- Email: [email protected]
Sindbis virus mutant ts20 of complementation group E contains a lesion in glycoprotein E2
VIROLOGY
151, lo-20 (1986)
Sindbis Virus Mutant ts20 of Complementation Group E Contains a Lesion in Glycoprotein E2 BJijRN
L?ivLsim
H. LINDQVIST; JERRY DISALVO,~ JAMES H. STRAUSS, AND ELLEN
CHARLES G. STRAUSS4
of Biology,
Pasadena,
Cakjbka
Received
November
Institute
of Technology,
20, 1985; accepfxd
January
M. RICE,
Cal$wnia
91125
2.4, 1986
A technique has been devised to readily obtain the entire structural protein region of Sindbis virus cloned into a plasmid vector. This method uses the fact that the nearest site for restriction enzyme Hind111 to the 3’ terminal poly(A) occurs at nucleotides 6266-6271 in the genomie RNA. Inserts extending from the poly(A) tract to this Hind111 site are 5438 nucleotides long (excluding the poly A tract) and contain the entire 4106-nucleotide structural protein region. Using an oligo(dT)-tailed vector as a primer for first strand cDNA synthesis such clones could be obtained in high yield. We were interested in a precise determination of the mutation responsible for the temperature-sensitive phenotype of ts20, a mutant belonging to complementation group E which has a defect in the function of glycoprotein E2 at the nonpermissive temperature. Using this technique we have cloned and sequenced the structural protein region of t-320and of several revertants and concluded that the mutation was a changefrom histidine to Ieucineat amino acid 291of E2. Reversion to temperature insensitivity occurred by same site reversion to the parental nucleotide, restoring the original hi&dine as amino acid 291. Thus, complementation group E of Sindbis virus results from changes in glycoprotein E2 and together with previous results from our laboratory (Arias et aL, 1983;Hahn et aL, 1985) demonstratesthat the three RNA+ complementation groups of Sindbis virus, C, D, and E, result from changes in the three structural proteins of the virus, capsid, glycoprotein El, and glycoprotein E2, respectively. 0 1986 Academic Press. Inc.
INTRODUCTION
proteins are translated from a viral subgenomic mRNA (26 S RNA), cotranslationally inserted into the endoplasmic reticulum, glycosylated, and transported via the Golgi apparatus to the plasma membrane, where they can be used for budding. A large number of temperature-sensitive mutants of Sindbis virus, the type alphavirus, have been isolated and characterized (reviewed in Strauss and Strauss, 1980). They have been grouped by complementation analysis into seven groups, of which three (C, D, and E) are RNA+, that is, they make significant quantities of virus specific RNA at the nonpermissive temperature, and four (A, B, F, G) are RNA-, synthesizing little or no RNA at the nonpermissive temperature. The former are assumed to have lesions in structural proteins found in mature virions, the latter in nonstructural proteins required to replicate the
The alphaviruses contain an icosahedral nucleocapsid consisting of multiple copies of a single species of basic capsid protein complexed with a plus-stranded RNA of 11,703 nucleotides (49 S RNA); the nucleocapsid is surrounded by a lipid envelope containing two virus encoded integral membrane glycoproteins, called El and E2. The envelope is acquired when the capsid, assembled in the cytoplasm, buds through the plasma membrane. The virus glyco‘Present address: Institute of Medical Biology, University of TromsG, 9001 TromsG, Norway. ’ Present address: Merck, Sharp & Dohme Research Laboratories, P.O. Box 2000, Rahway, N.J. 07065. a Present address: Department of Microbiology and Immunology, Washington University, School of Medicine, Box 8093, St. Louis, MO. 63110. ’ To whom reprint requests should be addressed. 0042-6822/86 $3.00 Copyright Q 1986 by Academic Press, Inc. All rights of reproductionin any form reserved.
10
GLYCOPROTEIN
E2 MUTANT
viral RNA. We have previously reported that three group C mutants contain lesions in the capsid protein (Hahn et a& 1985) and that two group D mutants contain lesions in glycoprotein El (Arias et aZ., 1983). Here we report that mutant ts20, the single representative of complementation group E in the collections of Burge and Pfefferkorn (1966a, 1966b) or of Strauss et al. (1976), has a lesion in glycoprotein E2.
OF SINDBIS
VIRUS
11
stock (ts20(2)) and a new revertant, ts20(2)R, was obtained from this stock. ts20(2) and its revertant were not cloned, but were sequenced directly from the RNA by chain termination methods described by Ou et al. (1982) using dideoxynucleoside triphosphates and reverse transcriptase (Sanger et aZ., 1977; Zimmern and Kaesberg, 1978).
Isolation and pur$ication of intracellular RNA. Five to six roller bottles of
Sindbis MATERIALS
AND
METHODS
Growth media, bufers, and chemicals. Growth media and buffers for the production of intracellular Sindbis RNA have been described previously (Ou et aL, 1982). Oligo(dT)-cellulose was purchased from Collaborative Research Inc. Low gelling temperature agarose (SeaPlaque) was obtained from FMC Corporation, Marine Colloids Division, Rockland, Maine. [a32P]Deoxynucleoside triphosphates were from Amersham-Searle and [y-q]adenosine triphosphate was from ICN. Restriction enzymes, T4 or Escherichia coli ligase, T4 polynucleotide kinase, and polymerase I (Klenow fragment) were purchased from New England Biolabs. Avian myeloblastosis virus (AMV) reverse transcriptase was from Life Sciences, Inc. RNase H was from P-L Biochemicals Inc. or Bethesda Research Labs. rH]Uridine was from New England Nuclear. Calf intestine alkaline phosphatase and RNase inhibitor (human placental) were from Boehringer-Mannheim Biochemicals and Promega Biotec, respectively. Terminal deoxynucleotide transferase was from P-L Biochemicals Inc. or New England Nuclear. Virus strains. Sindbis ts20 mutant was originally obtained from B. Burge. Revertants were isolated by plaquing virus stocks at 40” and growing stocks from individual plaques as previously described (Strauss et al, 1976). One revertant (ts20Rl) was found to contain a change in a region other than at the site of the ts lesion (see under Results), and a second revertant (ts20R2) was isolated from the original mutant stock. This proved to be identical to ts2ORl and may not have been an independently arising revertant. The mutant stock was then plaque purified twice to yield a new mutant
secondary chicken cells grown to confluence were infected with Sindbis virus at 30” in the presence of actinomycin D (1 pg/ml). After 3 hr of infection, fresh medium was added containing no actinomycin D and 25 &i/ml of [5,6-3H]uridine was added to one bottle in order to label the Sindbis RNA. At 6.5 hr after infection, the cells were recovered from the bottles, washed, and lysed as described (Ou et a& 1981,1982). The resulting cytoplasmic RNA was then phenolchloroform extracted twice followed by a chloroform extraction and two ether extractions. Finally, the RNA was precipitated in the presence of 0.2 M Na acetate with 2.5 vol of absolute ethanol overnight at -20’. The RNA was collected by centrifugation at 12,000 rpm in a Sorvall SS34 rotor for 30 min. The RNA was resuspended in water, SDS was added to 0.2%, and the solution was heated to 56” for 10 min. The ionic strength was then adjusted to 0.2 M NaCl, 0.01 M EDTA, 0.01 M Tris, pH 7.6, and the solution was applied to an oligo(dT)-cellulose column in the same buffer to select for poly(A)-containing Sindbis 49 S and 26 S RNAs. The poly(A)containing RNA was eluted with glass distilled water containing 0.2% SDS and rechromatographed on the oligo(dT)-cellulose column as before. The RNA was ethanol precipitated twice as before, the last traces of ethanol were removed with a stream of nitrogen gas, and resuspended in CB buffer (1 mM EDTA, 10 mM NaCl, 10 mM Tris, pH 7.5) containing 0.2% SDS. Separation of the 49 S and 26 S viral RNAs was performed on a gradient of 15 to 30% sucrose in CB buffer by centrifugation at 40,000 rpm for 6.5 hr at 20”. The fractions containing 49 S or 26 S RNA respectively, were pooled. The pooled RNA was precipitated twice with absolute ethanol and re-
12
LINDQVIST
suspended in ZO-~1 glass distilled water. Recovery of RNA was 2-4 pg for the 49 S and lo-30 pg for the 26 S. Construction of cDNA clones. Essentially the strategy of Okayama and Berg (1982) was used. First strand synthesis starting from the poly(A) tail of Sindbis virion 49 S RNA was performed with an oligo(dT)tailed plasmid vector. The reaction mixture contained 50 mM Tris-HCl (pH 8.3), 8 mM MgClz, 50 mM KCl, 10 mM DTT, dNTPs at 1 mM. To this was added 0.75 pg of 49 S Sindbis RNA, 100 ng of Proteus 1 T-tailed primer (both heated to 56” for 2 min and rapidly chilled before addition to the reaction mixture), 0.8 pg of actinomycin D, 6.8 units of human placental RNase inhibitor, 2 ~1 of [%P]dATP equal to 20 &i, and finally 13 units of AMV reverse transcriptase. Final volume was 20 ~1, and incubation was at 42.5” for 30 min. The reaction was stopped by addition of 2 ~1 0.25 M EDTA, 1~110% SDS. The mixture was then phenol-chloroform extracted, resuspended in 10 mM Tris, 0.1 mM EDTA, adjusted to 2.0 MNH1-acetate, precipitated twice with 2 vol of ethanol, washed once with 70% ethanol, and dried. The sample was resuspended in 15 ~1 5 mM Tris-HCl, 0.05 mM EDTA (pH 7.4). Second strands were synthesized using E. coli PolI in the presence of E. coli RNase H and DNA ligase. The reaction was performed by addition of PolI buffer to a final concentration of 100 mM HEPES, pH 6.9, 10 mM MgClz, 6 mM P-mercaptoethanol, 70 mM KU, 1.5 mM P-NAD, dNTPs to a final concentration of 1 PM, 20 &i of r2P]dATP, followed by the addition of 2 units of E. coli ligase, 24 units of E. coli PolI and 1 unit of E. coli RNase H. The total volume was 40 ~1. Incubation took place at 12” for 1 hr followed by another hour at room temperature. The reaction was stopped by addition of EDTA and SDS to 0.025 M and 1%) respectively, and 5 pg yeast tRNA. The mixture was then extracted with phenol-chloroform and precipitated twice with ethanol as described for the first strand synthesis. For Hind111 digestion and ligation, the double-stranded cDNA was digested with Hind111 and closed with either T4 or E. coli DNA ligase. The DNA from the second
ET AL.
