A Novel Viral RNA Species in Sindbis Virus-Infected Cells

JOURNAL OF VIROLOGY, Dec. 1997, p. 9108–9117 0022-538X/97/$04.0010 Copyright © 1997, American Society for Microbiology

Vol. 71, No. 12

A Novel Viral RNA Species in Sindbis Virus-Infected Cells MATTHEW M. WIELGOSZ

AND

HENRY V. HUANG*

Department of Molecular Microbiology, Washington University School of Medicine, St. Louis, Missouri 63110-1093 Received 3 March 1997/Accepted 8 August 1997

Sindbis virus (SIN), the type alphavirus, has been studied extensively to identify the viral cis-acting sequences and proteins involved in RNA transcription and replication. However, very little is known about how these processes are coordinated. For example, synthesis of the genomic RNA and the subgenomic mRNA depends on the minus strand. Do these activities occur independently on different templates, or can replication and transcription take place simultaneously on the same template? We describe the appearance of a SINspecific, plus-sense RNA that is intermediate in size between the genomic and subgenomic RNA species. This RNA, designated RNA II, is observed in a number of different cell lines, both early and late in infection. The number of RNA II species, their sizes, and their abundances are influenced by the subgenomic promoter. We have mapped the 3* end of RNA II to a site within the subgenomic promoter, four nucleotides before the initiation site of the subgenomic mRNA. Our results indicate that the appearance of RNA II is correlated with subgenomic mRNA transcription, such that strong or active promoters tend to increase the abundance of RNA II, relative to weak or less active promoters. RNA II is most abundantly detected with the full promoter and is at much lower abundance with the minimal promoter. The possible origins of RNA II are discussed. The alphaviruses belong to the Togaviridae family. They have a global distribution and infect insect, avian, and mammalian hosts with various degrees of virulence (20, 46, 50, 62). The identification of viral cis-acting sequences and trans-acting proteins required for alphavirus replication and transcription has led to the development of wide-host-range expression vectors (4, 42, 73) and potential vaccines for disease-transmitting mosquitoes (6). In Sindbis virus (SIN), the first two-thirds of the incoming genomic RNA is translated to yield two nonstructural polyproteins (nsPs), nsP123 and nsP1234. The polyproteins and/or their proteolytic products constitute the viral contribution to the RNA-dependent RNA polymerase (R-dRp). The exact composition of the R-dRp is not known, and it may differ depending on whether the R-dRp is engaged in replication or transcription. However, several functions have been identified or assigned to the individual nsPs which participate in viral RNA synthesis. nsP1 is thought to be the enzyme that caps viral RNA, since it can bind guanylate nucleotides (1) and exhibits methyltransferase activity (45). nsP1 also functions in minus-strand RNA synthesis (69). nsP2 has sequence motifs characteristic of nucleoside triphosphate binding domains (17, 19, 27) and helicase domains (18). Indeed, nsP2 of Semliki Forest virus (SFV) exhibits GTPase and ATPase activities (52). However, attempts to demonstrate helicase activity in this protein have not been successful (52). nsP2 also functions as the nsP protease (68) and is critical to subgenomic mRNA synthesis (24, 58, 59). Specific roles for nsP3 have not been defined. nsP3 is serine/threonine phosphorylated in vivo (41) and appears to be involved in genomic replication (24, 33, 35, 36, 64) and subgenomic mRNA transcription (33). nsP4 contains the GDD motif that is common to many viral RNA polymerases (28) and is considered to be the polymerase of SIN. Several temperature-sensitive (ts) mutations which affect

viral RNA synthesis have been mapped to nsP4 (23, 56), and one of these renders the virus completely defective in RNA synthesis (3, 30, 60). The nsPs and/or their proteolytic products are believed to assemble on the 39 end of the genomic RNA to synthesize a complementary minus-strand RNA (31, 38). The minus strand contains two cis-acting sequences that are critical to viral replication. The first is located at the 39 end of the minus strand, and it directs the replication of genomic RNA, which, in turn, is used for continued minus-strand synthesis and nsP translation (31, 40). The second cis-acting sequence is the subgenomic promoter. It is located approximately 7.6 kb downstream from the 39 end of the minus strand, and it is required for the transcription of subgenomic mRNA (39, 47). Transcription initiates at the promoter and by runoff transcription produces a subgenomic mRNA that is 39 coterminal with the genomic RNA. The subgenomic mRNA encodes the viral structural proteins, and these package the genomic RNA into progeny virions (22, 44, 67, 70, 71). As the SIN infection cycle progresses, genomic RNA is routed predominantly into the virion assembly pathway, concurrent with an increased rate of plus-strand RNA synthesis, cessation of minus-strand synthesis, and a decreased rate of nsP translation. The switch from minus-strand synthesis early in infection to plus-strand synthesis later in infection appears to involve the processing of the nsPs into their individual nonstructural proteins and/or proteolytic intermediates (12, 34–36). It is not known how transcription and replication are coordinated on the minus strand. There is some evidence that these activities occur simultaneously on the same template (57, 63). Three double-stranded (ds) replicative forms (RFs) have been identified upon the treatment of partially ds replicative intermediates with RNase (5, 13, 57, 63, 66). RF I constitutes a ds form of the full-length genome, while RFs II and III correspond to the ds forms of the first two-thirds (the nonstructural region) and the remaining one-third (the structural region) of the genome, respectively (66). Sawicki et al. found that an SFV ts mutant exhibited decreased levels of subgenomic RNA relative to genomic RNA at the nonpermissive temperature and displayed a correlated decrease in RFs II and III relative to RF

* Corresponding author. Mailing address: Department of Molecular Microbiology, Washington University School of Medicine, Box 8230, 660 S. Euclid Ave., St. Louis, MO 63110-1093. Phone: (314) 362-2755. Fax: (314) 362-1232. 9108

VOL. 71, 1997

A NOVEL SIN-SPECIFIC RNA SPECIES

I (57). Those authors concluded that a virus-specific protein which functions to promote transcription and block genomic RNA replication was responsible for the appearance of RNasesensitive sites within replicative intermediates, such that RNase treatment of these species gave rise to RFs II and III (57). Thus, when the virus-specific protein could not function normally at the nonpermissive temperature, the RNase-sensitive sites of the replicative intermediates were masked and RF I became the dominant species. Similar results were obtained by Segal and Sreevalsan by using a ts mutant that was defective in subgenomic mRNA transcription in SIN (63). Although ds RNA can be obtained from infected cells upon treatment with RNase, the predominant SIN-specific RNA species in an infected cell are the single-stranded genomic and subgenomic RNAs (65). Other single-stranded viral RNA species have been observed in SIN (9, 26, 49, 53) and SFV (8, 29, 37). One species, in particular, was found to have a sedimentation coefficient of 38S and an electrophoretic mobility between those of genomic and subgenomic RNAs under nondenaturing conditions (9, 37). When the SFV 38S species was resolved under denaturing conditions, it was found to comigrate with the genomic 42S RNA (29). RNase T1 oligonucleotide fingerprinting confirmed that the 38S species was most likely a conformer of the genomic 42S RNA (29). Recent observations in our laboratory (26) and others (49) have shown that a glyoxal-denatured RNA species that is intermediate in size between genomic and subgenomic RNAs is present in SIN-infected cells. This RNA, which we call RNA II, may represent the single-stranded equivalent of RF II. Very little is known about RNA II, other than that it is SIN specific (unpublished observations) (49) and its appearance depends on the SIN subgenomic promoter (26). We report here the characterization of RNA II. MATERIALS AND METHODS Reagents, chemicals, isotopes, recombinant DNA materials, and methods. Trizol reagent was obtained from Bethesda Research Laboratories (Grand Island, N.Y.). Cycloheximide was purchased from Sigma. [a-32P]GTP was from Amersham (Arlington Heights, Ill.). [32P]orthophosphate and [g-32P]ATP were obtained from ICN (Costa Mesa, Calif.). Restriction endonucleases and other recombinant DNA materials were purchased from New England Biolabs (Beverly, Mass.), Bethesda Research Laboratories, Epicentre Technologies (Madison, Wis.), or Boehringer Mannheim. Poly(A) polymerase was acquired from United States Biochemical (Cleveland, Ohio). A version of the DTaq DNAdependent DNA polymerase was obtained at Washington University (St. Louis, Mo.), but the sequencing reaction was performed as described in the DTaq sequencing kit (United States Biochemical) for kinased primers. All molecular biology procedures were performed as described by Sambrook et al. (54), unless otherwise noted. Cell lines. Unless otherwise noted, all vertebrate and invertebrate cells were grown at 37 and 30°C, respectively. Baby hamster kidney (BHK-21) cells (ATCC CCL10) were grown in minimal essential medium with Earle’s salts (MEM) supplemented with 10% heat-inactivated (HI) fetal calf serum (FCS). Cells between passages 5 and 15 were used. Primary chicken embryo fibroblasts (CEF) were prepared as described previously (48) and grown in MEM supplemented with 3% HI FCS, 100-U/ml penicillin, and 100-mg/ml streptomycin. Secondary CEF cells were used. C7-10 cells and C6-36 cells (ATCC 1660-CRL), originally isolated from Aedes albopictus larvae (55), were grown in MEM supplemented with 10% HI FCS and 10% tryptone phosphate buffer. Cells between passages 30 and 40 (C7-10) or 124 and 132 (C6-36) were used. African green monkey kidney (Vero) cells (ATCC 81-CCL) were grown in Medium 199 supplemented with 5% FCS (not HI). Cells between passages 125 and 132 were used. Virus constructs. Toto1000 contains the entire SIN genome [11,703 nucleotides, excluding the poly(A) tail] downstream of the SP6 promoter (51). Toto1000 has one subgenomic promoter that is positioned ;7.6 kb downstream from the virus’s 59 end. This promoter is required for transcription of the subgenomic mRNA which encodes the virion’s structural proteins. It is ;4.2 kb in size and is 39 coterminal with the genomic RNA. TCS is similar to Toto1000, in that it contains the nonstructural (nsP) and structural (STR) coding regions of SIN. However, the nsP and STR coding regions of TCS are separated by the chloramphenicol acetyltransferase (CAT) gene (25). Additionally, TCS contains two subgenomic promoters (25). The first is the CAT promoter, and it is located ;7.6 kb downstream from the 59 end of

