold) in CTV-ΔCla (b) is shown below the CTV-ΔClaR. The expanded view of ORF 1a is shown below CTV-ΔClaR with the location of +1 or +2 insertion, and −1 deletion mutations in ORF 1a (c–h). The inserted nucleotides are shown in bold. The nucleotide in bold and underlined was deleted in ΔClaR-M6 (g). The different domains of ORF 1a are as outlined in the legend for Fig. 1A. (B) Northern blot analysis of accumulation of positive-stranded genomic and sgRNAs from N. benthamiana protoplasts transfected with insertion or deletion mutants at 3 and 4 dpi. The blots were hybridized with a 3′ positive-stranded RNA-specific riboprobe. The position of the genomic and sgRNAs are indicated by an arrowhead and an arrow, respectively.

Examination of frameshifts in oligo(U) stretches

In contrast to the frameshifts examined above, the +1 insertion found in the original infectious cDNA clones that allowed infectivity and greatly reduced the toxicity of the cDNA in E. coli was in a run of 6 U’s, which is known to be a “slippery sequence” associated with frameshifting in other viral systems Wilson et al 1988, Brierley et al 1992. However, frameshifting also has been correlated with downstream structures Jacks et al 1988, Chamorro et al 1992. It is possible that the 6 U’s at 3732 had an adventitious accessory structure downstream, and the frameshifting at this position was unusual. To examine whether other U-rich regions of CTV would similarly function as slippery-sequence frameshifting signals to recover infectivity and reduce toxicity in E. coli, we created frameshifts in other oligo(U) sites. A run of 6 U’s occurs at four different places in ORF 1a at nts 1103–1108, 3726–3731, 4957–4962, and 7078–7083. Mutations to alter the reading frame of ORF 1a in each of the U-rich regions were created (Fig. 4A). The plasmids were evaluated for toxicity to E. coli, and the RNA transcripts were evaluated for replication in protoplasts.

A +1 insertion mutation was introduced at nt 1109 in the L-ProI region to obtain ΔClaR-M3 (Fig. 4A, c). A +2 insertion mutation was introduced at nt 3732 in the MT domain of ORF 1a to obtain ΔClaR-M4 (Fig. 4A, d). These mutants replicated in protoplasts and produced genomic and sgRNAs at reduced levels, 30- to 50-fold lower when compared to that of CTV-ΔClaR, but with approximately comparable levels to that of CTV-ΔCla, which contained a +1 insertion at nt 3732 (Fig. 4B, b–d). Similarly, a +1 insertion was introduced at the 6 U’s at nt 4963 in the 5′ half of IDR to obtain ΔClaR-M5, and an adenylate was deleted after the 6 U’s at nt 7084 in the 3′ half of IDR to create ΔClaR-M6 (Fig. 4A, f and g). Mutant ΔClaR M6 replicated at levels comparable with that of the original CTV-ΔCla, while mutant ΔClaR-M5 replicated slightly less. Both accumulated 30- to 50-fold less viral RNAs than that of the wild-type CTV-ΔClaR (Fig. 4B, f and g). These results suggested that the homopolymeric stretch of U’s in ORF 1a are shift-prone sequences.

We examined the requirements of U-rich regions as shift-prone sequences by introducing silent mutations plus a +1 insertion to disrupt the U-run of the original U-rich region between nts 3726 and 3732 (UUU UUU U changed to UUC UUC U), resulting in mutant CTV-ΔCla-M7 (Fig. 5A, b). As a control, similar silent mutations were introduced in CTV-ΔClaR between nts 3726 and 3731 without the reading frame mutation (UUU UUU A changed to UUC UUC A), resulting in CTV-ΔClaR-M8 (Fig. 5A, d). Mutant CTV-ΔCla-M7 failed to replicate to detectable levels in protoplasts, suggesting that the homopolymeric stretch of U’s facilitated frameshifting (Fig. 5B, b). As expected, CTV-ΔClaR-M8 replicated approximately similarly to that of CTV-ΔClaR, demonstrating that the silent mutations did not affect the replication (Fig. 5B, d).

Fig. 5.