strand synthesis was resuspended in 9 ~1 of 5 mM Tris-HCl, 0.05 mM EDTA, 1~1 of 10X medium salt restriction enzyme buffer (as recommended by the manufacturer), and 1.4 units of HindIII. Incubation was at 37” for 60 min. The reaction mixture was extracted with phenol-chloroform, precipitated twice, and ligations were performed under standard conditions. Competent E. coli. (strain MC1061) were transformed with the ligation mixture and plated on petri plates containing 25 @g/ml ampicillin. Characterization of plasmid DNA. The ampicillin resistant colonies were screened for the presence of the appropriate insert by preparing plasmid DNA from 2 ml overnight cultures by phenol extraction and examining the size of the insert after digestion with Hind111 and EcoRI. Alternatively, an alkaline lysis procedure (Maniatis et al., 1982) was used to prepare supercoiled plasmid DNA from cells adhering to a toothpick touched to the surface of resistant colonies and examining the size of the supercoiled DNA. In either case the DNA was electrophoresed on small 0.8% agarose gels (5 X 7.5 cm) and visualized by uv light after staining with ethidium bromide. Sequence determination of DNA. DNA sequencing was carried out by the procedure of Maxam and Gilbert (1980) using 40- and 80-cm-long gels (Sanger and Coulson, 1978). Fragments generated by restriction endonucleases were end labeled either with [y-32P]dATP using T4 polynueleotide kinase or when possible were labeled by fill-in synthesis of 5’ overhangs using the Klenow fragment of PolI and the appropriate [a-32P]dNTP. After a second restriction digestion to generate fragments labeled on only one end, the DNA fragments were recovered from acrylamide gels by electroelution (Strauss et aL, 1984) or from low gelling agarose gels by phenol extraction of the melted gel (Maniatis et al, 1982). RESULTS
Constructicm of Sindbis cDNA clones for sequence analysis. Clones containing cDNA copies of the structural region of ts20 and its revertants were constructed by a mod-
GLYCOPROTEIN
E2 MUTANT
ification of the methods of Okayama and Berg (1982). The plasmid vector used was constructed by H. V. Huang (unpublished). It was derived originally from pBR322 but is only 2332 bp long. It contains an origin of replication, an ampicillin resistance gene as a selectable marker, a promoter region for Sp6 RNA polymerase, and a polylinker containing unique restriction sites for a number of restriction enzymes including HindIII, S&I, and SmuI. Purified plasmid DNA was cleaved with SmaI, which leaves blunt ends, and (dT)-tailed with terminal deoxynucleotidyl transferase to an average length of 52 nucleotides (Fig. 1). One of the oligo(dT) tails was removed by cutting (in the polylinker) with restriction enzyme SalI, and the singly tailed vector was purified by preparative electrophoresis and chromatography on oligo(dA)-cellulose. This oligo(dT)-tailed vector was then used as a primer for synthesis of cDNA on Sindbis virion poly(A)+ RNA templates. Priming of first strand synthesis occurs with poly(T) at the poly(A) tract of the viral RNA, and second strand synthesis is primed by oligoribonucleotides produced by RNase H treatment of the RNA:DNA hybrids, using the methods of Okayama and Berg (1982). The double-stranded cDNA synthesized must be longer than 5.5 kb in order to extend beyond the first Hind111 site in the viral RNA. RNase H and E. coli ligase are included in the second strand synthesis reaction to remove the original RNA template and to close single strand nicks, respectively (such that the double-stranded cDNA product is covalently attached to the plasmid primer in both strands). The product is then cut with Hind111 and reclosed with E. coli ligase. The use of E. eoli ligase in this step is preferable to T4 ligase since the E. coli enzyme ligates blunt ends inefficiently and requires cohesive ends such as those produced by Hind111 for efficient activity under the conditions used. Thus, molecules containing Hind111 sites at both ends are selectively closed and the background of clones produced by blunt end ligation when T4 DNA ligase is used is effectively eliminated. After transformation of E. coli and selection for ampicillin resistance, resistant colonies were picked at random and
OF SINDBIS
13
VIRUS
~01~ (dT1 TAILING
+ 3’
5’ -Aa
495 AMEN. RUOP
RNA
FIRST STRAND SYNTHESIS
SECOND STRAND SYNTHESIS
Hind111 E. COLI LISASE
SINOSIS
cONA
VECTOR
FIG. 1. Diagrammatic representation of the cloning strategy for structural protein mutants. The vector primer is made by digesting plasmid DNA with .%uI, adding oligo(dT) to each end using terminal deoxynucleotide transferase and removing one tail with SalI. Following purification as described under Results, the vector was annealed to the poly(A) tail at the 3’ end of purified 49 S virion RNA. First strand and second strand cDNA synthesis are described under Materials and Methods. Following Hind111 digestion and ligation with E. co15ligase the DNA is used to transform competent E. coli cells.
the plasmid DNA was examined by agarose gel electrophoresis in minigels for the presence of a correctly sized insert. For the best preparations, more than 90% of resistant colonies were found to have a plasmid with an insert of 5438 nucleotides (plus a variable length poly(A) tract) that extended from the poly(A) tract at the 3’ end of virion RNA to the Hind111 site at nucleotides 6266-6271 of the virion RNA; this
LINDQVIST
14
ET AL.
structural proteins are encoded by nucleotides 7646 to 11381 of the genomic RNA. The proteins are encoded in order 5’ to 3’ as follows: capsid (nucleotides 7646 to 8438), E3 (8439 to 8630), E2 (8631 to 9899), 6K (9900 to 10064) and El (10065 to 11381). Cloned cDNA from ts20 and ts+ revertants isolated from ts20 was sequenced by the methods of Maxam and Gilbert (1980). The complete structural region of ts20 was sequenced. A number of changes were found that could have arisen from cloning of variants in the population and/or cloning artifacts and these are discussed in more detail below. Only changes in E2 were found which correlated with the ts20 phenotype, and the details of changes found in E2 are shown schematically in Fig. 2. Each of the changes presented in Fig. 2 was shown to be present in more than one clone of cDNA from any virus stock so as to eliminate minor variants or cloning artifacts. Sequences are compared to the sequence deduced for the ancestral HR strain [denoted HR(66)] from which the mutants were derived (Arias et aa, 1983); this sequence is the same as that previously determined for an HR small plaque strain of Sindbis virus (Strauss et al, 1984). Mutant ts20 contains two nucleotide
region encompasses the genes for the three structural proteins of Sindbis virus, which are encoded in nucleotides 7947 to 11381 of the genomic RNA. When T4 ligase was used in the closure step, however, only about 10% of resistant colonies contained an insert of the proper size. With appropriate choice of restriction enzymes and plasmid vectors, this method is generally applicable to any alphavirus or to any poly(A)-containing RNA in order to selectively clone a portion of the 3’ terminal region. Sequence analysis of ts.20. The complete nucleotide sequence of the genome of Sindbis virus has been determined (Strauss et ah, 1984) and the overall genome organization and translation strategy of the alphaviruses is known (reviewed in Strauss and Strauss, 1986). The three structural proteins are translated from a subgenomic 26 S mRNA whose sequence has been published (Rice and Strauss, 1981), and which begins with nucleotide 7599 of the genomic RNA and continues to the end of the genomic RNA at nucleotide 11,703. The amino acid sequence obtained from the N termini of the structural proteins (Bell et aZ., 1978) has been aligned with the amino acid sequence deduced from the nucleotide sequence. From these data we know the
HR (66) HR ISPI
/
ts20
I
HIS
GLU 200
291
u
u I
LEU 291 ts20R1 ts20R2
A
,
I LYS
ts20
I21
ts20
(21 R
u 200 Y LEU 291
FIG. 2. Changed nucleotides found in glycoprotein E2 (nt 8631-9899). The top line indicates the ancestral sequence of Sindbis HR which in this region is coincident with the HR(SP) sequence previously reported (Strauss et aL, 1934). For ts20, ts20R1, and ts2OR2, which were sequenced as cloned copies, only changes found in at least two independent clones are shown; for these strains the entire E2 region was sequenced. For ts20(2) and ts20(2)R, dideoxy sequencing was performed on an RNA template (and thus the sequence found represents the average sequence in the RNA population); for these strains only the regions delineated by the short lines were sequenced. In the representation for HR(66) all nucleotides altered in any strain below are indicated, for HR(66) the ancestral nucleotide is shown above the line and the encoded amino acid (if altered) is shown below the line. In the strains below, any change from the ancestral sequence is shown; if no change is indicated, the nucleotide (and amino acid) is the same as in the parental HR(66) sequence. Numbering of nucleotides is from the 5’ terminus of the Sindbis genome; numbering of amino acids is from the N terminus of E2.