9109

the genomic RNA (;12.6 kb). It is required for transcription of the CAT subgenomic mRNA, which is ;5.0 kb in size and 39 coterminal with the genomic plus-strand RNA. The second promoter is the STR promoter, and it is located ;8.4 kb downstream from the 59 end of the genomic RNA. It is required for transcription of the STR subgenomic mRNA (identical to that of Toto1000), which is ;4.2 kb in size and 39 coterminal with the genomic plus-strand RNA. The CAT and STR promoters of TCS contain the 298/114 subgenomic promoter sequences (298 refers to the number of nucleotides upstream of the start site of the subgenomic mRNA, and 114 refers to the number of nucleotides downstream of the start site of the subgenomic mRNA). The B2, III, and 219/15 clones have been described previously and differ from TCS only in that their STR promoters are mutant (25, 26). The B2 STR promoter sequence was isolated from a library of viruses which contained random sequences between the 213 and 29 positions of the STR subgenomic promoter. The B2 virus has nucleotide changes at the 213, 210, and 29 positions, where the wild-type A, G, and U nucleotides are replaced with G, A, and A, respectively. The III virus was another virus from the 213 to 29 library, and its STR promoter contained nucleotide changes at positions 211 and 210, where the wild-type G and G nucleotides have been replaced with C and C, respectively. The 219/15 clone was constructed such that its STR subgenomic promoter contains the 219 to 15 promoter sequences in place of the 298/114 promoter. TCIS is identical to TCS, except that a 979-bp insert is positioned immediately downstream of the CAT gene and immediately upstream of the STR promoter. Thus, the CAT subgenomic mRNA of TCIS is ;6 kb in size and is 39 coterminal with the genomic RNA. The STR subgenomic mRNA of TCIS is identical to that of Toto1000 (;4.2 kb in size and coterminal with the 39 end of the genomic plus-strand RNA). TCIS was constructed as follows. TCS was digested with XhoI, and its ends were rendered blunt with T4 DNA polymerase and dephosphorylated with calf alkaline phosphatase. A 979-bp insert was generated by digesting pSPORT1 (Bethesda Research Laboratories) with TaqI. The 979-bp DNA fragment was isolated, and its ends were rendered blunt. The XhoI-digested TCS vector and the 979-bp insert were ligated together by using T4 DNA ligase. The 240/114, 240/15, and 298/15 clones are identical to TCS, but their STR promoters (240/114, 240/15, or 298/15) have replaced the 298/114 STR promoter of TCS. The construction of these clones was similar to that described for the 219/15 clone (25), with some differences. First, the region of interest (i.e., 240/114, 240/15, or 298/15) was PCR amplified from the Toto1000 template by using oligonucleotides that contained the appropriate 59 or 39 promoter sequences, flanked by either the XhoI or XbaI site, respectively. The PCR products were digested with XhoI and XbaI and directionally cloned into the Pneo S shuttle vector (25). The subcloned promoters were then digested with XhoI and BssHII and directionally cloned into TDV (25). The 219/114 clone differs from the 219/15 clone in that the STR promoter contains the 219/114 promoter in place of the 219/15 promoter. The construction of 219/114 was done as follows. The 219/15 clone was used as template DNA in the PCR amplification of a 211-nucleotide sequence using oligonucleotides 1940 and 1914. Oligonucleotide 1940 (59 CCCGTTTTCACCATGGGCA AATA 39) hybridizes to the noncoding strand of CAT 266 nucleotides upstream of the STR subgenomic mRNA start site. Oligonucleotide 1914 (59 CTAGTCT AGAACTATGCTGACTATTTAGG 39) contains the 25 to 114 sequence and hybridizes to the coding strand of 219/15, encompassing nucleotides 25 to 15 relative to the STR subgenomic RNA start site. The PCR product was digested with XhoI and XbaI and directionally cloned into the 219/15 clone. For the construction of TCS-21, Toto:ts 21 A2 was digested with BglII and ClaI and the 423-bp fragment was directionally cloned into TCS. Toto:ts 21 A2 contains a single G-to-A mutation at nucleotide 2590, changing Cys 304 of nsP2 to Tyr (24). RNA transcription and transfection and preparation of viral stocks. SstIlinearized plasmids were transcribed in vitro with SP6 DNA-dependent RNA polymerase. The DNA templates were digested with DNase I, and the RNA transcripts were phenol, chloroform, and ether extracted prior to ethanol precipitation. The transfection procedure was similar to that described by Liljestro ¨m et al. (43). A 5-mg sample of a given transcript was mixed with 107 BHK-21 cells, and the cells were electroporated in a 0.2-cm gap electrocuvette using a T820 Electro Square Porator (BTX Inc., San Diego, Calif.). Transfected BHK-21 cells were added to 5 ml of growth medium, transferred to a culture dish, and incubated at 37°C for 16 h. The medium was collected and centrifuged at 12,000 3 g for 20 min at 4°C to pellet cell debris. Aliquots of the medium were placed into 0.5-ml microcentrifuge tubes and stored at 280°C until use. Infection and RNA labeling. Cells (;0.4 3 106 to 0.6 3 106) were seeded onto 60-mm-diameter culture dishes. Vertebrate and invertebrate cells were maintained at 37 or 30°C, respectively, for approximately 24 h. After this time, vertebrate cells approached 90% confluence while invertebrate cells approached 50% confluence. Vertebrate and invertebrate cells were both infected with virus at a multiplicity of infection (MOI) of 5 at either 37°C (vertebrate cells) or 30°C (invertebrate cells) in solution A (phosphate-buffered saline minus Mg21 and Ca21) supplemented with 1% HI FCS. At 1 h postinfection (hpi), the inoculum was removed and the cells were washed twice with solution A (23°C) supplemented with 1% HI FCS. The cell cultures then received 3 ml of the appropriate medium and were maintained at 37 or 30°C. For RNA labeling in vertebrate cells, dactinomycin (1-mg/ml final concentration) was added to the medium 1 h before labeling. The medium was then