Fig. 5

Open in a new tab

The oligo (U) stretch in ORF 1a of CTV facilitated frameshifting in vivo. (A) The schematic diagram of CTV-ΔCla showing the sequence around nt 3732 containing an additional nt “U” (in bold) (a), the codons UUU UUU in CTV-ΔCla encoding phenylalanine were changed to UUC UUC without changing the amino acid sequence (nts in bold and italics) to create CTV-ΔCla-M7 (b). As a control the condons UUU UUU in CTV-ΔClaR (corrected clone: without a +1 U insertion) (c) were also changed to UUC UUC (nts in bold and italics) in CTV-ΔClaR-M8 (d). The domains of ORF 1a are as outlined in the legend for Fig. 1A. (B) Northern blot analysis of accumulation of positive-stranded RNAs from protoplasts transfected with CTV-ΔCla and CTV-ΔClaR and their silent mutants at 3 and 4 dpi. The blots were hybridized with a 3′ positive-stranded RNA-specific riboprobe. The genomic and sgRNAs are indicated with an arrowhead and an arrow, respectively.

Effect of frameshift mutations in ORF 1a on toxicity to E. coli

E. coli transformed with intact, supercoiled plasmids containing the original cDNA clones (pCTV9 and pCTV-ΔCla) with the +1 insertion mutation produced visible colonies after overnight incubation, whereas 2–3 days were required to see visible colonies with plasmids containing the repaired clones CTV9R and CTV-ΔClaR (Fig. 2, Fig. 6), suggesting that the frameshift mutation present in the original cDNA clones reduced the toxicity to E. coli. Since the original full-length (CTV9) and replicon (CTV-ΔCla) cDNA clones were equally toxic to E. coli (the repaired clones equally more toxic) (Fig. 6), the toxic region appeared to be located within ORFs 1a or 1b. The frameshift at nt 3732 reduced the toxicity of the original clones, suggesting that the toxicity domain was between nts 3732 and 11011. This location is supported by the observation that, when constructing cDNA clones for assembly of the full-length cDNA clone, we were able to clone the sequences between NcoI (nt 3685) and BstEII (nt 7775) only in the reverse orientation in pUC119 (pT36-Nco-Bst; Satyanarayana et al., 1999).

Fig. 6.

Fig. 6

Open in a new tab

Effect of insertion (+1 and +2) and deletion (−1) mutations in ORF 1a of CTV on E. coli strain JM109 colony size on LB-agar plate. The purified plasmid DNAs of pCTV9, pCTV9R, pCTV-ΔCla, and pCTV-ΔClaR, and frameshift mutants were used to transform E. coli. The location of frameshift mutations introduced in ORF 1a are indicated in upper right corner of the graph. The domains of ORF 1a are as described in the legend for Fig. 1A. The data represent mean and standard deviation of 25–30 colonies at 1 and 2 days posttransformation. NV, colonies not visible.

We examined the effects of the other frameshift mutations on toxicity to E. coli. The purified plasmids containing the frameshift mutations were used to transform E. coli strain JM109, and bacterial growth on LB agar plates was compared with that of E. coli transformed with pCTV9, pCTV-ΔCla, pCTV9R, pCTV-ΔClaR, pUC119, and pTMV (Dawson et al., 1988). Mutants with frameshift mutations between nts 3732 and 7084 (pΔClaR-M1, -M4, -M5, -M6, pCTV9, and pCTV-ΔCla) produced colonies of 0.2–0.9 mm diameter at 1 day posttransformation (dpt) and 0.9–1.6 mm diameter at 2 dpt, respectively (Fig. 2, Fig. 6), compared to 1.5–1.8 and 2.6–2.7 mm diameter colonies containing pUC119 and pTMV at 1 and 2 dpt, respectively. However, we failed to observe visible colonies under the microscope with the pΔClaR-M3 (with a frameshift mutation at nt 1109), pΔClaR-M2 (with a frameshift at nt 8390), pCTV9R, or pCTV-ΔClaR after overnight incubation at 37°C; however, the colonies grew to 0.2–0.3 mm diameter after 2 days of incubation (Fig. 6). Thus, the frameshift mutations in ORF 1a around the IDR (pΔClaR-M1, -M4, -M5, -M6) minimized the toxicity to E. coli, while the mutations introduced before (pΔClaR-M3) or after (pΔClaR-M2) failed to minimize the toxicity (Fig. 6), suggesting that the toxicity region was located between nts 3732 and 8390.