GLYCOPROTEIN
E2 MUTANT
changes from the HR(66) sequence in the E2 region. One is a G - IJ change at nucleotide 9890, which is a silent change in the codon for Ser-420 of E2 (UCG - UCU); this change was also present in all revertants of ts20 examined and is characteristic of the ts20 strain. The second change is in the codon for His-291 of E2, leading to its replacement by Leu-291 (CAC - CUC). In three different revertants examined, this nucleotide has reverted to the parental nucleotide, restoring His at position 291. We conclude that the ts lesion in ts20 is the replacement of His-291 in glycoprotein E2 by Leu. The first revertant of ts20 examined, labeled tsBOR1, had an additional coding change, the substitution of Glu-200 by Lys (GAG --t AAG). Although this constitutes a significant change in charge on E2, it seemed unlikely to have anything to do with the ts lesion because ts20 possessed the ancestral sequence in this region and because the region around Glu-200 is quite variable in E2, being involved in antigenic variation in Sindbis virus (E. G. Strauss and A. L. Schmaljohn, unpublished data). Nevertheless, this change was examined in more detail to ensure that it represented strain variation unrelated to temperature sensitivity. A second, independent revertant was isolated from the ts20 stock and sequenced through this region. This revertant had the same sequence as tsZOR1, with His as amino acid 291 and Lys as amino acid 200 (strain ts20R2 in Fig. 2). To further probe the change at amino acid 200, the ts20 mutant was plaque purified, grown, and a new revertant isolated from this plaque purified stock. This new isolate of ts20, referred to as ts20(2), and its revertant, ts20(2)R, were sequenced in the E2 region through the two changes of interest, using dideoxy sequencing (Fig. 2). In this procedure, synthetic DNA oligonucleotides were used as primers for reverse transcription of the RNA templates with dideoxynucleotides in specific chain termination reactions (Zimmern and Kaesberg, 1978; Ou et aZ., 1982). ts20(2) again showed the change from His-291 to Leu, which reverted to His in t.s20(2)R. Both ts20(2) and ts20(2)R possessed the ancestral
OF SINDBIS
15
VIRUS
Glu at residue 200. Thus, although at least two independent revertants isolated from stock ts20(1) possessed Lys instead of Glu at residue 200, plaque purification of the stock followed by isolation of a new revertant showed that this change was a strain difference unrelated to temperature sensitivity or reversion to temperature insensitivity. Characteristics of glycoprotein E2 relative to the ts20 mutation. In Fig. 3 we have plotted the ts20 lesion on a schematic representation of the E2 glycoprotein which illustrates some of the key features. In the
-4 I 1
I 2
I
I
3
4
FIG. 3. Analysis of E2 showing several features of the protein. The upper panel is a similarity comparison of the amino acid sequences of the E2 glycoproteins of Sindbis (Rice and Strauss, 1981) and Ross River (Dalgarno et al, 1983) viruses plotted as a moving average using a string length of 20 residues. Circles with C or S indicate glycosylation sites which in Sindbis normally have complex and simple carbohydrate chains attached, respectively (Mayne et aL, 1985). The site of the t&Xl mutation is indicated and the unlabeled arrow indicates the location of the ts20R change. The shaded area is a variable region in which a number of antigenic epitopes have been localized (E. G. Strauss and A. L. Schmaljohn, unpublished). The lower panels are a hydrophobicity plots of Sindbis E2 (middle panel) and Ross River E2 (lower panel) using a string length of 7 and the algorithm of Kyte and Doolittle (1982). Hydrophobic regions are above the center line; hydrophilic domains are plotted below the line. Amino acids are numbered in hundreds from the beginning of E2.
16
LINDQVIST
upper panel is shown a homology profile for the E2 glycoproteins of two distantly related alphaviruses, Sindbis and Ross River viruses. The ts20 mutation is in a region of moderate homology (-50%). E2 is the least conserved of the three structural proteins of alphaviruses. The average homology between Sindbis and Ross River viruses for the entire structural region is 48% (Dalgarno et al, 1983), whereas for E2 it is 42% .The ts20 mutation is near the second glycosylation site in Sindbis E2, which usually carries a simple oligosaecharide chain (circled S) (Mayne et al., 1985) and is not conserved as to position in alphaviruses (Rice and Strauss, 1981). In the bottom panels it can be seen that the ts20 lesion is localized in a region that on balance is not markedly hydrophobic or hydrophilic but just C terminal to a hydrophobic domain and N terminal to a hydrophilic domain. The unlabeled arrow in Fig. 3 is the location of the additional change found in ts20Rl and ts20R2. This is close to the conserved glycosylation site carrying complex carbohydrates (circled C). This change is located within a more variable region of E2 (shown by shaded overlay) which is highly charged and contains neutralization epitopes of the virus (E. G. Strauss and A. L. Schmaljohn, unpublished). Strain variation and cloning artifacts. In addition to the changes in E2 discussed above, a number of other changes were found that can be classified into three categories: minor variants (present in only one clone and probably arising from cloning of variants in the population, although reverse transcriptase errors or other cloning artifacts cannot be excluded); strain differences (changes found in all clones from a stock, such as the silent change in the codon for Ser-420 of E2 discussed above, that may have been induced during the original mutagenesis or that might arise when variants in the population are plaque purified during work with the virus mutants); and cloning artifacts (in two cases discussed below, changes were found that seem unlikely to have been present in the original RNA population and that may have arisen during cloning). In the first category, we found the following four
ET AL.