9110

WIELGOSZ AND HUANG

removed and replaced with 1.5 ml of labeling medium (1-mg/ml dactinomycin and 100-mC/ml [32P]orthophosphate in the appropriate phosphate-free cell medium). The cells were incubated at 37°C for 2 h. RNA labeling for invertebrate cells was essentially the same as that for the vertebrate cells. Dactinomycin (5-mg/ml final concentration) was added to the medium 0.5 h before labeling (either 20.5 or 44.5 hpi). The medium was then removed and replaced with 1.5 ml of labeling medium (identical to the vertebrate labeling medium except for 5-mg/ml dactinomycin). The cells were incubated at 30°C for 3 h. RNA isolation and analysis. The medium overlying each cell monolayer was removed, and the cells were washed twice with ice-cold solution A. The cells were solubilized with 3 ml of Trizol reagent and stored at 280°C. RNA was isolated in accordance with the manufacturer’s protocol and dissolved in 10 to 30 ml of diethylpyrocarbonate-treated water. Approximately 1/10 of the total RNA from a given cell sample was denatured with glyoxal and dimethyl sulfoxide (7) and electrophoresed through 1% agarose gels in 10 mM sodium phosphate (pH 7.0) (54). Gels were fixed with methanol and dried. 32P-labeled RNA was detected by autoradiography with film that was flash hypersensitized up to an A440 of ;0.15 above that of nonflashed film (32, 54). Exposures were done at 280°C for 1 to 5 days without an intensifying screen. 32 P-labeled RNA was quantitated on a Bio-Rad PhosphorImager. The background counts were estimated by the average counts above and below those of the RNA species and subtracted from the counts for the RNA species. The relative molar quantity of each viral RNA was determined by multiplying its background-subtracted counts by a size correction factor. For example, the relative amount of the structural subgenomic RNA of TCS compared to genomic RNA was calculated by multiplying the actual counts of the subgenomic RNA by 3.03 (length of TCS genome divided by the length of the STR subgenomic RNA, i.e., 12,589/4,156). Northern blot analysis. The Northern blotting procedure was performed as previously described (54), with minor changes. Samples (3 mg) of total RNAs from TCS-, Toto1000-, and mock-infected cells were denatured and resolved on a 1% agarose gel. The gel was soaked in 20 mM NaOH for 20 min and then washed extensively with water. The gel was neutralized with 203 SSC (13 SSC is 0.15 M NaCl plus 0.015 M sodium citrate), and the RNA was blot transferred overnight to a Hybond N1 membrane (Amersham). The membrane was dried at room temperature and irradiated with 120 mJ of UV light (254 nm) with a Stratagene auto-cross-linker (Stratagene, La Jolla, Calif.). All probes used in the Northern analyses were in vitro transcribed from DNA templates by using the T7 DNA-dependent RNA polymerase and were radiolabeled with [a-32P]GTP. The STR probe template contains the capsid coding sequences of Toto1000 between nucleotides 7612 and 7871 cloned into the HindIII site of pSPORT1. The CAT probe template contains nucleotides 195 to 442 of the bacterial CAT gene (2) directionally cloned into pSPORT1 at the BamHI and EcoRI sites. The nonstructural (NSP) probe template contains the nsP4 coding sequences of Toto1000 between nucleotides 7338 and 7593 cloned at the HindIII site of pSPORT1. The runoff site for the STR and NSP probes was MluI, and that for the CAT probe was BamHI. The RNA probes were each of minus-sense polarity. To reprobe a blot, probes were removed from the membrane by twice pouring an ;98°C solution of 1% sodium dodecyl sulfate onto the membrane. Poly(A) tailing. Total RNA (4 mg in a volume of 7 ml) was incubated at 50°C for 10 min, and then cooled on ice before the addition of 8 ml of poly(A) reaction buffer [37.5 mM Tris HCl [pH 7.0], 1.68 mM MnCl2, 3.0 mM MgCl2, 1.87 mM ATP, 94 mM dithiothreitol, 1.5-mg/ml bovine serum albumin, 112.3 mM KCl, 18.7% glycerol, 18.8-U/ml poly(A) polymerase]. Poly(A) reactions were carried out at 30°C for 20 min with 10-U/ml (final concentration) poly(A) polymerase and were stopped by addition of phenol. Total RNA was chloroform and ether extracted and then ethanol precipitated with a tRNA carrier. RT and PCR. Total RNA from the poly(A) reactions was solubilized in 10 ml of a NotI oligo(dT) mixture [32 mM NotI oligo(dT); 59 GATCTAGAGCGGCC GCCCTTTTTTTTTTTTTTT 39 in diethylpyrocarbonate-treated water), heated at 70°C for 10 min, and then cooled on ice. The reverse transcription (RT) reactions were performed in accordance with the Superscript II protocol at 48°C for 1 h. The reactions were terminated by heat denaturation at 70°C for 10 min. Half of each RT reaction mixture was used in a 100-ml PCR mixture with the 1940 and NotI oligo(dT) oligonucleotides at 1 mM as primers and 2.5 U of Tth polymerase. Thirty-five PCR cycles were performed (one cycle was 40 s at 95°C, 30 s at 60°C, and 40 s at 72°C). Cycle sequencing analysis. PCR-amplified products were resolved on 2.5% NuSieve GTG gels and gel purified (54). A 5-fmol sample of the PCR product was used per dideoxynucleotide termination reaction, in accordance with the DTaq cycle sequencing kit instructions for kinased primers. The 219/15 PCR product was sequenced with the 1941 primer (59 GAATTACAACAGTACTGC GATGA 39), which hybridizes to the noncoding strand of CAT 152 nucleotides upstream of the STR subgenomic mRNA start site. The PCR products obtained for the XbaI TCS transcript, TCS, and B2 samples were sequenced with primer 1913 (59 CCGCTCGAGGGGGCCCATTACACCTGTCCTAC 39), which hybridizes to the noncoding strand of TCS, encompassing nucleotides 298 to 283 relative to the STR subgenomic mRNA start site. The sequencing reaction products were resolved on 8% denaturing polyacrylamide (19:1 acrylamidebisacrylamide ratio) gels (54) and analyzed by autoradiography.

J. VIROL.

FIG. 1. Properties of RNA II that are determined by the subgenomic promoter. (A) BHK-21 cells were infected with the appropriate virus at an MOI of 5 and treated with dactinomycin 1 h before labeling with [32P]orthophosphate at 3 hpi. At 5 hpi, total RNA was isolated, denatured, and resolved on a 1% agarose gel. Exposures were at 280°C for 2 to 5 days. G is genomic RNA. Lg and Sm are the large and small RNA II species, respectively. The CAT and STR subgenomic RNAs are designated CAT and STR, respectively. (B) Same as in A but the in vivo-incorporated label of each RNA (relative to the genomic RNA of the virus) was quantitated on a Bio-Rad PhosphorImager and used to calculate the STR/ CAT and Lg/Sm ratios shown. These relative molar ratios have been normalized against those obtained for TCS.

RESULTS Gel analyses of SIN intracellular RNA reveal two prominent bands: the 49S genomic and 26S subgenomic RNAs. A third RNA species is frequently observed (see Toto1000 in Fig. 1A). We call this RNA II. Similar RNAs have been observed and shown to be isomers of the 49S genomic RNA (e.g., through denaturation) (29). We previously showed that insertion of an additional promoter into the SIN genome results in the transcription of an additional mRNA. The new mRNA, like the normal SIN subgenomic mRNA, initiates at the promoter and, by runoff transcription, produces a new subgenomic mRNA that is 39 coterminal with the genomic RNA (see CAT and STR subgenomic mRNAs in Fig. 1A). Interestingly, these viruses produce two species of RNA II. Indeed, several physical characteristics of RNA II are determined by the subgenomic promoter (26). Toto1000 has one subgenomic promoter which directs the transcription of the STR subgenomic RNA. TCS and its derivatives (B2, 219/15, III, and TCIS) have two—the CAT promoter, which directs the transcription of the CAT subgenomic RNA, and the STR promoter, which directs the transcription of the STR subgenomic RNA. As shown in Fig. 1A, Toto1000 has only one RNA II species, while TCS and its derivatives have two. These results show that the number of different RNA II species in the cell correlates with the number of promoters in the infecting virus. The second property of RNA II that is affected by the subgenomic promoter is size. The STR promoter of Toto1000 and the CAT promoters of TCS and its derivatives are each positioned ;7.6 kb downstream from the 59 ends of each virus.