Discussion

We found that frameshifts within U stretches in CTV were repaired during transcription or replication, resulting in infection of approximately 2–3% as many protoplasts as the repaired transcripts. Oligo(U) stretches have been identified as slippery sequences that are involved in frameshifting of other viruses Wilson et al 1988, Brierley et al 1992, Honda et al 1995, Kim et al 2001. However, frameshifting also is often associated with downstream structures in the RNA Jacks et al 1988, Brierley et al 1992, Chamorro et al 1992, Vickers and Ecker 1992, Chen et al 1995, Marczinke et al 1998, Barry and Miller 2002. In HIV, a run of 6 U’s is required for efficient −1 frameshifting to produce the gag–pol proteins with an extended run of U’s resulting in slightly increased levels of frameshifting Wilson et al 1988, Honda et al 1995. However, Parkin et al. (1992) reported that a stem-loop structure 3′ of the HIV −1 shift site was important for wild-type levels of frameshifting in vivo. Nevertheless, downstream structures appeared not to be needed in the observed repair of the frameshift mutations after replication in N. benthamiana protoplasts. We did not identify any common structures by examining the sequences with the MFOLD program (data not shown). Also, since we examined all of the 6 U-regions, none of which are thought to naturally be involved in frameshifting, there is no expectation that all would conserve some accessory structure needed for frameshifting.

Slippery sequences identified in retrovirus frameshifting include AAAAAAC (Mouse mammary tumor virus); AAAUUUA (Rous sarcoma virus); UUUUUUA (HIV-1 and HIV-2); and UUUAAAC in coronavirus frameshifting (IBV) Jacks et al 1987, Jacks et al 1988, Wilson et al 1988, Brierley et al 1992. Additionally, numerous other slippery sequences have been identified Brierley et al 1992, Marczinke et al 1998, Kim et al 2001 and some of them occur throughout the CTV genome. We did not examine whether other potentially slippery sequences effect frameshifting in CTV. However, we examined six oligo(U) sites that allowed frameshifting when a mutation was created to destroy the reading frame. It would be expected that each of these sites also shift to an incorrect reading frame in the absence of mutations. If there are additional slippery sequences scattered throughout the CTV genome that also occasionally shift the reading frame, it is fortunate that sufficient full-length translation product is produced to allow efficient replication. Yet it is possible that slippery sequences serve as a survival mechanism for large RNA viruses. Viral RNA polymerases are known to be error prone (Drake and Holland, 1999) without an error repair mechanism, defining viruses as pseudospecies of sequence variants (Domingo et al., 1995). It has been argued that recombination and selection are the viral mechanisms to repair errors. Slippery sequences could serve as an additional repair mechanism. If a mutation occurs upstream of a slippery sequence, a frameshift could repair the mistake if the intervening altered amino acids are tolerated. Perhaps there is an advantage for CTV to have slippery sequences scattered throughout ORF 1a. This could be one of the reasons some CTV isolates have been observed to have a high degree of evolutionary stasis (Albiach-Martí et al., 2000).

The frameshifts inserted into the oligo(U) regions of CTV created a compromise between infectivity and toxicity of the cDNA in high-copy-number plasmids in E. coli. The number of protoplasts that became infected with RNA transcripts from the frameshifted clone was reduced by 30- to 50-fold when compared to that of the corrected clone. However, the toxicity in E. coli was greatly reduced. Bacterial colonies were visible after 1 day, even though they were minute, compared to 2–3 days with the repaired clone. The progeny virus from the frameshifted clone was repaired; it did not contain the original frameshift caused by the inserted nucleotide and had a phenotype identical to the parental wild-type virus (Satyanarayana et al., 2001). Thus, the frameshift mutation appeared to have no negative impact on the value of the clones other than the initial reduction in infectivity. Developing an infectious, full-length cDNA clone of CTV was difficult. It is unlikely that we would have obtained an infectious full-length clone without a fortuitous frameshift mutation in the original full-length cDNA clone.