changes: a U - C change at nucleotide 8063 of the virion RNA in one clone of ts2ORl (a silent change in the codon for Phe-139 of the capsid protein); an A - G change at nucleotide 9173 in one clone of ts20(1) (a silent change in Glu-181 of glycoprotein E2); a U - C change at nucleotide 9647 of one clone of ts20(1) (a silent change in Tyr339 of E2); and a G - U change at nucleotide 10379 of one strain of ts20Rl (which results in a change of Glu-105 to Asp in glycoprotein El). We cannot rule out cloning artifacts as the source of these changed nucleotides. However, since three are silent changes and one results in a conservative amino acid substitution (Asp in place of Glu) and since these changes were each observed in only a single clone and were not present in other clones derived from the same cDNA preparation, we feel it is more likely that they represent minor variants in the population, rather than cloning artifacts. In the second category, in addition to the silent change in the codon for Ser-420 of E2 of all ts20 strains and revertants sequenced and the Glu - Lys change in the E2 of some of the revertants that were discussed above, we found two other changes in ts20. Both are silent changes and although only the mutant clones were examined in these regions these changes appear to be characteristic of the ts20 strains used. These two changes were a U - C at nucleotide 7664 (a silent change in the codon for Phe-6 of the capsid protein), and a C - U change in nucleotide 11030 (a silent change in the codon for Ser-323 of glycoprotein El). The three silent changes in ts20 found could have been produced by the original mutagenesis or could have resulted from plaque purification of ts20 stocks when variants could have been isolated. Finally, we found changes in two clones that seem likely to have resulted from artifacts that arose during some stage of cloning, possibly during synthesis of double-stranded cDNA. Both were deletions of an A residue in a string of A residues in the RNA; one in the sequence CAAAG (nucleotides 9187-9191) [cloned sequence CAAG in one clone from ts20(1)] and one in the sequence CAAAAAC (nucleotides
GLYCOPROTEIN
E2 MUTANT
11272-11278) [cloned sequence CAAAAC in one clone from tsZORl]. These deletions were not sequencing artifacts caused by compressions since samples from other clones present on the same gel clearly showed the correct sequence. It is possible that these clones arose from defective RNA molecules produced by the viral replicase, but such an RNA would clearly be nonviable because translation of protein would be out of phase past the deletion, and thus these RNAs could not replicate and persist in the population. Therefore, it seems more likely that the deletion occurred at some stage involved in cDNA synthesis or cloning. DISCUSSION
The lesicm in ts.20.Mutant ts20 was isolated by Burge and Pfefferkorn (1966a) after treatment of an HR stock with nitrosoguanidine. The work here demonstrates that the temperature sensitivity of ts20 results from an A - U change at nucleotide 9502 of the virion RNA that results in His291 of glycoprotein E2 being changed to Leu-291. This change is not one expected to arise from the action of nitrosoguanidine (Drake, 19’70), but as we have discussed previously, this agent causes many different types of changes (Arias et al, 1983) and an A - U change after nitrosoguanidine treatment has been found to be the cause of temperature sensitivity in Sindbis mutant ts13 (Hahn et ar, 1985). ts20 was also found to have three silent changes in the region sequenced, one of U - C at nucleotide 7664, one of C - U at nucleotide 11030, and one of G - U at nucleotide 9890. These changes may also have arisen during the original mutagenesis, or they may represent spontaneous changes that were immortalized during plaque purification of stocks. In cells infected with ts20 at the nonpermissive temperature, glycoproteins PE2 (the precursor to E2) and El are made, inserted into the endoplasmic reticulum, and transported to the glyeosylated, plasma membrane (reviewed in Strauss and Strauss, 1980). Nucleocapsids are
OF SINDBIS
VIRUS
17
formed in the cytoplasm and can interact specifically with the cytoplasmic domains of the glycoproteins, of which only PE2 has a significant cytoplasmic domain (Wirth et al, 1977; Garoff et al, 1980; Rice and Strauss, 1981), as shown by an accumulation of capsids on the inner surface of the plasma membrane (Brown, 1980). No budding occurs, however, and infectious virions are not produced. Thus, many of the functions of PE2 are normal, but the ts lesion prevents the cleavage of PE2 to E2 and E3 and the final maturation of progeny virions. The His - Leu change at residue 291 presumably leads to incorrect folding of PE2 at the nonpermissive temperature that still allows it to be transported to the plasma membrane and to interact with capsids (and El?), but that prevents the final maturation process. It is unclear whether an inability of PE2 to be cleaved is the cause of the phenotype or whether the inability to bud, caused perhaps by faulty interactions with El, leads to noncleavage. It has been found that PE2 can be cleaved in the absence of nucleocapsid formation (Huth et al., 1984; Rice et al, 1985) or under conditions where little virus is formed (Mayne et aZ., 1984) suggesting that cleavage precedes budding, but the issue is not completely resolved at present. Mutants in complementation group E are comparatively rare. ts20 is the only member of this group found by Burge and Pfefferkorn (1966b) or by Strauss et aZ.(1976); Durbin and Stoller (1984) have recently reported on a second member of this group. Although it is possible that the lack of ts mutations in group E reflects a failure to complement on the part of most mutations in E2 (or PE2), it seems more likely that the window for ts changes in E2 is small. E2 is the least conserved of the structural proteins among alphaviruses. E2 may have arisen during evolution as a changeable protein that interacts with the vertebrate immune system. E2 has been found to carry the major neutralization epitopes of alphaviruses (Dalrymple et a& 1976; Cole et al., 1982), and changes in sequence leading to changes in interactions with monoclonal antibodies are readily accommodated without apparent effect on the growth of
18
LINDQVIST
the virus (A. L. Schmaljohn, personal communication). Mutants in the three RNA+ complementation groups of Sindbis, C, D, and E, have now been sequenced. Group C mutants possess changes in the capsid protein (Hahn et aL, 1985), group D mutants are changed in El (Arias et aL, 1983), and group E mutants are changed in E2 (this work; see also Durbin and Stollar, 1984), and thus each RNA+ complementation group corresponds to one of the virion structural proteins. Reversion to temperature insensitivity. In all of the ts20 revertants examined, reversion has been accomplished by a change in the mutated nucleotide to the parental nucleotide, restoring the parental His-291. This has been our experience with Sindbis ts mutants in general (Arias et aZ., 1983; Hahn et al, 1985). Of six mutants sequenced to date which possessed a total of seven amino acid changes because one mutant contained two changes, reversion in all cases but one has been accomplished by restoring the parental sequence. The one exception involved a mutation in El that resulted from two nucleotide changes in a single codon, such that a Lys in the parental sequence was replaced by a Gly in the mutant (Arias et al, 1983). Reversion to the parental sequence thus would require two nucleotide changes, and the revertant sequenced was found to have a single nucleotide change in the affected codon such that Arg replaced the Gly; in this case the pseudorevertant in which the parental Lys was replaced with Arg was more probable than the true revertant. The revertant of ts20 that was accompanied by a change of Glu-200 to Lys might provide an explanation as to why pseudorevertants, whether same site or second site, have not been seen more often. The finding that revertant populations from ts20 derived from two independent plaques were accompanied by the exogenous change of Glu-200 to Lys suggests that these two plaques were clonally related; that is, that reversion in this case first occurred in a variant in which (an innocuous) change at position 200 had already occurred, and this revertant was then amplified in the population. RNA vi-
ET AL.
ruses are highly evolved for rapid growth; any change that enables a virus to grow even slightly faster will be positively selected for during virus evolution. Thus, any change from this sequence, especially one which causes conformational changes in the protein rendering them inactive at 40”, will be more or less disabling. A true revertant, once it arises, will in general outgrow the mutant, even at the permissive temperature, and be amplified in the population. If pseudorevertants (that Iead to plaque formation at 40”) do occur, they may not possess this replication advantage leading to amplification, and thus would represent a small fraction of t.s+revertants in a population. The isolation of true revertants has made the interpretation of which lesions are responsible for temperature sensitivity quite straightforward. Variants and cloning artifacts. In this study, a number of variants, mostly containing silent mutations, have been cloned. This suggests that, given the high rate of RNA mutation because of lack of proofreading functions by RNA replicases, silent changes occur often and persist in the RNA population. Changes that seriously impair the ability of the virus to replicate are quickly screened out during growth and are unlikely to be isolated. This means that Sindbis genome RNA consists of a population of molecules with a defined average sequence but in which each individual RNA molecule can possesschanges from this sequence. Those changes which confer only slight selection advantage or disadvantage can be fixed in the population during plaque purification. In addition, as culture conditions are changed, the selective pressures operating may be different, leading to selection of variants (Strauss and Strauss, 1980). We also isolated two clones that contained deletions which may be due to the cloning procedure. The deletion of a nucleotide in these clones, whether due to a cloning artifact or to the cloning of a defective variant in the population, would lead to a serious misinterpretation of the coding potential of the RNA if only a single clone were being examined, and points out the need to confirm sequence results from