VOL. 71, 1997

Accordingly, the sole RNA II species of Toto1000 and the small RNA II species of TCS and its derivatives have identical mobilities, corresponding to a size of 7.6 kb (Fig. 1A). The same relationship holds for the STR promoters of TCS and its derivatives, which are positioned ;8.43 kb downstream from each virus’s 59 end (except for that of TCIS). The STR promoter of TCIS is positioned 979 nucleotides downstream from the original STR promoter position of TCS. As shown in Fig. 1A, the new position of the STR promoter correlated with an increase in the size of the large RNA II species, from ;8.43 to ;9.41 kb. These results not only demonstrate that the size of RNA II is determined by the position of the subgenomic promoter but also suggest that RNA II is colinear with the 59 end of genomic RNA and has a 39 end that is close to the position of the subgenomic promoter. A third characteristic of RNA II that is affected by the subgenomic promoter is abundance. We previously found that the abundance of RNA II decreases when the strength of the subgenomic promoter decreases (26). Indeed, the 219/15 virus, whose STR promoter contains the 219/15 promoter sequence in place of the 298/114 STR promoter of TCS, has a weaker STR promoter than TCS (as assessed by the relative decrease in the STR-to-CAT subgenomic RNA ratios), and this correlated with decreased large (Lg)-to-small (Sm) RNA II ratios compared to TCS (Fig. 1A). Similar results were obtained with the B2 virus, which contains a weaker STR promoter than TCS (see Materials and Methods). The relative molar ratio of STR to CAT subgenomic mRNAs for B2 is ;0.6 (as assessed by PhosphorImager scans), while that for TCS is 1.9. Correspondingly, the ratio of Lg to Sm RNA II for B2 is ;0.5, while that for TCS is ;1.5. In contrast, the STR promoter of virus III has activity comparable to that of the TCS STR promoter (26), where the relative molar ratios of STR to CAT subgenomic mRNAs are 1.7 for III and 1.9 for TCS. This corresponded to a Lg-to-Sm RNA II ratio similar to that obtained for TCS (;1.0 for III and ;1.5 for TCS). We explored the association of RNA II abundance and subgenomic promoter activity further, with derivatives of TCS that contained various deletions in their respective STR promoters. The 240/114, 240/15, 298/15, and 219/114 viruses are identical to TCS, except that their STR promoter sequences have replaced the 298/114 STR promoter of TCS. As shown in Fig. 1B, the Lg-to-Sm RNA II ratios of the 219/15 and 219/114 viruses are lower than that obtained for TCS, and these correlate with decreased STR-to-CAT subgenomic mRNA ratios (relative to TCS). Note that the STR/CAT and Lg/Sm ratios listed in Fig. 1B are normalized against those obtained for TCS. The addition of the 16 to 114 promoter sequences to the minimal 219/15 promoter in 219/114 decreased its STR-to-CAT subgenomic mRNA ratio nearly threefold compared to that of 219/15, and this corresponded to a Lg-to-Sm RNA II ratio that was lower than that obtained for 219/15. In contrast, the STR-to-CAT subgenomic mRNA ratios of the 240/15 and 298/15 viruses are not appreciably different from that of 219/15, but their Lg-to-Sm RNA II ratios are four- to fivefold greater than that of 219/15 (Fig. 1B). These results indicate that the 240 to 220 promoter sequences, when placed upstream of the minimal 219/15 promoter, can increase the abundance of RNA II in a way that does not appear to be related to promoter strength. Interestingly, the 240 to 220 promoter sequences do enhance promoter strength when they are placed upstream of the 219/114 STR promoter. Indeed, the activity of the 240/114 STR promoter is nearly indistinguishable from that of the 298/114 STR promoter of TCS (Fig. 1B).


9111

FIG. 2. Appearance of RNA II in different vertebrate (A) and invertebrate (B) cells, both early and late in infection. Cells were infected with TCS at an MOI of 5 and treated with dactinomycin for 1 h (vertebrate) or 0.5 h (invertebrate) before labeling with [32P]orthophosphate. The cells were labeled for 2 h (vertebrate) or 3.5 h (invertebrate) before total RNA was isolated at the times shown. For definitions of abbreviations, see the legend to Fig. 1.

Appearance of RNA II in different cells. RNA II has previously been observed in BHK-21 hamster and C7-10 mosquito cells (25, 26). We used Vero, CEF, and C6-36 cells as additional hosts for infection by TCS and analyzed the appearance of RNA II at different times during the infection cycle. The times at which total RNA was harvested were determined by selecting the stages of TCS growth that were comparable among the different cell types. As shown in Fig. 2, RNA II was detected in all of the vertebrate (Fig. 2A) and invertebrate (Fig. 2B) cells tested, both early and late in infection. The mock-infected samples of the vertebrate and invertebrate cells did not yield any radiolabeled RNA species similar in size to RNA II (data not shown). RNA II was observed in BHK-21 cells infected at 30°C and as early as 3 or 8 hpi in BHK-21 and C7-10 cells, respectively (data not shown). These experiments demonstrate that the appearance of RNA II is not limited to a particular cell type, stage of infection, or temperature and suggest that the appearance of this RNA in infected cells is a general phenomenon. We performed several experiments like those depicted in Fig. 2 to measure the relative molar abundance of RNA II compared to the genomic and subgenomic RNAs of TCS and Toto1000. Intracellular viral RNAs were labeled for 2 h before isolation and gel analysis. The results are summarized in Table 1. Under these conditions, the relative molar abundance of RNA II changes by about twofold between 5 and 8 hpi (Table 1). On average, there is approximately 1 RNA II species to every 7 genomic or 50 subgenomic RNAs in Toto1000-infected cells. These are minimal estimates, as the stability of RNA II may be less than that of the other viral RNAs. Northern blot analysis of RNA II. We previously found that a negative-sense probe complementary to the 59 end of genomic RNA could detect RNA II by Northern blot hybridization (unpublished results). This indicated that RNA II is of

9112

WIELGOSZ AND HUANG

J. VIROL.

TABLE 1. Relative molar ratios of RNA II in TCS and Toto1000-infected BHK-21 cells at 37°C Virus, hpi

Toto1000, 5 Toto1000, 8 TCS, 5 TCS, 8

Ratio 6 SDa G/STR

0.10 6 0.03 0.14 6 0.04 0.33 6 0.05 0.25 6 0.07

G/CAT

G/Sm

0.56 6 0.14 0.56 6 0.16

561 862 15 6 4 10 6 4

G/Lg

18 6 4 10 6 1

STR/Sm

47 6 10 55 6 6

STR/Lg

59 6 10 37 6 7

CAT/Sm

27 6 8 17 6 6

nb

3 3 5 3

a The in vivo incorporated label of each RNA species (relative to the genomic RNA of the virus) was quantitated on a Bio-Rad PhosphorImager and used to calculate the relative molar ratios shown. G refers to the genomic RNA, STR and CAT refer to the STR and CAT subgenomic RNAs, and Sm and Lg refer to the small and large RNA II species, respectively. SD, standard deviation of each RNA ratio, from n experiments. b n, number of independent experiments performed.

plus-sense polarity and suggested that its 59 end is coterminal with genomic RNA. Furthermore, the results of Fig. 1A suggested that the 39 end of RNA II is close to the position of the subgenomic promoter. We determined the boundary of each RNA II species in TCS- and Toto1000-infected cells by Northern blot analysis using negative-sense probes that hybridize to positions both upstream and downstream of the subgenomic promoter. We first used an STR probe that is complementary to sequences immediately downstream of the STR subgenomic promoter, between nucleotides 115 and 1259 of Toto1000 or nucleotides 121 and 1264 of TCS. There are 35 nucleotides in the STR subgenomic untranslated region (downstream of the

STR promoter) which are identical to those in the CAT subgenomic untranslated region (downstream of the TCS CAT promoter). The Lg RNA II species of TCS contains these nucleotides and may be detected by the STR probe. As shown in Fig. 3A, the STR probe detected the genomic RNA of each virus, as well as the CAT subgenomic mRNA of TCS and the STR subgenomic mRNA of both viruses. However, the STR probe failed to detect the RNA II species of either virus. This result indicates that the 35 nucleotides of complementarity between the Lg RNA II species of TCS and the STR probe was insufficient for its detection and suggests that any RNA II species whose 39 end extends beyond the STR subgenomic promoter to include only the 35 nucleotides of the STR or

FIG. 3. Northern blot analysis of RNA II. BHK-21 cells were infected with Toto1000 or TCS at an MOI of 5 or mock infected. Total RNA was harvested at 5 hpi. RNA (3 mg per sample) was resolved on a 1% agarose gel and transferred to a nylon membrane. The membrane was probed with an in vitro-transcribed minus-sense STR probe (A), CAT probe RNA (B), or NSP probe (C). (D) Target RNAs of the different probes for TCS between nucleotides 7250 and 8750. The coding regions of TCS are designated NSP, CAT, and STR and are shown as filled boxes (black, textured gray, and white, respectively). The white boxes represent the untranslated region downstream of the CAT and STR promoters. The CAT and STR promoters are represented by arrows and the subscript P. The hash marks designate 150-nucleotide intervals. The open area upstream of the STR promoter represents the XhoI and ApaI sites of TCS. The probes are labeled and represented as shaded boxes which correspond to their target sequences. For definitions of abbreviations, see the legend to Fig. 1.