These results suggest that purposely inserting frameshifts at slippery sequences of viral cDNAs that are toxic to E. coli could be an additional strategy for cloning these viruses in convenient high-copy-number plasmids. The frameshift provides a compromise: a reduction of infectivity for a decrease in toxicity in E. coli. If the host system is susceptible enough to provide infection with a couple of magnitudes of decrease in infectivity, the frameshift should allow growth of an infectious clone in E. coli. If the host cells are even more susceptible, it is possible that two frameshifts could be used. As is evident from this work, all frameshifts in slippery sequences will not minimize the toxicity to E. coli. The frameshift must be positioned to prevent expression of the toxicity domain, similar to the insertion of introns for the same purpose. We examined only stretches of U’s, but if the frameshifting occurred during translation, it is likely that other identified slippery sequences would also work. Since the repair of the frameshift mutations was due to in vitro transcription or translation in protoplasts, neither of which are virus-specific functions, the purposeful introduction of frameshift mutations into frameshift-prone sequences in front of toxic regions should work for any recalcitrant positive-stranded RNA virus. Thus, this strategy provides an additional approach in the virologist’s repertoire for developing reverse-genetic systems for difficult viruses.

Materials and methods

Plasmid construction

The assembly of a full-length infectious cDNA clone of CTV T36, pCTV9, and a replicon, pCTV-ΔCla, behind the SP6 RNA polymerase promoter in pUC119 was reported previously (Satyanarayana et al., 1999). The +1 insertion mutation found at nt 3732 in pCTV9 and pCTV-ΔCla was corrected by reverse transcription of the region between nts 3559 and 4630 followed by amplification by PCR from the double-stranded RNA extracted from CTV T36 isolate-infected bark tissue. The amplified cDNA fragment was digested with NcoI (nt 3685) and Bsu36I (nt 4430) restriction endonucleases and followed by ligation into similarly digested pCTV9 and pCTV-ΔCla to obtain pCTV9R and pCTV-ΔClaR, respectively (Fig. 1A). CTV-GFP-p23 was obtained by precisely exchanging the CP ORF with that of GFP in CTV-CP/p23 (Satyanarayana et al., 2002a). Subsequently, the +1 insertion mutation found at nt 3732 in CTV-GFP-p23 was removed by exchanging a DNA fragment between NcoI and Bsu36I from pCTV9R to obtain CTVR-GFP-p23 (Fig. 3A).

The frameshift mutations in ORF 1a were introduced by PCR using a set of positive- and negative-sense oligonucleotides that contained the intended mutation, followed by overlap-extension PCR (Ho et al., 1989) of the region between unique restriction sites in pCTV-ΔClaR (Fig. 4A). The +2 and +1 insertion mutations at nts 3732 and 4079, respectively, were introduced by amplifying DNA fragments that contained the mutations between nts 3559 and 4630 and then ligated between NcoI (nt 3685) and Bsu36I (nt 4430) restriction sites into pCTV-ΔClaR to obtain ΔClaR-M4 and ΔClaR-M1, respectively (Fig. 4A). Insertion mutations (+1) were introduced at nt 8390 by amplifying a PCR product between nts 7747 and 19293 using pCTV-ΔClaR as a template and ligated between BstEII (nt 7775) and NotI (nt 19293) restriction sites into pCTV-ΔClaR to obtain ΔClaR-M2, and at nt 1109 by overlap-extension PCR, and the PCR product was ligated between Bg/II (nt 1029) and NcoI (nt 3685) restriction sites into pCTV-ΔClaR to obtain ΔClaR-M3 (Fig. 4A). The PCR products amplified between nts 3559 and 7803 that contained +1 and −1 mutations at nts 4963 and 7084 were ligated between NcoI and BstEII sites into pCTV-ΔClaR to obtain ΔClaR-M5 and ΔClaR-M6, respectively (Fig. 4A). Silent mutations in homopolymeric run of U’s between nts 3726 and 3732 (UUU UUU U changed to UUC UUC U) were introduced by overlap-extension PCR using mutagenized primers between nts 3559 and 4630 and then ligated into pCTV-ΔCla between NcoI and Bsu36I restriction sites to obtain pCTV-ΔCla-M7 that contained a +1 insertion mutation at nt 3732 (Fig. 5A). pCTV-ΔClaR-M8 contained silent mutations in wild-type pCTV-ΔClaR between nts 3726 and 3731 (UUU UUU changed to UUC UUC) (Fig. 5A).