GLYCOPROTEIN
E2 MUTANT
cloned copies of RNA by sequencing more than one clone or by other methods. ACKNOWLEDGMENTS We acknowledge the expert technical assistance of Ms. Edith Lenches who maintained the virus stocks and isolated the revertants studied here. One of us (B.H.L.) acknowledges a travel grant from the Norwegian Research Council for Science and Humanities in connection with a sabbatical stay at Caltech. This work was supported in part by Grants AI 10793 and AI 20612 from the National Institutes of Health and by Grant PCM 8316856 from the National Science Foundation, and by Biomedical Research Support Grant RR 07003. REFERENCES ARIAS, C., BELL, J. R., LENCHES, E. M., STRAUSS, E. G., and STRAUSS, J. H. (1983). Sequence analysis of two mutants of Sindbis virus defective in the intracellular transport of their glycoproteins. J. MoL Bid 168,87-102. BELL, J. R., HUNKAPILLER, M. W., HOOD, L. E., and STRAUSS, J. H. (19’78). Amino-terminal sequence analysis of the structural proteins of Sindbis virus. Prw. Natl. Acad Sk USA 75,2722-2726. BROWN, D. T. (1980). The assembly of alphaviruses. In “The Togaviruses: Biology, Structure, Replication” (R. W. Schlesinger, ed.), Chap. 17, pp. 4’73-501. Academic Press, New York. BURGE, B. W., and PFEFFERKORN, E. R. (1966a). Isolation and characterization of conditional-lethal mutants of Sindbis virus. Virology 30,204-213. BURGE, B. W., and PFEFFERKORN, E. R. (1966b). Complementation between temperature-sensitive mutants of Sindbis virus. Virology 30,214-223. COLE, G. A., SCHMALJOHN, A. L., and DALRYMPLE, J. M. (1982). Protection against lethal Sindbis virus infection with specific monoclonal antibodies. In “Viral Diseases in South-east Asia and the Western Pacific” (J. S. Mackenzie, ed.), pp. 54-545. Academic Press, Orlando, Fla./Sydney. DALGARNO, L., RICE, C. M., and STRAUSS, J. H. (1983). Ross River virus 26s RNA: Complete nucleotide sequence and deduced sequence of the encoded structural proteins. Virology 129,170-187. DALRYMPLE, J., SCHLESINGER, S., and RUSSELL, P. K. (1976). Antigenic characterization of two Sindbis envelope glycoproteins separated by isoelectric focusing. Virology 69,93-103. DRAKE, G. R. (1970). “The Molecular Basis of Mutation,” pp. 146-159. Holden-Day, San Francisco. DURBIN, R. K., and STOLLAR, V. (1984). A mutant of Sindbis virus with a host-dependent defect in mat-
OF SINDBIS
VIRUS
19
uration associated with hyperglycosylation of E2. Virobgy 135,331~344. GAROFF, H., FRISCHAUF, A.-M., SIMONS, K., LEHRACH, H., and DELIUS, H. (1980). Nucleotide sequence of cDNA coding for Semliki Forest virus membrane glycoproteins. Nature (Lund& 288,235-241. HAHN, C. S., STRAUSS,E. G., and STRAUSS,J. H. (1985). Sequence analysis of three Sindbis virus mutants temperature-sensitive in the capsid protein autoprotease. Proc. Natl. Ad Sci USA 82,4648-4652. HUTH, A., RAPOPORT,T. A., and KAARIAINEN, L. (1984). Envelope proteins of Semliki Forest virus synthesized in Xenopus oocytes are transported to the cell surface. EMBO J. 3,767-771. KYTE, J., and DOOLITTLE, R. F. (1982). A simple method for displaying the hydropathic character of a protein. J. Mot Biol 157,105-132. MANIATIS, T., FRITSCH, E. F., and SAMBROOK,J. (1982). “Molecular Cloning, a Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. MAXAM, A. M., and GILBERT, W. (1980). Sequencing end-labeled DNA with base-specific chemical cleavages. In “Methods in Enzymology” (L. Grossman and K. Moldave, eds.), Vol. 65, pp. 499-560. Academic Press, Orlando, Fla. MAYNE, J. T., BELL, J. R., STRAUSS,E. G., and STRAUSS, J. H. (1985). Pattern of gylcosylation of Sindbis virus envelope proteins synthesized in hamster and chicken cells. Virology 142, 121-133. MAYNE, J. T., RICE, C. M., STRAUSS, E. G., HUNKAPILLER, M. W., and STRAUSS, J. H. (1984). Biochemical studies of the maturation of the small Sindbis virus glycoprotein E3. Virology 134,338-357. OKAYAMA, H., and BERG, P. (1982). High-efficiency cloning of full-length cDNA. Mol. CeU Biol. 2,161170. Ou, J.-H., STRAUSS, E. G., and STRAUSS, J. H. (1981). Comparative studies of the 3’ terminal sequences of several alphavirus RNAs. Virology 109,281-289. Ou, J.-H., TRENT, D. W., and STRAUSS, J. H. (1982). The 3’ non-coding regions of alphavirus RNAs contain repeating sequences. J. Mol. Biol 156,719-730. RICE, C. M., FRANKE, C. A., STRAUSS,J. H., and HRUBY, D. E. (1985). Expression of Sindbis virus structural proteins via recombinant vaccinia virus: Synthesis, processing, and incorporation into mature Sindbis virions. J. Vird 56,227-239. RICE, C. M., and STRAUSS, J. H. (1981). Nucleotide sequence of the 26s mRNA of Sindbis virus and deduced sequence of the encoded virus structural proteins. Proc. Natl Ad Sci USA 78,2062-2066. SANGER, F., and COULSON,A. R. (1978). The use of thin acrylamide gels for DNA sequencing. FEBS Lett 87,107-110. SANGER, F., NICKLEN, S., and COULSON,R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Amd. Sci. USA ?4,5463-5467.
20
LINDQVIST
STRAUSS, E. G., LENCHES, E. M., and STRAUSS, J. H. (1976). Mutants of Sindbis virus. I. Isolation and partial characterization of 89 new temperaturesensitive mutants. virology 74,154-168. STRAUSS, E. G., RICE, C. M., and STRAUSS,J. H. (1984). Complete nucleotide sequence of the genomic RNA of Sindbis virus. Vidogy 133, 92-110. STRAUSS, E. G., and STRAUSS, J. H. (1980). Mutants of alphaviruses: Genetics and physiology. In “The Togaviruses: Biology, Structure, Replication” (R. W. Schlesinger, ed.), Chap. 14, pp. 393-426. Academic Press, New York. STRAUSS, E. G., and STRAUSS, J. H. (1986). Structure
ET AL. and replication of the alphavirus genome. In “The Togaviridae and Flaviviridae” (S. Schlesinger and M. J. Schlesinger, eds.), Chap. 3, pp. 35-90. Plenum, New York. WIRTH, D. F., KATZ, F., SMALL, B., and LODISH, H. (19’77). How a single Sindbis virus mRNA directs the synthesis of one soluble protein and two integral membrane glycoproteins. Cell 10,253-263. ZIMMERN, D., and KAESBERG, P. (1978). 3’-Terminal nucleotide sequence of encephalomyocarditis virus RNA determined by reverse transcriptase and chain-terminating inhibitors. Proc. Nati Auw! Sk. USA 75,4257-4261.
151, lo-20 (1986)
Sindbis Virus Mutant ts20 of Complementation Group E Contains a Lesion in Glycoprotein E2 BJijRN
L?ivLsim
H. LINDQVIST; JERRY DISALVO,~ JAMES H. STRAUSS, AND ELLEN
CHARLES G. STRAUSS4
of Biology,
Pasadena,
Cakjbka
Received
November
Institute
of Technology,
20, 1985; accepfxd
January
M. RICE,
Cal$wnia
91125
2.4, 1986
A technique has been devised to readily obtain the entire structural protein region of Sindbis virus cloned into a plasmid vector. This method uses the fact that the nearest site for restriction enzyme Hind111 to the 3’ terminal poly(A) occurs at nucleotides 6266-6271 in the genomie RNA. Inserts extending from the poly(A) tract to this Hind111 site are 5438 nucleotides long (excluding the poly A tract) and contain the entire 4106-nucleotide structural protein region. Using an oligo(dT)-tailed vector as a primer for first strand cDNA synthesis such clones could be obtained in high yield. We were interested in a precise determination of the mutation responsible for the temperature-sensitive phenotype of ts20, a mutant belonging to complementation group E which has a defect in the function of glycoprotein E2 at the nonpermissive temperature. Using this technique we have cloned and sequenced the structural protein region of t-320and of several revertants and concluded that the mutation was a changefrom histidine to Ieucineat amino acid 291of E2. Reversion to temperature insensitivity occurred by same site reversion to the parental nucleotide, restoring the original hi&dine as amino acid 291. Thus, complementation group E of Sindbis virus results from changes in glycoprotein E2 and together with previous results from our laboratory (Arias et aL, 1983;Hahn et aL, 1985) demonstratesthat the three RNA+ complementation groups of Sindbis virus, C, D, and E, result from changes in the three structural proteins of the virus, capsid, glycoprotein El, and glycoprotein E2, respectively. 0 1986 Academic Press. Inc.
INTRODUCTION
proteins are translated from a viral subgenomic mRNA (26 S RNA), cotranslationally inserted into the endoplasmic reticulum, glycosylated, and transported via the Golgi apparatus to the plasma membrane, where they can be used for budding. A large number of temperature-sensitive mutants of Sindbis virus, the type alphavirus, have been isolated and characterized (reviewed in Strauss and Strauss, 1980). They have been grouped by complementation analysis into seven groups, of which three (C, D, and E) are RNA+, that is, they make significant quantities of virus specific RNA at the nonpermissive temperature, and four (A, B, F, G) are RNA-, synthesizing little or no RNA at the nonpermissive temperature. The former are assumed to have lesions in structural proteins found in mature virions, the latter in nonstructural proteins required to replicate the
The alphaviruses contain an icosahedral nucleocapsid consisting of multiple copies of a single species of basic capsid protein complexed with a plus-stranded RNA of 11,703 nucleotides (49 S RNA); the nucleocapsid is surrounded by a lipid envelope containing two virus encoded integral membrane glycoproteins, called El and E2. The envelope is acquired when the capsid, assembled in the cytoplasm, buds through the plasma membrane. The virus glyco‘Present address: Institute of Medical Biology, University of TromsG, 9001 TromsG, Norway. ’ Present address: Merck, Sharp & Dohme Research Laboratories, P.O. Box 2000, Rahway, N.J. 07065. a Present address: Department of Microbiology and Immunology, Washington University, School of Medicine, Box 8093, St. Louis, MO. 63110. ’ To whom reprint requests should be addressed. 0042-6822/86 $3.00 Copyright Q 1986 by Academic Press, Inc. All rights of reproductionin any form reserved.
10
GLYCOPROTEIN
E2 MUTANT
viral RNA. We have previously reported that three group C mutants contain lesions in the capsid protein (Hahn et a& 1985) and that two group D mutants contain lesions in glycoprotein El (Arias et aZ., 1983). Here we report that mutant ts20, the single representative of complementation group E in the collections of Burge and Pfefferkorn (1966a, 1966b) or of Strauss et al. (1976), has a lesion in glycoprotein E2.