VOL. 71, 1997

CAT subgenomic untranslated region (which are complementary to the STR probe) would not be detected. Thus, the 39 end of RNA II does not extend significantly beyond the STR promoter. We removed the STR probe from the membrane and reprobed the blot a second time with a CAT probe that is complementary to the 59 untranslated region of CAT and a portion of the CAT coding region between nucleotides 7657 and 7876 of TCS, several hundred nucleotides upstream of the STR promoter and 45 nucleotides downstream of the CAT promoter. As shown in Fig. 3B, the genomic and CAT subgenomic RNA species of TCS were detected but the STR subgenomic mRNA of TCS was not. The CAT probe did not detect any RNA species in the Toto1000 sample, since this virus does not contain any CAT sequence. The CAT probe did detect the Lg RNA II species of TCS but failed to detect the Sm RNA II species. This indicates that the 39 end of the Lg RNA II species extends sufficiently beyond the CAT promoter for its detection but the 39 end of the Sm RNA II species does not. To map the 39 end of RNA II more precisely, we stripped off the CAT probe and reprobed the blot a third time with an NSP probe that is complementary to nucleotides 2263 to 25 of the subgenomic promoter. The STR and CAT promoters of TCS have identical sequences between nucleotides 298 and 25. Thus, the NSP probe is complementary to 93 nucleotides of the CAT subgenomic mRNA of TCS and may detect this species. The results presented in Fig. 3C demonstrate that the genomes of TCS and Toto1000 were detected, as expected. The NSP probe did not hybridize to the STR subgenomic mRNA but did give a signal for the CAT subgenomic mRNA of TCS—indicating that the 93 nucleotides of complementary between the NSP probe and the CAT subgenomic mRNA of TCS were sufficient for its detection. The NSP probe also detected the RNA II species of each virus. This result, in combination with those obtained from Fig. 3A and B, demonstrates that the 39 end of each RNA II species lies in close proximity to the position of the subgenomic promoter. 3* end of RNA II. Because the abundance of RNA II is low (Table 1), we polyadenylated the 39 end of this species to facilitate its amplification for sequence analysis. A control experiment was first performed by using runoff transcripts that were synthesized from the XbaI-digested TCS plasmid. In TCS, the XbaI site is 15 nucleotides downstream of the STR promoter. Thus, transcripts which terminate at this site are similar in size to the large RNA II species of TCS. We used 100 fmol of transcript in two separate polyadenylation reactions, either in the presence or in the absence of RNA from uninfected BHK-21 or C7-10 cells. After the polyadenylation reactions, the NotI oligo(dT) primer was used to prime cDNA synthesis and the products were PCR amplified by using the 1940 and NotI oligo(dT) primers (Fig. 4A). Any RNA having a free 39-OH should be polyadenylated and competent for subsequent cDNA synthesis. However, only those products which contain CAT sequences complementary to the 1940 primer should be amplified. The distance from the 59 end of the 1940 primer to the XbaI runoff site is 284 nucleotides, and the length of the NotI oligo(dT) primer is 33 nucleotides. Consequently, a PCR product of no less than 317 bp was expected (Fig. 4A). Indeed, we obtained from each transcript sample (either in the presence or in the absence of RNA from uninfected cells) a PCR product that was slightly larger than 310 bp (data not shown). The results of Fig. 4B show that the sequence of each PCR product is identical to that of the TCS STR promoter and the poly(A) tail was added exactly at the XbaI site. We then treated RNA from TCS-infected cells with this


9113

amplification method. Total RNA (4 mg) was used in the polyadenylation reaction, and RT-PCR was performed as described before. A PCR product of approximately 300 bp was generated (Fig. 4A). The results presented in Fig. 4C show that the sequence of the PCR product was that of the TCS STR promoter, and the poly(A) tail was added to position 24 of the STR promoter. In related experiments, we found that the small RNA II species of TCS and the large RNA II species of TCS-infected CEF and C7-10 cells have sequences identical to that of the STR subgenomic promoter, with poly(A) tails added to position 24 of the subgenomic promoter (data not shown). Taken together, these results suggest that similar mechanisms generate the 39 end of RNA II in different cell types and different subgenomic promoters. We also tested if the 39 end of the large RNA II species in 219/15- and B2-infected cells were different from that obtained with TCS. The size of the PCR product of the B2 sample was approximately 300 bp, while that of 219/15 was ;220 bp (since nucleotides 298 to 220 of the STR promoter are deleted) (data not shown). The abundance of the B2 and 219/15 PCR products was lower than that of the TCS product (data not shown), probably because these viruses have relatively less of the Lg RNA II species than the Sm RNA II species, compared with TCS (Fig. 1A). As shown in Fig. 4D, the sequence of each PCR product is identical to that of the STR promoter of each virus. In both cases, the poly(A) tail was added to position 24 of the STR promoter. These results indicate that the 219/15 subgenomic promoter alone, in the absence of flanking promoter sequences, determines the 39 end of RNA II. Similarly, nucleotide substitutions at positions 213, 210, and 29 of the B2 STR promoter do not affect the 39 end of RNA II. Possible origins of RNA II. Since RNA II is of plus-sense polarity (Fig. 3 and 4), it is conceivable that it might originate from the cleavage of genomic RNA at the position of the subgenomic promoter at any time during the isolation or analysis of viral RNA. We tested this possibility by adding radiolabeled TCS transcripts to Trizol-solubilized monolayers of infected or mock-infected cells. The results of Fig. 5A show that the samples which received TCS transcripts did not generate prominent, RNA II-like species and indicate that the procedures used to isolate and analyze the viral RNA of TCS do not cleave genomic RNA at the position of the subgenomic promoter to yield significant amounts of RNA II-like species. It should be noted that RNA species similar in size to the STR subgenomic mRNA of TCS do accumulate, and these species appear to be in vitro transcripts that prematurely terminate within the nsP coding region (as assessed by Northern blot analysis). The generation of RNA II by some unknown mechanism may require active transcription at the subgenomic promoter, since the appearance of RNA II is correlated with subgenomic mRNA synthesis (Fig. 1 and 2). Thus, inhibition of subgenomic mRNA synthesis should prevent the appearance of RNA II. This hypothesis was tested two ways. Cycloheximide prevents or limits the formation of first-time subgenomic mRNA transcription complexes during the early part of infection but does not affect replication complexes which have formed prior to treatment (61). We used this drug between 2 and 3 hpi (a time at which the formation of subgenomic transcription complexes should be sensitive to cycloheximide treatment) to observe its effect on the appearance of RNA II. At 5 hpi, total RNA was isolated. As shown in Fig. 5B, the samples that received cycloheximide within the first 140 min pi had substantially less subgenomic RNA than did samples receiving later treatments. This correlated with a decrease

9114

WIELGOSZ AND HUANG

J. VIROL.

FIG. 4. Sequence determination of the 39 end of RNA II. (A) Primers used to amplify and sequence the 39 end of the XbaI TCS transcript and the Lg RNA II species of TCS. The 1940 and NotI oligo(dT) primers were used for RT-PCR amplification, and the 1913 primer was used for sequencing. The 59 ends of each primer are shown as gray-filled circles. The lines that are labeled RNA or single-stranded DNA (ss DNA) represent intermediates used in the amplification procedure for each sample but are scaled with respect to the Lg RNA II species of TCS. The coding regions in each PCR product are represented by filled boxes (textured gray for CAT and black for NSP). The open box for the XbaI TCS transcript represents the region downstream of the STR subgenomic mRNA start site (indicated with an arrow). The cDNA species are extended minimally by 33 nucleotides [the length of the NotI oligo(dT) primer]. The line which makes up the right side of each PCR product represents this 33-nucleotide extension. (B) 39-end sequence of XbaI-linearized TCS transcripts, either in the presence or in the absence of RNA from C7-10 cells. Each sequence reads from left to right (G, A, T, and C, respectively). (C) 39-end sequence of the large RNA II species of TCS. (D) 39-end sequences of the large RNA II species of B2 and 219/15, respectively.