Protoplasts transfection and Northern blot analysis of viral RNAs

SP6 RNA polymerase-derived transcripts from NotI-linearized CTV cDNAs were generated in vitro as described by Satyanarayana et al. (1999). The procedures for the isolation of mesophyll protoplasts from N. benthamiana leaves and polyethylene glycol mediated transfections were carried out as described previously (Satyanarayana et al., 1999). The total RNA isolated from 3 and 4 day post inoculated (dpi) protoplasts were analyzed by Northern blot hybridization using a 3′ positive-stranded RNA-specific riboprobe (Satyanarayana et al., 1999). Bands from different exposures of Northern hybridization blots of total RNA from protoplasts at 4 dpi were quantified by scanning and densitometry with the OS-SCAN program (Oberlin Scientific, Oberlin, OH). Results represented at least three to four independent protoplasts transfections using two to three independent clones for each construct.

E. coli strain, media, and plasmid DNA isolation

Competent cells of E. coli strain JM109 prepared as described by Hanahan (1995) were transformed and plated on LB agar-amp plates (1.0% tryptone, 0.5% yeast extract, and 1.0% sodium chloride in water, pH 7.0 plus 1.5% bacto-agar containing 100 μg/ml ampicillin). The plates were incubated overnight at 37°C followed by 2 days incubation at room temperature. The size of E. coli colonies was measured with a microscope at 1 and 2 dpt. 2xYT liquid medium (1.6% tryptone, 1.0% yeast extract, and 0.5% sodium chloride in water, pH 7.0) was used to grow E. coli harboring recombinant DNAs for the plasmid DNA isolation (Sambrook and Russell, 2001).

Nucleotide sequencing

Sequencing of the full-length infectious clone pCTV9 and the frameshift mutants to confirm the mutations was performed with an automatic sequencer (Applied Biosystems, model 373) at the Interdisciplinary Center for Biotechnology Research DNA sequencing core facility of the University of Florida, Gainesville, FL.

Nucleotide sequence accession number

The complete nucleotide sequence of the infectious full-length cDNA clone of CTV T36 isolate, pCTV9R, was deposited in the GenBank database under the Accession No. AY170468. The nucleotide numbering and positions mentioned in this investigation correspond to the pCTV9R sequence.

Acknowledgements

We thank John Cook, Cecile Robertson, and Judy Harber for excellent technical assistance. This research was supported in part by an endowment from the J.R. and Addie S. Graves family and grants from the Florida Citrus Production Research Advisory Council, the National Citrus Research Council, the U.S.–Israel Binational Agricultural Research and Development Fund, USDA/ARS Cooperative Agreement, the USDA/NRI, and the Florida Agricultural Experiment Station, and approved for publication as Journal Series No. R-09320.