OF SINDBIS
VIRUS
11
stock (ts20(2)) and a new revertant, ts20(2)R, was obtained from this stock. ts20(2) and its revertant were not cloned, but were sequenced directly from the RNA by chain termination methods described by Ou et al. (1982) using dideoxynucleoside triphosphates and reverse transcriptase (Sanger et aZ., 1977; Zimmern and Kaesberg, 1978).
Isolation and pur$ication of intracellular RNA. Five to six roller bottles of
Sindbis MATERIALS
AND
METHODS
Growth media, bufers, and chemicals. Growth media and buffers for the production of intracellular Sindbis RNA have been described previously (Ou et aL, 1982). Oligo(dT)-cellulose was purchased from Collaborative Research Inc. Low gelling temperature agarose (SeaPlaque) was obtained from FMC Corporation, Marine Colloids Division, Rockland, Maine. [a32P]Deoxynucleoside triphosphates were from Amersham-Searle and [y-q]adenosine triphosphate was from ICN. Restriction enzymes, T4 or Escherichia coli ligase, T4 polynucleotide kinase, and polymerase I (Klenow fragment) were purchased from New England Biolabs. Avian myeloblastosis virus (AMV) reverse transcriptase was from Life Sciences, Inc. RNase H was from P-L Biochemicals Inc. or Bethesda Research Labs. rH]Uridine was from New England Nuclear. Calf intestine alkaline phosphatase and RNase inhibitor (human placental) were from Boehringer-Mannheim Biochemicals and Promega Biotec, respectively. Terminal deoxynucleotide transferase was from P-L Biochemicals Inc. or New England Nuclear. Virus strains. Sindbis ts20 mutant was originally obtained from B. Burge. Revertants were isolated by plaquing virus stocks at 40” and growing stocks from individual plaques as previously described (Strauss et al, 1976). One revertant (ts20Rl) was found to contain a change in a region other than at the site of the ts lesion (see under Results), and a second revertant (ts20R2) was isolated from the original mutant stock. This proved to be identical to ts2ORl and may not have been an independently arising revertant. The mutant stock was then plaque purified twice to yield a new mutant
secondary chicken cells grown to confluence were infected with Sindbis virus at 30” in the presence of actinomycin D (1 pg/ml). After 3 hr of infection, fresh medium was added containing no actinomycin D and 25 &i/ml of [5,6-3H]uridine was added to one bottle in order to label the Sindbis RNA. At 6.5 hr after infection, the cells were recovered from the bottles, washed, and lysed as described (Ou et a& 1981,1982). The resulting cytoplasmic RNA was then phenolchloroform extracted twice followed by a chloroform extraction and two ether extractions. Finally, the RNA was precipitated in the presence of 0.2 M Na acetate with 2.5 vol of absolute ethanol overnight at -20’. The RNA was collected by centrifugation at 12,000 rpm in a Sorvall SS34 rotor for 30 min. The RNA was resuspended in water, SDS was added to 0.2%, and the solution was heated to 56” for 10 min. The ionic strength was then adjusted to 0.2 M NaCl, 0.01 M EDTA, 0.01 M Tris, pH 7.6, and the solution was applied to an oligo(dT)-cellulose column in the same buffer to select for poly(A)-containing Sindbis 49 S and 26 S RNAs. The poly(A)containing RNA was eluted with glass distilled water containing 0.2% SDS and rechromatographed on the oligo(dT)-cellulose column as before. The RNA was ethanol precipitated twice as before, the last traces of ethanol were removed with a stream of nitrogen gas, and resuspended in CB buffer (1 mM EDTA, 10 mM NaCl, 10 mM Tris, pH 7.5) containing 0.2% SDS. Separation of the 49 S and 26 S viral RNAs was performed on a gradient of 15 to 30% sucrose in CB buffer by centrifugation at 40,000 rpm for 6.5 hr at 20”. The fractions containing 49 S or 26 S RNA respectively, were pooled. The pooled RNA was precipitated twice with absolute ethanol and re-
12
LINDQVIST
suspended in ZO-~1 glass distilled water. Recovery of RNA was 2-4 pg for the 49 S and lo-30 pg for the 26 S. Construction of cDNA clones. Essentially the strategy of Okayama and Berg (1982) was used. First strand synthesis starting from the poly(A) tail of Sindbis virion 49 S RNA was performed with an oligo(dT)tailed plasmid vector. The reaction mixture contained 50 mM Tris-HCl (pH 8.3), 8 mM MgClz, 50 mM KCl, 10 mM DTT, dNTPs at 1 mM. To this was added 0.75 pg of 49 S Sindbis RNA, 100 ng of Proteus 1 T-tailed primer (both heated to 56” for 2 min and rapidly chilled before addition to the reaction mixture), 0.8 pg of actinomycin D, 6.8 units of human placental RNase inhibitor, 2 ~1 of [%P]dATP equal to 20 &i, and finally 13 units of AMV reverse transcriptase. Final volume was 20 ~1, and incubation was at 42.5” for 30 min. The reaction was stopped by addition of 2 ~1 0.25 M EDTA, 1~110% SDS. The mixture was then phenol-chloroform extracted, resuspended in 10 mM Tris, 0.1 mM EDTA, adjusted to 2.0 MNH1-acetate, precipitated twice with 2 vol of ethanol, washed once with 70% ethanol, and dried. The sample was resuspended in 15 ~1 5 mM Tris-HCl, 0.05 mM EDTA (pH 7.4). Second strands were synthesized using E. coli PolI in the presence of E. coli RNase H and DNA ligase. The reaction was performed by addition of PolI buffer to a final concentration of 100 mM HEPES, pH 6.9, 10 mM MgClz, 6 mM P-mercaptoethanol, 70 mM KU, 1.5 mM P-NAD, dNTPs to a final concentration of 1 PM, 20 &i of r2P]dATP, followed by the addition of 2 units of E. coli ligase, 24 units of E. coli PolI and 1 unit of E. coli RNase H. The total volume was 40 ~1. Incubation took place at 12” for 1 hr followed by another hour at room temperature. The reaction was stopped by addition of EDTA and SDS to 0.025 M and 1%) respectively, and 5 pg yeast tRNA. The mixture was then extracted with phenol-chloroform and precipitated twice with ethanol as described for the first strand synthesis. For Hind111 digestion and ligation, the double-stranded cDNA was digested with Hind111 and closed with either T4 or E. coli DNA ligase. The DNA from the second
ET AL.
strand synthesis was resuspended in 9 ~1 of 5 mM Tris-HCl, 0.05 mM EDTA, 1~1 of 10X medium salt restriction enzyme buffer (as recommended by the manufacturer), and 1.4 units of HindIII. Incubation was at 37” for 60 min. The reaction mixture was extracted with phenol-chloroform, precipitated twice, and ligations were performed under standard conditions. Competent E. coli. (strain MC1061) were transformed with the ligation mixture and plated on petri plates containing 25 @g/ml ampicillin. Characterization of plasmid DNA. The ampicillin resistant colonies were screened for the presence of the appropriate insert by preparing plasmid DNA from 2 ml overnight cultures by phenol extraction and examining the size of the insert after digestion with Hind111 and EcoRI. Alternatively, an alkaline lysis procedure (Maniatis et al., 1982) was used to prepare supercoiled plasmid DNA from cells adhering to a toothpick touched to the surface of resistant colonies and examining the size of the supercoiled DNA. In either case the DNA was electrophoresed on small 0.8% agarose gels (5 X 7.5 cm) and visualized by uv light after staining with ethidium bromide. Sequence determination of DNA. DNA sequencing was carried out by the procedure of Maxam and Gilbert (1980) using 40- and 80-cm-long gels (Sanger and Coulson, 1978). Fragments generated by restriction endonucleases were end labeled either with [y-32P]dATP using T4 polynueleotide kinase or when possible were labeled by fill-in synthesis of 5’ overhangs using the Klenow fragment of PolI and the appropriate [a-32P]dNTP. After a second restriction digestion to generate fragments labeled on only one end, the DNA fragments were recovered from acrylamide gels by electroelution (Strauss et aL, 1984) or from low gelling agarose gels by phenol extraction of the melted gel (Maniatis et al, 1982). RESULTS
Constructicm of Sindbis cDNA clones for sequence analysis. Clones containing cDNA copies of the structural region of ts20 and its revertants were constructed by a mod-
GLYCOPROTEIN
E2 MUTANT
ification of the methods of Okayama and Berg (1982). The plasmid vector used was constructed by H. V. Huang (unpublished). It was derived originally from pBR322 but is only 2332 bp long. It contains an origin of replication, an ampicillin resistance gene as a selectable marker, a promoter region for Sp6 RNA polymerase, and a polylinker containing unique restriction sites for a number of restriction enzymes including HindIII, S&I, and SmuI. Purified plasmid DNA was cleaved with SmaI, which leaves blunt ends, and (dT)-tailed with terminal deoxynucleotidyl transferase to an average length of 52 nucleotides (Fig. 1). One of the oligo(dT) tails was removed by cutting (in the polylinker) with restriction enzyme SalI, and the singly tailed vector was purified by preparative electrophoresis and chromatography on oligo(dA)-cellulose. This oligo(dT)-tailed vector was then used as a primer for synthesis of cDNA on Sindbis virion poly(A)+ RNA templates. Priming of first strand synthesis occurs with poly(T) at the poly(A) tract of the viral RNA, and second strand synthesis is primed by oligoribonucleotides produced by RNase H treatment of the RNA:DNA hybrids, using the methods of Okayama and Berg (1982). The double-stranded cDNA synthesized must be longer than 5.5 kb in order to extend beyond the first Hind111 site in the viral RNA. RNase H and E. coli ligase are included in the second strand synthesis reaction to remove the original RNA template and to close single strand nicks, respectively (such that the double-stranded cDNA product is covalently attached to the plasmid primer in both strands). The product is then cut with Hind111 and reclosed with E. coli ligase. The use of E. eoli ligase in this step is preferable to T4 ligase since the E. coli enzyme ligates blunt ends inefficiently and requires cohesive ends such as those produced by Hind111 for efficient activity under the conditions used. Thus, molecules containing Hind111 sites at both ends are selectively closed and the background of clones produced by blunt end ligation when T4 DNA ligase is used is effectively eliminated. After transformation of E. coli and selection for ampicillin resistance, resistant colonies were picked at random and
OF SINDBIS
13
VIRUS
~01~ (dT1 TAILING
+ 3’
5’ -Aa
495 AMEN. RUOP
RNA
FIRST STRAND SYNTHESIS
SECOND STRAND SYNTHESIS
Hind111 E. COLI LISASE
SINOSIS
cONA
VECTOR
FIG. 1. Diagrammatic representation of the cloning strategy for structural protein mutants. The vector primer is made by digesting plasmid DNA with .%uI, adding oligo(dT) to each end using terminal deoxynucleotide transferase and removing one tail with SalI. Following purification as described under Results, the vector was annealed to the poly(A) tail at the 3’ end of purified 49 S virion RNA. First strand and second strand cDNA synthesis are described under Materials and Methods. Following Hind111 digestion and ligation with E. co15ligase the DNA is used to transform competent E. coli cells.