in the abundance of RNA II to nearly undetectable levels. At 2 h 40 min pi, the effect of cycloheximide on subgenomic mRNA synthesis was minimal and this corresponded to an increased abundance of RNA II compared to that in samples treated earlier. At 3 hpi, the effect of cycloheximide was not apparent. Indeed, the relative abundances of the subgenomic mRNAs and RNA II to the genomic RNA for the 3-h cycloheximide sample was indistinguishable from that of the untreated sample. We also found that cycloheximide-treated samples that were isolated at 3 versus 5 hpi exhibited similar patterns of inhibited subgenomic mRNA synthesis, and this correlated with a decrease in the abundance of RNA II (data not shown). These results suggest that treatment with cycloheximide during the early part of the infection cycle inhibits the translation of proteins that are required for subgenomic

mRNA synthesis and correspondingly reduces the abundance of RNA II to nearly undetectable levels. The results of Fig. 5B also suggest that genomic RNA, despite having an increased abundance relative to subgenomic mRNA in this experiment, is not cleaved during the isolation and analysis of viral RNA to yield RNA II. The second method we employed to determine the effect of defective subgenomic mRNA synthesis on the abundance of RNA II involved the use of a mutant virus. Toto:ts 21 A2 is RNA negative at the nonpermissive temperature (58) and is defective in subgenomic mRNA synthesis in a temperatureindependent manner (11). The causal mutation of Toto:ts 21 A2 maps to residue 304 of nsP2 but does not affect nsP processing (24). We introduced this mutation into the TCS construct to create TCS-21. Cells were infected with TCS and

VOL. 71, 1997


9115

DISCUSSION

FIG. 5. Possible origins of RNA II. (A) Nonradioactively labeled TCS and mock-infected BHK-21 cells were solubilized with Trizol at 5 hpi. Full length in vitro transcripts of TCS (radiolabeled with [a-32P]GTP) were immediately added to each sample, and RNA was analyzed as described in Materials and Methods). The TCS RNA sample (labeled in vivo with [32P]orthophosphate) was used as a standard to determine the positions of the Lg and Sm RNA II species. (B) BHK-21 cells were infected with TCS and labeled as described in the legend to Fig. 1. At the times listed, cycloheximide was added to samples marked with a plus sign (final concentration, 10 mg/ml). Total RNA was isolated at 5 hpi and analyzed as described in the legend to Fig. 1A. (C) TCS and TCS-21 were used to infect BHK-21 cells at the temperatures shown. At 4 hpi, duplicate cell cultures grown at 30°C were shifted to 40°C (labeled 30-40). All cell cultures were treated with dactinomycin 1 h before labeling with [32P]orthophosphate. Cell labeling was terminated 2 h later, at 8 hpi, and total RNA was isolated and analyzed as described before. For definitions of abbreviations, see the legend to Fig. 1.

TCS-21 at either 30°C (in duplicate) or 40°C. Duplicate cell cultures which were infected at 30°C were shifted to 40°C at 4 hpi. At 8 hpi, total RNA was isolated from each sample. As shown in Fig. 5C, the TCS-21-infected samples maintained at 30°C or shifted to 40°C were noticeably defective in subgenomic mRNA synthesis compared to those of TCS. This corresponded to a substantial decrease in the abundance of RNA II for TCS-21, relative to that of TCS. The TCS-21 sample infected and maintained at 40°C exhibited defective RNA synthesis for all viral RNAs. These results demonstrate that an alteration in the coding region of nsP2 adversely affects subgenomic mRNA transcription, and this correlates to a reduction in the abundance of RNA II. These results are also consistent with those in Fig. 5B and further support the hypothesis that genomic RNA is not cleaved at the position of the promoter to generate RNA II during the isolation and analysis of viral RNA.

The results presented here demonstrate that RNA II is of plus-sense polarity and show that its 39 end maps to the 24 position of the subgenomic promoter. We do not know how the promoter acts to give rise to RNA II. It is possible that RNA secondary structure in the promoter causes the generation of RNA II. However, the 39 end of RNA II is the same, whether it terminates at the TCS, B2, or 219/15 promoter. Thus, any secondary structure which generates RNA II must be present in the minimal promoter itself and be unaffected by mutations in the B2 promoter. Computer analyses failed to detect any secondary structure in the minimal promoter. Alternatively, the promoter may act indirectly to facilitate the binding of specific proteins which cause the generation of RNA II. The observation that both cycloheximide treatment early in infection and the mutation of TCS-21 reduce the amount of subgenomic RNA synthesized relative to genomic RNA, with a corresponding decrease in the abundance of RNA II, implies that proteins are more likely to be involved in the generation of RNA II than is secondary structure inherent to the promoter. There are several mechanisms by which proteins bound at the subgenomic promoter might cause the generation of RNA II. One class of models invokes RNA cleavage as a possible mechanism. Proteins bound at the promoter might stimulate the cleavage of any genomic RNA (at the 24 position of the promoter) that becomes at least locally hybridized to the minus strand. Another possibility is that proteins may recognize and cleave the subgenomic promoter at position 24 or close to it on the minus strand. Thus, the replication of genomic RNA on cleaved minus-strand templates would then terminate at position 24. Yet another mechanism is that proteins bound to the promoter cleave nascent genomic RNA once a replication complex reaches the promoter. All of the cleavage mechanisms must explain the specificity of cleavage to leave a free 39-OH at position 24 of the promoter. Although the generation of RNA II via a cleavage mechanism cannot be ruled out, we favor a model in which the generation of RNA II is a consequence of competition between the replication of genomic RNA and the initiation of subgenomic mRNA synthesis. In this model, it is assumed that replication and transcription occur simultaneously on the same minus-strand template and that subgenomic transcriptional initiation is slow compared to genomic RNA elongation. Three outcomes can be envisioned. First, replication complexes may physically displace the proteins bound to the promoter that were engaged in transcriptional initiation. The net effect may be a reduction in the levels of transcribed subgenomic RNA, and RNA II would not be generated. Second, replication complexes may pause at the promoter because initiation complexes at the promoter would physically block continued genomic RNA elongation. Once transcription had initiated, replication complexes could then resume elongation. The net effect may be a reduction in the levels of replicated genomic RNA. When cells are lysed for RNA isolation, the partially completed genomic RNAs in the stalled replication complexes are released as RNA II molecules. Finally, replication complexes may pause on the promoter but then dissociate from the minus-strand template. Once dissociated, the replication complexes may fall apart and release their nascent genomic RNA as RNA II. RNA II molecules may therefore be dead-end products. It is interesting that if the pausing or stalling model is correct, then the 39 end of RNA II is essentially a “toe print” of whatever protein was bound to the promoter. Given the abundance of RNA II relative to the genomic RNA (one to