References

  1. Ahlquist P, French R, Janda M, Loesch-Fries L.S. Multicomponent RNA plant virus infection derived from cloned viral cDNA. Proc. Natl. Acad. Sci. USA. 1984;81:7066–7070. doi: 10.1073/pnas.81.22.7066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Albiach-Martí M.R, Mawassi M, Gowda S, Satyanarayana T, Hilf M.E, Shanker S, Almira E.C, Vives M.C, Lopez C, Guerri J, Flores R, Moreno P, Garnsey S.M, Dawson W.O. Sequences of citrus tristeza virus separated in time and space are essentially identical. J. Virol. 2000;74:6856–6865. doi: 10.1128/jvi.74.15.6856-6865.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Almazán F, González J.M, Pénzes Z, Izeta A, Calvo E, Plana-Durán J, Enjuanes L. Engineering the largest RNA virus genome as an infectious bacterial artificial chromosome. Proc. Natl. Acad. Sci. USA. 2000;97:5516–5521. doi: 10.1073/pnas.97.10.5516. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Barry J.K, Miller W.A. A -1 ribosomal frameshift element that requires base pairing across four kilobases suggests a mechanism of regulating ribosome and replicase traffic on a viral RNA. Proc. Natl. Acad. Sci. USA. 2002;99:11133–11138. doi: 10.1073/pnas.162223099. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Boyer J.-C, Haenni A.-L. Infectious transcripts and cDNA clone of RNA viruses. Virology. 1994;198:415–426. doi: 10.1006/viro.1994.1053. [DOI] [PubMed] [Google Scholar]
  6. Brierley I, Jenner A.J, Inglis S.C. Mutational analysis of the “slippery-sequence” component of a coronavirus ribosomal frameshifting signal. J. Mol. Biol. 1992;227:463–479. doi: 10.1016/0022-2836(92)90901-U. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Casais R, Thiel V, Siddell S.G, Cavanagh D, Britton P. Reverse genetics system for the avian coronavirus infectious bronchitis virus. J. Virol. 2001;75:12359–12369. doi: 10.1128/JVI.75.24.12359-12369.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chamorro M, Parkin N, Varmus H.E. An RNA pseudoknot and an optimal heptameric shift site are required for highly efficient ribosomal frameshifting on a retroviral messenger RNA. Proc. Natl. Acad. Sci. USA. 1992;89:713–717. doi: 10.1073/pnas.89.2.713. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chen X, Chamorro M, Lee S.I, Shen L.X, Hines J.V, Tinoco I, Jr., Varmus H.E. Structural and functional studies of retroviral RNA pseudoknots involved in ribosomal frameshifting: nucleotides at the junction of the two stems are important for efficient ribosomal frameshifting. EMBO J. 1995;14:842–852. doi: 10.1002/j.1460-2075.1995.tb07062.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dawson W.O, Bubrick P, Grantham G.L. Modification of the tobacco mosaic virus coat protein gene affects replication, movement, and symptomatology. Phytopathology. 1988;78:783–789. [Google Scholar]
  11. Domingo E, Holland J.J, Biebricher C, Eigen M. Quasispecies: the concept and the word. In: Gibbs A.J, Calisher C.H, García-Arenal F, editors. Molecular Basis of Virus Evolution. Cambridge Univ. Press; Cambridge, UK: 1995. pp. 181–191. [Google Scholar]
  12. Drake J.W, Holland J.J. Mutation rates among RNA viruses. Proc. Natl. Acad. Sci. USA. 1999;96:13910–13913. doi: 10.1073/pnas.96.24.13910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. González J.M, Pénzes P, Almazán F, Calvo E, Enjuanes L. Stabilization of a full-length infectious cDNA clone of transmissible gastroenteritis coronavirus by insertion of an intron. J. Virol. 2002;76:4655–4661. doi: 10.1128/JVI.76.9.4655-4661.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gowda S, Satyanarayana T, Ayllón M.A, Albiach-Martí M.R, Mawassi M, Rabindran S, Garnsey S.M, Dawson W.O. Characterization of the cis-acting elements controlling subgenomic mRNAs of citrus tristeza virus: production of positive- and negative-stranded 3′-terminal and positive-stranded 5′-terminal RNAs. Virology. 2001;286:134–151. doi: 10.1006/viro.2001.0987. [DOI] [PubMed] [Google Scholar]