the plasmid DNA was examined by agarose gel electrophoresis in minigels for the presence of a correctly sized insert. For the best preparations, more than 90% of resistant colonies were found to have a plasmid with an insert of 5438 nucleotides (plus a variable length poly(A) tract) that extended from the poly(A) tract at the 3’ end of virion RNA to the Hind111 site at nucleotides 6266-6271 of the virion RNA; this
LINDQVIST
14
ET AL.
structural proteins are encoded by nucleotides 7646 to 11381 of the genomic RNA. The proteins are encoded in order 5’ to 3’ as follows: capsid (nucleotides 7646 to 8438), E3 (8439 to 8630), E2 (8631 to 9899), 6K (9900 to 10064) and El (10065 to 11381). Cloned cDNA from ts20 and ts+ revertants isolated from ts20 was sequenced by the methods of Maxam and Gilbert (1980). The complete structural region of ts20 was sequenced. A number of changes were found that could have arisen from cloning of variants in the population and/or cloning artifacts and these are discussed in more detail below. Only changes in E2 were found which correlated with the ts20 phenotype, and the details of changes found in E2 are shown schematically in Fig. 2. Each of the changes presented in Fig. 2 was shown to be present in more than one clone of cDNA from any virus stock so as to eliminate minor variants or cloning artifacts. Sequences are compared to the sequence deduced for the ancestral HR strain [denoted HR(66)] from which the mutants were derived (Arias et aa, 1983); this sequence is the same as that previously determined for an HR small plaque strain of Sindbis virus (Strauss et al, 1984). Mutant ts20 contains two nucleotide
region encompasses the genes for the three structural proteins of Sindbis virus, which are encoded in nucleotides 7947 to 11381 of the genomic RNA. When T4 ligase was used in the closure step, however, only about 10% of resistant colonies contained an insert of the proper size. With appropriate choice of restriction enzymes and plasmid vectors, this method is generally applicable to any alphavirus or to any poly(A)-containing RNA in order to selectively clone a portion of the 3’ terminal region. Sequence analysis of ts.20. The complete nucleotide sequence of the genome of Sindbis virus has been determined (Strauss et ah, 1984) and the overall genome organization and translation strategy of the alphaviruses is known (reviewed in Strauss and Strauss, 1986). The three structural proteins are translated from a subgenomic 26 S mRNA whose sequence has been published (Rice and Strauss, 1981), and which begins with nucleotide 7599 of the genomic RNA and continues to the end of the genomic RNA at nucleotide 11,703. The amino acid sequence obtained from the N termini of the structural proteins (Bell et aZ., 1978) has been aligned with the amino acid sequence deduced from the nucleotide sequence. From these data we know the
HR (66) HR ISPI
/
ts20
I
HIS
GLU 200
291
u
u I
LEU 291 ts20R1 ts20R2
A
,
I LYS
ts20
I21
ts20
(21 R
u 200 Y LEU 291
FIG. 2. Changed nucleotides found in glycoprotein E2 (nt 8631-9899). The top line indicates the ancestral sequence of Sindbis HR which in this region is coincident with the HR(SP) sequence previously reported (Strauss et aL, 1934). For ts20, ts20R1, and ts2OR2, which were sequenced as cloned copies, only changes found in at least two independent clones are shown; for these strains the entire E2 region was sequenced. For ts20(2) and ts20(2)R, dideoxy sequencing was performed on an RNA template (and thus the sequence found represents the average sequence in the RNA population); for these strains only the regions delineated by the short lines were sequenced. In the representation for HR(66) all nucleotides altered in any strain below are indicated, for HR(66) the ancestral nucleotide is shown above the line and the encoded amino acid (if altered) is shown below the line. In the strains below, any change from the ancestral sequence is shown; if no change is indicated, the nucleotide (and amino acid) is the same as in the parental HR(66) sequence. Numbering of nucleotides is from the 5’ terminus of the Sindbis genome; numbering of amino acids is from the N terminus of E2.
GLYCOPROTEIN
E2 MUTANT
changes from the HR(66) sequence in the E2 region. One is a G - IJ change at nucleotide 9890, which is a silent change in the codon for Ser-420 of E2 (UCG - UCU); this change was also present in all revertants of ts20 examined and is characteristic of the ts20 strain. The second change is in the codon for His-291 of E2, leading to its replacement by Leu-291 (CAC - CUC). In three different revertants examined, this nucleotide has reverted to the parental nucleotide, restoring His at position 291. We conclude that the ts lesion in ts20 is the replacement of His-291 in glycoprotein E2 by Leu. The first revertant of ts20 examined, labeled tsBOR1, had an additional coding change, the substitution of Glu-200 by Lys (GAG --t AAG). Although this constitutes a significant change in charge on E2, it seemed unlikely to have anything to do with the ts lesion because ts20 possessed the ancestral sequence in this region and because the region around Glu-200 is quite variable in E2, being involved in antigenic variation in Sindbis virus (E. G. Strauss and A. L. Schmaljohn, unpublished data). Nevertheless, this change was examined in more detail to ensure that it represented strain variation unrelated to temperature sensitivity. A second, independent revertant was isolated from the ts20 stock and sequenced through this region. This revertant had the same sequence as tsZOR1, with His as amino acid 291 and Lys as amino acid 200 (strain ts20R2 in Fig. 2). To further probe the change at amino acid 200, the ts20 mutant was plaque purified, grown, and a new revertant isolated from this plaque purified stock. This new isolate of ts20, referred to as ts20(2), and its revertant, ts20(2)R, were sequenced in the E2 region through the two changes of interest, using dideoxy sequencing (Fig. 2). In this procedure, synthetic DNA oligonucleotides were used as primers for reverse transcription of the RNA templates with dideoxynucleotides in specific chain termination reactions (Zimmern and Kaesberg, 1978; Ou et aZ., 1982). ts20(2) again showed the change from His-291 to Leu, which reverted to His in t.s20(2)R. Both ts20(2) and ts20(2)R possessed the ancestral
OF SINDBIS
15
VIRUS
Glu at residue 200. Thus, although at least two independent revertants isolated from stock ts20(1) possessed Lys instead of Glu at residue 200, plaque purification of the stock followed by isolation of a new revertant showed that this change was a strain difference unrelated to temperature sensitivity or reversion to temperature insensitivity. Characteristics of glycoprotein E2 relative to the ts20 mutation. In Fig. 3 we have plotted the ts20 lesion on a schematic representation of the E2 glycoprotein which illustrates some of the key features. In the
-4 I 1
I 2
I
I
3
4
FIG. 3. Analysis of E2 showing several features of the protein. The upper panel is a similarity comparison of the amino acid sequences of the E2 glycoproteins of Sindbis (Rice and Strauss, 1981) and Ross River (Dalgarno et al, 1983) viruses plotted as a moving average using a string length of 20 residues. Circles with C or S indicate glycosylation sites which in Sindbis normally have complex and simple carbohydrate chains attached, respectively (Mayne et aL, 1985). The site of the t&Xl mutation is indicated and the unlabeled arrow indicates the location of the ts20R change. The shaded area is a variable region in which a number of antigenic epitopes have been localized (E. G. Strauss and A. L. Schmaljohn, unpublished). The lower panels are a hydrophobicity plots of Sindbis E2 (middle panel) and Ross River E2 (lower panel) using a string length of 7 and the algorithm of Kyte and Doolittle (1982). Hydrophobic regions are above the center line; hydrophilic domains are plotted below the line. Amino acids are numbered in hundreds from the beginning of E2.