9116

WIELGOSZ AND HUANG

every seven; Table 1), both pausing scenarios predict that the promoter poses a significant impediment to genomic RNA synthesis, and this may be a novel mechanism to coordinate replication and transcription. Alternatively, the dissociated replication complex may bind to another minus-strand template to resume and complete the synthesis of the nascent genomic RNA. In this case, RNA II may be a precursor to a recombinant genome that is generated during plus-strand synthesis. Regardless of the mechanism of RNA II generation, RNA II may have an effect on the SIN infection cycle. For example, RNA II may or may not be capped or polyadenylated in vivo, but there are examples in the literature in which RNA species devoid of a 59 cap, poly(A) tail, or both are translation competent (10, 15, 16, 21, 72). If RNA II were translated, nsP4 would be deficient in at least two amino acids at its carboxy terminus. Modified nsP4 could function differently from fulllength nsP4 with respect to viral RNA synthesis. Some of these differences may include altered cis-acting sequence specificity or processivity. Another effect RNA II may have on the SIN infection cycle is decreasing the number of virion particles with full-length genomic RNA. The 39 end of RNA II extends well beyond the putative encapsidation signal (70), so it is not unreasonable to suggest that it can be packaged. Indeed, an RNA species similar in size to that expected for RNA II has been observed in virion particles (14). If RNA II is packaged and is present in a molar ratio of 1 to 10 compared to genomic RNA (Table 1), a 10% decrease in the number of virion particles that contain full-length genomic RNA could lower the efficiency with which a virus population initiates infection. Finally, as mentioned before, RNA II may be a precursor to a recombinant genome. Recombination is one of several mechanisms that a virus population can employ to increase its diversity. If template switching for a replication complex occurred (due to a block imposed by initiation complexes on the subgenomic promoter), one might expect a recombinant genome containing the nsP coding region from one virus genome and an STR coding region from another. ACKNOWLEDGMENTS This work was supported by Public Health Service grant AI26763. We thank C. M. Rice for Toto:ts 21 A2 and the SIN virologists at Washington University for helpful discussions. REFERENCES 1. Ahola, T., and L. Ka ä ¨ria ïnen. 1995. Reaction in alphavirus mRNA capping: formation of a covalent complex of nonstructural protein nsP1 with 7-methyl-GMP. Proc. Natl. Acad. Sci. USA 92:507–511. 2. Alton, N. K., and D. Vapnek. 1979. Nucleotide sequence analysis of the chloramphenicol resistance transposon Tn9. Nature 282:864–869. 3. Barton, D. J., S. G. Sawicki, and D. L. Sawicki. 1988. Demonstration in vitro of temperature-sensitive elongation of RNA in Sindbis virus mutant ts6. J. Virol. 62:3597–3602. 4. Bredenbeek, P. J., I. Frolov, C. M. Rice, and S. Schlesinger. 1993. Sindbis virus expression vectors: packaging of RNA replicons by using defective helper RNAs. J. Virol. 67:6439–6446. 5. Bruton, C. J., and S. I. Kennedy. 1975. Semliki Forest virus intracellular RNA: properties of the multi-stranded RNA species and kinetics of positive and negative strand synthesis. J. Gen. Virol. 28:111–127. 6. Carlson, J., K. Olson, S. Higgs, and B. Beaty. 1995. Molecular genetic manipulation of mosquito vectors. Annu. Rev. Entomol. 40:359–388. 7. Carmichael, G. G., and G. K. McMaster. 1980. The analysis of nucleic acids in gels using glyoxal and acridine orange. Methods Enzymol. 65:380–391. 8. Clegg, J. C., and S. I. Kennedy. 1974. Polyadenylic acid sequences in the virus RNA species of cells infected with Semliki Forest virus. J. Gen. Virol. 22:331–345. 9. Czarniecki, W. C., and T. Sreevalsan. 1979. Sindbis virus RNA replication. I. Properties of the 38S RNA species. J. Gen. Virol. 44:759–771. 10. Danthinne, X., J. Seurinck, F. Meulewaeter, M. M. Van Montagu, and M. Cornelissen. 1993. The 39 untranslated region of satellite tobacco necrosis virus RNA stimulates translation in vitro. Mol. Cell. Biol. 13:3340–3349.

J. VIROL. 11. De´, I., S. G. Sawicki, and D. L. Sawicki. 1996. Sindbis virus RNA-negative mutants that fail to convert from minus-strand to plus-strand synthesis: role of the nsP2 protein. J. Virol. 70:2706–2719. 12. de Groot, R. J., W. R. Hardy, Y. Shirako, and J. H. Strauss. 1990. Cleavagesite preferences of Sindbis virus polyproteins containing the non-structural proteinase. Evidence for temporal regulation of polyprotein processing in vivo. EMBO J. 9:2631–2638. 13. Friedman, R. M., H. B. Levy, and W. B. Carter. 1966. Replication of Semliki Forest virus: three forms of viral RNA produced during infection. Proc. Natl. Acad. Sci. USA 56:440–446. 14. Frolova, E., I. Frolov, and S. Schlesinger. 1997. Packaging signals in alphaviruses. J. Virol. 71:248–258. 15. Gallie, D. R., and M. Kobayashi. 1994. The role of the 39-untranslated region of non-polyadenylated plant viral mRNAs in regulating translational efficiency. Gene 142:159–165. 16. Gallie, D. R., R. L. Tanguay, and V. Leathers. 1995. The tobacco etch viral 59 leader and poly(A) tail are functionally synergistic regulators of translation. Gene 165:233–238. 17. Gorbalenya, A. E., E. V. Koonin, A. P. Donchenko, and V. M. Blinov. 1988. A conserved NTP-motif in putative helicases. Nature 333:22. 18. Gorbalenya, A. E., E. V. Koonin, A. P. Donchenko, and V. M. Blinov. 1989. Two related superfamilies of putative helicases involved in replication, recombination, repair and expression of DNA and RNA genomes. Nucleic Acids Res. 17:4713–4730. 19. Gorbalenya, A. E., E. V. Koonin, and Y. I. Wolf. 1990. A new superfamily of putative NTP-binding domains encoded by genomes of small DNA and RNA viruses. FEBS Lett. 262:145–148. 20. Griffin, D. E. 1986. The Togaviridae and Flaviviridae. Plenum Publishing Corp., New York, N.Y. 21. Gunnery, S., and M. B. Mathews. 1995. Functional mRNA can be generated by RNA polymerase III. Mol. Cell. Biol. 15:3597–3607. 22. Hahn, C. S., C. M. Rice, E. G. Strauss, E. M. Lenches, and J. H. Strauss. 1989. Sindbis virus ts103 has a mutation in glycoprotein E2 that leads to defective assembly of virions. J. Virol. 63:3459–3465. 23. Hahn, Y. S., A. Grakoui, C. M. Rice, E. G. Strauss, and J. H. Strauss. 1989. Mapping of RNA2 temperature-sensitive mutants of Sindbis virus: complementation group F mutants have lesions in nsP4. J. Virol. 63:1194–1202. 24. Hahn, Y. S., E. G. Strauss, and J. H. Strauss. 1989. Mapping of RNA2 temperature-sensitive mutants of Sindbis virus: assignment of complementation groups A, B, and G to nonstructural proteins. J. Virol. 63:3142–3150. 25. Hertz, J. M., and H. V. Huang. 1995. Evolution of the Sindbis virus subgenomic mRNA promoter in cultured cells. J. Virol. 69:7768–7774. 26. Hertz, J. M., and H. V. Huang. 1995. Host-dependent evolution of the Sindbis virus promoter for subgenomic mRNA synthesis. J. Virol. 69:7775– 7781. 27. Hodgman, T. C. 1988. A new superfamily of replicative proteins. Nature 333:22–23. (Erratum, 333:578, 1988.) 28. Kamer, G., and P. Argos. 1984. Primary structural comparison of RNAdependent polymerases from plant, animal and bacterial viruses. Nucleic Acids Res. 12:7269–7282. 29. Kennedy, S. I. 1976. Sequence relationships between the genome and the intracellular RNA species of standard and defective-interfering Semliki Forest virus. J. Mol. Biol. 108:491–511. 30. Kera ¨nen, S., and L. Ka ä ¨ria ïnen. 1979. Functional defects of RNA-negative temperature-sensitive mutants of Sindbis and Semliki Forest viruses. J. Virol. 32:19–29. 31. Kuhn, R. J., Z. Hong, and J. H. Strauss. 1990. Mutagenesis of the 39 nontranslated region of Sindbis virus RNA. J. Virol. 64:1465–1476. 32. Laskey, R. A., and A. D. Mills. 1975. Quantitative film detection of 3H and 14 C in polyacrylamide gels by fluorography. Eur. J. Biochem. 56:335–341. 33. LaStarza, M. W., J. A. Lemm, and C. M. Rice. 1994. Genetic analysis of the nsP3 region of Sindbis virus: evidence for roles in minus-strand and subgenomic RNA synthesis. J. Virol. 68:5781–5791. 34. Lemm, J. A., and C. M. Rice. 1993. Assembly of functional Sindbis virus RNA replication complexes: requirement for coexpression of P123 and P34. J. Virol. 67:1905–1915. 35. Lemm, J. A., and C. M. Rice. 1993. Roles of nonstructural polyproteins and cleavage products in regulating Sindbis virus RNA replication and transcription. J. Virol. 67:1916–1926. 36. Lemm, J. A., T. Rumenapf, E. G. Strauss, J. H. Strauss, and C. M. Rice. 1994. Polypeptide requirements for assembly of functional Sindbis virus replication complexes: a model for the temporal regulation of minus- and plus-strand RNA synthesis. EMBO J. 13:2925–2934. 37. Levin, J. G., and R. M. Friedman. 1971. Analysis of arbovirus ribonucleic acid forms by polyacrylamide gel electrophoresis. J. Virol. 7:504–514. 38. Levis, R., H. Huang, and S. Schlesinger. 1987. Engineered defective interfering RNAs of Sindbis virus express bacterial chloramphenical acetyltransferase in avian cells. Proc. Natl. Acad. Sci. USA 84:4811–4815. 39. Levis, R., S. Schlesinger, and H. V. Huang. 1990. Promoter for Sindbis virus RNA-dependent subgenomic RNA transcription. J. Virol. 64:1726–1733. 40. Levis, R., B. G. Weiss, M. Tsiang, H. Huang, and S. Schlesinger. 1986. Deletion mapping of Sindbis virus DI RNAs derived from cDNAs defines