  15. Hanahan D. Techniques for transformation of E. coli. In: Glover D.M, Hames B.D, editors. Vol. 1. IRL Press; Oxford, UK: 1995. pp. 1–36. (DNA Cloning: a practical approach, Second ed, Core Techniques). [Google Scholar]
  16. Hilf M.E, Karasev A.V, Pappu H.R, Gumpf D.J, Niblett C.L, Garnsey S.M. Characterization of citrus tristeza virus subgenomic RNAs in infected tissue. Virology. 1995;208:576–582. doi: 10.1006/viro.1995.1188. [DOI] [PubMed] [Google Scholar]
  17. Ho S.N, Hunt H.D, Horton R.M, Pullen J.K, Pease L.R. Site-directed mutagenesis by overlap extension using polymerase chain reaction. Gene. 1989;77:51–59. doi: 10.1016/0378-1119(89)90358-2. [DOI] [PubMed] [Google Scholar]
  18. Honda A, Nakamura T, Nishimura S. RNA signals for translation frameshift: influence of stem size and slippery sequence. Biochem. Biophys. Res. Commun. 1995;213:575–582. doi: 10.1006/bbrc.1995.2170. [DOI] [PubMed] [Google Scholar]
  19. Jacks T, Madhani H.D, Masiarz F.R, Varmus H.E. Signals for ribosomal frameshifting in the Rous sarcoma virus gag-pol region. Cell. 1988;55:447–458. doi: 10.1016/0092-8674(88)90031-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jacks T, Townsley K, Varmus H.E, Majors J. Two efficient ribosomal frameshifting events are required for synthesis of mouse mammary tumor virus gag-related polyproteins. Proc. Natl. Acad. Sci. USA. 1987;84:4298–4302. doi: 10.1073/pnas.84.12.4298. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Johansen I.E. Intron insertion facilitates amplification of cloned virus cDNA in Escherichia coli while biological activity is reestablished after transcription in vivo. Proc. Natl. Acad. Sci. USA. 1996;93:12400–12405. doi: 10.1073/pnas.93.22.12400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Karasev A.V, Boyko V.P, Gowda S, Nikolaeva O.V, Hilf M.E, Koonin E.V, Niblett C.L, Cline K, Gumpf D.J, Lee R.F, Garnsey S.M, Lewandowski D.J, Dawson W.O. Complete sequence of the citrus tristeza virus RNA genome. Virology. 1995;208:511–520. doi: 10.1006/viro.1995.1182. [DOI] [PubMed] [Google Scholar]
  23. Kim Y.G, Maas S, Rich A. Comparative mutational analysis of cis-acting RNA signals for translational frameshifting in HIV-1 and HTLV-2. Nucleic Acids Res. 2001;29:1125–1131. doi: 10.1093/nar/29.5.1125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lai M.M.C. The making of infectious viral RNA: no size limit in sight. Proc. Natl. Acad. Sci. USA. 2000;97:5025–5027. doi: 10.1073/pnas.97.10.5025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. López-Moya J.J, García J.A. Construction of stable and highly infectious intron-containing cDNA clone of plum pox potyvirus and its use to infect plants by particle bombardment. Virus Res. 2000;68:99–107. doi: 10.1016/s0168-1702(00)00161-1. [DOI] [PubMed] [Google Scholar]
  26. Marczinke B, Fisher R, Vidakovic M, Bloys A.J, Brierley I. Secondary structure and mutational analysis of the ribosomal frameshift signal of Rous sarcoma virus. J. Mol. Biol. 1998;284:205–225. doi: 10.1006/jmbi.1998.2186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pappu H.R, Karasev A.V, Anderson E.J, Pappu S.S, Hilf M.E, Febres V.J, Eckloff R.M.G, McCaffery M, Boyko V.P, Gowda S, Dolja V.V, Koonin E.V, Gumpf D.J, Cline K.C, Garnsey S.M, Dawson W.O, Lee R.F, Niblett C.L. Nucleotide sequence and organization of eight 3′ open reading frames of the citrus tristeza closterovirus genome. Virology. 1994;199:35–46. doi: 10.1006/viro.1994.1095. [DOI] [PubMed] [Google Scholar]
  28. Parkin N.T, Chamorro M, Varmus H.E. Human immunodeficiency virus type 1 gag-pol frameshifting is dependent on downstream mRNA secondary structure: demonstration by expression in vivo. J. Virol. 1992;66:5147–5151. doi: 10.1128/jvi.66.8.5147-5151.1992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Quillet L, Guilley H, Jonard G, Richards K. In vitro synthesis of biologically active beet necrotic yellow vein virus RNA. Virology. 1989;172:293–301. doi: 10.1016/0042-6822(89)90131-1. [DOI] [PubMed] [Google Scholar]
  30. Racaniello V.R, Baltimore D. Cloned poliovirus cDNA is infectious in mammalian cells. Science. 1981;214:916–919. doi: 10.1126/science.6272391. [DOI] [PubMed] [Google Scholar]
  31. Rice C.M, Grakoui A, Galler R, Chambers T.J. Transcription of infectious yellow fever RNA from full-length cDNA templates produced by in vitro ligation. New Biol. 1989;1:285–296. [PubMed] [Google Scholar]
  32. Sambrook J, Russell D.W. Molecular Cloning: A Laboratory Manual. third ed. Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 2001. [Google Scholar]
  33. Satyanarayana T, Gowda S, Boyko V.P, Albiach-Martí M.R, Mawassi M, Navas-Castillo J, Karasev A.V, Dolja V, Hilf M.E, Lewandowski D.J, Moreno P, Bar-Joseph M, Garnsey S.M, Dawson W.O. An engineered closterovirus RNA replicon and analysis of heterologous terminal sequences for replication. Proc. Natl. Acad. Sci. USA. 1999;96:7433–7438. doi: 10.1073/pnas.96.13.7433. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Satyanarayana T, Gowda S, Mawassi M, Albiach-Martí M.R, Ayllón M.A, Robertson C, Garnsey S.M, Dawson W.O. Closterovirus encoded HSP70 homolog and p61 in addition to both coat proteins function in efficient virion assembly. Virology. 2000;278:253–265. doi: 10.1006/viro.2000.0638. [DOI] [PubMed] [Google Scholar]
  35. Satyanarayana T, Bar-Joseph M, Mawassi M, Albiach-Martí M.R, Ayllón M.A, Gowda S, Hilf M.E, Moreno P, Garnsey S.M, Dawson W.O. Amplification of Citrus tristeza virus from a cDNA clone and infection of citrus trees. Virology. 2001;280:87–96. doi: 10.1006/viro.2000.0759. [DOI] [PubMed] [Google Scholar]
  36. Satyanarayana T, Gowda S, Ayllón M.A, Albiach-Martí M.R, Dawson W.O. Mutational analysis of the replication signals in the 3′-nontranslated region of Citrus tristeza virus. Virology. 2002;300:140–152. doi: 10.1006/viro.2002.1550. [DOI] [PubMed] [Google Scholar]
  37. Satyanarayana T, Gowda S, Ayllón M.A, Albiach-Martí M.R, Rabindran S, Dawson W.O. The p23 protein of Citrus tristeza virus controls asymmetrical RNA accumulation. J. Virol. 2002;76:473–483. doi: 10.1128/JVI.76.2.473-483.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Taniguchi T, Palmieri M, Weissmann C.Q. DNA -containing hybrid plasmids giving rise to Q phage formation in the bacterial host. Nature (London) 1978;274:223–228. doi: 10.1038/274223a0. [DOI] [PubMed] [Google Scholar]
  39. Thiel V, Herold J, Schelle B, Siddell S.G. Infectious RNA transcribed in vitro from a cDNA copy of the human coronavirus genome cloned in vaccinia virus. J. Gen. Virol. 2001;82:1273–1281. doi: 10.1099/0022-1317-82-6-1273. [DOI] [PubMed] [Google Scholar]
  40. Turpen T.H. Ph.D. dissertation. A Molecular Genetic Analysis of Host/Viral Interactions, Implications for the Use of Plant Viruses as Gene Vectors. University of California; Riverside, CA: 1992. [Google Scholar]
  41. Vickers T.A, Ecker D.J. Enhancement of ribosomal frameshifting by oligonucleotides targeted to the HIV gag-pol region. Nucleic Acids Res. 1992;20:3945–3953. doi: 10.1093/nar/20.15.3945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wilson W, Braddock M, Adams S.E, Rathjen P.D, Kingsman S.M, Kingsman A.J. HIV expression strategies: ribosomal frameshifting is directed by a short sequence in both mammalian and yeast systems. Cell. 1988;55:1159–1169. doi: 10.1016/0092-8674(88)90260-7. [DOI] [PubMed] [Google Scholar]
  43. Yamshchikov V, Mishin V, Cominelli F. A new strategy in design of (+)RNA virus infectious clones enabling their stable propagation in E. coli. Virology. 2001;281:272–280. doi: 10.1006/viro.2000.0793. [DOI] [PubMed] [Google Scholar]
  44. Yount B, Curtis K.M, Baric R.S. Strategy for systematic assembly of large RNA and DNA genomes: transmissible gastroenteritis virus model. J. Virol. 2000;74:10600–10611. doi: 10.1128/jvi.74.22.10600-10611.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Yount B, Denison M.R, Weiss S.R, Baric R.S. Systematic assembly of a full-length infectious cDNA clone of mouse hepatitis virus strain A59. J. Virol. 2002;76:11065–11078. doi: 10.1128/JVI.76.21.11065-11078.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Virology are provided here courtesy of Elsevier

RESOURCES