VOL. 71, 1997 the sequences essential for replication and packaging. Cell 44:137–145. 41. Li, G. P., M. W. LaStarza, W. R. Hardy, J. H. Strauss, and C. M. Rice. 1990. Phosphorylation of Sindbis virus nsP3 in vivo and in vitro. Virology 179:416– 427. 42. Liljestro ¨m, P., and H. Garoff. 1991. A new generation of animal cell expression vectors based on the Semliki Forest virus replicon. Bio/Technology 9:1356–1361. 43. Liljestro ¨m, P., S. Lusa, D. Huylebroeck, and H. Garoff. 1991. In vitro mutagenesis of a full-length cDNA clone of Semliki Forest virus: the small 6,000-molecular-weight membrane protein modulates virus release. J. Virol. 65:4107–4113. 44. Lopez, S., J.-S. Yao, R. J. Kuhn, E. G. Strauss, and J. H. Strauss. 1994. Nucleocapsid-glycoprotein interactions required for assembly of alphaviruses. J. Virol. 68:1316–1323. 45. Mi, S., and V. Stollar. 1991. Expression of Sindbis virus nsP1 and methyltransferase activity in Escherichia coli. Virology 184:423–427. 46. Niklasson, B., and A. Espmark. 1986. Ockelbo disease: arthralgia 3–4 years after infection with a Sindbis virus related agent. Lancet i:1039–1040. 47. Ou, J. H., C. M. Rice, L. Dalgarno, E. G. Strauss, and J. H. Strauss. 1982. Sequence studies of several alphavirus genomic RNAs in the region containing the start of the subgenomic RNA. Proc. Natl. Acad. Sci. USA 79:5235–5239. 48. Pierce, J. S., E. G. Strauss, and J. H. Strauss. 1974. Effect of ionic strength on the binding of Sindbis virus to chick cells. J. Virol. 13:1030–1036. 49. Raju, R., S. V. Subramaniam, and M. Hajjou. 1995. Genesis of Sindbis virus by in vivo recombination of nonreplicative RNA precursors. J. Virol. 69: 7391–7401. 50. Rennels, M. B. 1984. Arthropod-borne virus infections of the central nervous system. Neurol. Clin. 2:241–254. 51. Rice, C. M., R. Levis, J. H. Strauss, and H. V. Huang. 1987. Production of infectious RNA transcripts from Sindbis virus cDNA clones: mapping of lethal mutations, rescue of a temperature-sensitive marker, and in vitro mutagenesis to generate defined mutants. J. Virol. 61:3809–3819. 52. Rikkonen, M., J. Pera ¨nen, and L. Ka ä ¨ria ïnen. 1994. ATPase and GTPase activities associated with Semliki Forest virus nonstructural protein nsP2. J. Virol. 68:5804–5810. 53. Rosemond, H., and T. Sreevalsan. 1973. Viral RNAs associated with ribosomes in Sindbis virus-infected HeLa cells. J. Virol. 11:399–415. 54. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 55. Sarver, N., and V. Stollar. 1977. Sindbis virus-induced cytopathic effect in clones of Aedes albopictus (Singh) cells. Virology 80:390–400. 56. Sawicki, D., D. B. Barkhimer, S. G. Sawicki, C. M. Rice, and S. Schlesinger. 1990. Temperature sensitive shut-off of alphavirus minus strand RNA synthesis maps to a nonstructural protein, nsP4. Virology 174:43–52. 57. Sawicki, D. L., L. Ka ä ¨ria ïnen, C. Lambek, and P. J. Gomatos. 1978. Mechanism for control of synthesis of Semliki Forest virus 26S and 42S RNA. J. Virol. 25:19–27.


9117

58. Sawicki, D. L., and S. G. Sawicki. 1985. Functional analysis of the A complementation group mutants of Sindbis HR virus. Virology 144:20–34. 59. Sawicki, D. L., and S. G. Sawicki. 1993. A second nonstructural protein functions in the regulation of alphavirus negative-strand RNA synthesis. J. Virol. 67:3605–3610. 60. Sawicki, D. L., S. G. Sawicki, S. Kera ¨nen, and L. Ka ä ¨ria ïnen. 1981. Specific Sindbis virus-coded function for minus-strand RNA synthesis. J. Virol. 39: 348–358. 61. Scheele, C. M., and E. R. Pfefferkorn. 1969. Inhibition of interjacent ribonucleic acid (26S) synthesis in cells infected by Sindbis virus. J. Virol. 4:117– 122. 62. Schlesinger, M. J., S. D. London, and C. Ryan. 1993. An in-frame insertion into the Sindbis virus 6K gene leads to defective proteolytic processing of the virus glycoproteins, a transdominant negative inhibition of normal virus formation, and interference in virus shut off of host-cell protein synthesis. Virology 193:424–432. 63. Segal, S., and T. Sreevalsan. 1974. Sindbis virus replicative intermediates: purification and characterization. Virology 59:428–442. 64. Shirako, Y., and J. H. Strauss. 1994. Regulation of Sindbis virus RNA replication: uncleaved P123 and nsP4 function in minus-strand RNA synthesis, whereas cleaved products from P123 are required for efficient plus-strand RNA synthesis. J. Virol. 68:1874–1885. 65. Simmons, D. T., and J. H. Strauss. 1972. Replication of Sindbis virus. I. Relative size and genetic content of 26 s and 49 s RNA. J. Mol. Biol. 71: 599–613. 66. Simmons, D. T., and J. H. Strauss. 1972. II. Multiple forms of doublestranded RNA isolated from infected cells. J. Mol. Biol. 71:615–631. 67. Strauss, E. G., C. R. Birdwell, E. M. Lenches, S. E. Staples, and J. H. Strauss. 1977. Mutants of Sindbis virus. II. Characterization of a maturationdefective mutant, ts103. Virology 82:122–149. 68. Strauss, E. G., R. J. de Groot, R. Levinson, and J. H. Strauss. 1992. Identification of the active site residues in the nsP2 proteinase of Sindbis virus. Virology 191:932–940. 69. Wang, Y.-F., S. G. Sawicki, and D. L. Sawicki. 1991. Sindbis virus nsP1 functions in negative-strand RNA synthesis. J. Virol. 65:985–988. 70. Weiss, B., G. U. Geigenmuller, and S. Schlesinger. 1994. Interactions between Sindbis virus RNAs and a 68 amino acid derivative of the viral capsid protein further define the capsid binding site. Nucleic Acids Res. 22:780–786. 71. Weiss, B., H. Nitschko, I. Ghattas, R. Wright, and S. Schlesinger. 1989. Evidence for specificity in the encapsidation of Sindbis virus RNAs. J. Virol. 63:5310–5318. 72. Wengler, G., and E. Castle. 1986. Analysis of structural properties which possibly are characteristic for the 39-terminal sequence of the genome RNA of flaviviruses. J. Gen. Virol. 67:1183–1188. 73. Xiong, C., R. Levis, P. Shen, S. Schlesinger, C. M. Rice, and H. V. Huang. 1989. Sindbis virus: an efficient, broad host range vector for gene expression in animal cells. Science 243:1188–1191.