16
LINDQVIST
upper panel is shown a homology profile for the E2 glycoproteins of two distantly related alphaviruses, Sindbis and Ross River viruses. The ts20 mutation is in a region of moderate homology (-50%). E2 is the least conserved of the three structural proteins of alphaviruses. The average homology between Sindbis and Ross River viruses for the entire structural region is 48% (Dalgarno et al, 1983), whereas for E2 it is 42% .The ts20 mutation is near the second glycosylation site in Sindbis E2, which usually carries a simple oligosaecharide chain (circled S) (Mayne et al., 1985) and is not conserved as to position in alphaviruses (Rice and Strauss, 1981). In the bottom panels it can be seen that the ts20 lesion is localized in a region that on balance is not markedly hydrophobic or hydrophilic but just C terminal to a hydrophobic domain and N terminal to a hydrophilic domain. The unlabeled arrow in Fig. 3 is the location of the additional change found in ts20Rl and ts20R2. This is close to the conserved glycosylation site carrying complex carbohydrates (circled C). This change is located within a more variable region of E2 (shown by shaded overlay) which is highly charged and contains neutralization epitopes of the virus (E. G. Strauss and A. L. Schmaljohn, unpublished). Strain variation and cloning artifacts. In addition to the changes in E2 discussed above, a number of other changes were found that can be classified into three categories: minor variants (present in only one clone and probably arising from cloning of variants in the population, although reverse transcriptase errors or other cloning artifacts cannot be excluded); strain differences (changes found in all clones from a stock, such as the silent change in the codon for Ser-420 of E2 discussed above, that may have been induced during the original mutagenesis or that might arise when variants in the population are plaque purified during work with the virus mutants); and cloning artifacts (in two cases discussed below, changes were found that seem unlikely to have been present in the original RNA population and that may have arisen during cloning). In the first category, we found the following four
ET AL.
changes: a U - C change at nucleotide 8063 of the virion RNA in one clone of ts2ORl (a silent change in the codon for Phe-139 of the capsid protein); an A - G change at nucleotide 9173 in one clone of ts20(1) (a silent change in Glu-181 of glycoprotein E2); a U - C change at nucleotide 9647 of one clone of ts20(1) (a silent change in Tyr339 of E2); and a G - U change at nucleotide 10379 of one strain of ts20Rl (which results in a change of Glu-105 to Asp in glycoprotein El). We cannot rule out cloning artifacts as the source of these changed nucleotides. However, since three are silent changes and one results in a conservative amino acid substitution (Asp in place of Glu) and since these changes were each observed in only a single clone and were not present in other clones derived from the same cDNA preparation, we feel it is more likely that they represent minor variants in the population, rather than cloning artifacts. In the second category, in addition to the silent change in the codon for Ser-420 of E2 of all ts20 strains and revertants sequenced and the Glu - Lys change in the E2 of some of the revertants that were discussed above, we found two other changes in ts20. Both are silent changes and although only the mutant clones were examined in these regions these changes appear to be characteristic of the ts20 strains used. These two changes were a U - C at nucleotide 7664 (a silent change in the codon for Phe-6 of the capsid protein), and a C - U change in nucleotide 11030 (a silent change in the codon for Ser-323 of glycoprotein El). The three silent changes in ts20 found could have been produced by the original mutagenesis or could have resulted from plaque purification of ts20 stocks when variants could have been isolated. Finally, we found changes in two clones that seem likely to have resulted from artifacts that arose during some stage of cloning, possibly during synthesis of double-stranded cDNA. Both were deletions of an A residue in a string of A residues in the RNA; one in the sequence CAAAG (nucleotides 9187-9191) [cloned sequence CAAG in one clone from ts20(1)] and one in the sequence CAAAAAC (nucleotides
GLYCOPROTEIN
E2 MUTANT
11272-11278) [cloned sequence CAAAAC in one clone from tsZORl]. These deletions were not sequencing artifacts caused by compressions since samples from other clones present on the same gel clearly showed the correct sequence. It is possible that these clones arose from defective RNA molecules produced by the viral replicase, but such an RNA would clearly be nonviable because translation of protein would be out of phase past the deletion, and thus these RNAs could not replicate and persist in the population. Therefore, it seems more likely that the deletion occurred at some stage involved in cDNA synthesis or cloning. DISCUSSION
The lesicm in ts.20.Mutant ts20 was isolated by Burge and Pfefferkorn (1966a) after treatment of an HR stock with nitrosoguanidine. The work here demonstrates that the temperature sensitivity of ts20 results from an A - U change at nucleotide 9502 of the virion RNA that results in His291 of glycoprotein E2 being changed to Leu-291. This change is not one expected to arise from the action of nitrosoguanidine (Drake, 19’70), but as we have discussed previously, this agent causes many different types of changes (Arias et al, 1983) and an A - U change after nitrosoguanidine treatment has been found to be the cause of temperature sensitivity in Sindbis mutant ts13 (Hahn et ar, 1985). ts20 was also found to have three silent changes in the region sequenced, one of U - C at nucleotide 7664, one of C - U at nucleotide 11030, and one of G - U at nucleotide 9890. These changes may also have arisen during the original mutagenesis, or they may represent spontaneous changes that were immortalized during plaque purification of stocks. In cells infected with ts20 at the nonpermissive temperature, glycoproteins PE2 (the precursor to E2) and El are made, inserted into the endoplasmic reticulum, and transported to the glyeosylated, plasma membrane (reviewed in Strauss and Strauss, 1980). Nucleocapsids are
OF SINDBIS
VIRUS
17
formed in the cytoplasm and can interact specifically with the cytoplasmic domains of the glycoproteins, of which only PE2 has a significant cytoplasmic domain (Wirth et al, 1977; Garoff et al, 1980; Rice and Strauss, 1981), as shown by an accumulation of capsids on the inner surface of the plasma membrane (Brown, 1980). No budding occurs, however, and infectious virions are not produced. Thus, many of the functions of PE2 are normal, but the ts lesion prevents the cleavage of PE2 to E2 and E3 and the final maturation of progeny virions. The His - Leu change at residue 291 presumably leads to incorrect folding of PE2 at the nonpermissive temperature that still allows it to be transported to the plasma membrane and to interact with capsids (and El?), but that prevents the final maturation process. It is unclear whether an inability of PE2 to be cleaved is the cause of the phenotype or whether the inability to bud, caused perhaps by faulty interactions with El, leads to noncleavage. It has been found that PE2 can be cleaved in the absence of nucleocapsid formation (Huth et al., 1984; Rice et al, 1985) or under conditions where little virus is formed (Mayne et aZ., 1984) suggesting that cleavage precedes budding, but the issue is not completely resolved at present. Mutants in complementation group E are comparatively rare. ts20 is the only member of this group found by Burge and Pfefferkorn (1966b) or by Strauss et aZ.(1976); Durbin and Stoller (1984) have recently reported on a second member of this group. Although it is possible that the lack of ts mutations in group E reflects a failure to complement on the part of most mutations in E2 (or PE2), it seems more likely that the window for ts changes in E2 is small. E2 is the least conserved of the structural proteins among alphaviruses. E2 may have arisen during evolution as a changeable protein that interacts with the vertebrate immune system. E2 has been found to carry the major neutralization epitopes of alphaviruses (Dalrymple et a& 1976; Cole et al., 1982), and changes in sequence leading to changes in interactions with monoclonal antibodies are readily accommodated without apparent effect on the growth of
18
LINDQVIST
the virus (A. L. Schmaljohn, personal communication). Mutants in the three RNA+ complementation groups of Sindbis, C, D, and E, have now been sequenced. Group C mutants possess changes in the capsid protein (Hahn et aL, 1985), group D mutants are changed in El (Arias et aL, 1983), and group E mutants are changed in E2 (this work; see also Durbin and Stollar, 1984), and thus each RNA+ complementation group corresponds to one of the virion structural proteins. Reversion to temperature insensitivity. In all of the ts20 revertants examined, reversion has been accomplished by a change in the mutated nucleotide to the parental nucleotide, restoring the parental His-291. This has been our experience with Sindbis ts mutants in general (Arias et aZ., 1983; Hahn et al, 1985). Of six mutants sequenced to date which possessed a total of seven amino acid changes because one mutant contained two changes, reversion in all cases but one has been accomplished by restoring the parental sequence. The one exception involved a mutation in El that resulted from two nucleotide changes in a single codon, such that a Lys in the parental sequence was replaced by a Gly in the mutant (Arias et al, 1983). Reversion to the parental sequence thus would require two nucleotide changes, and the revertant sequenced was found to have a single nucleotide change in the affected codon such that Arg replaced the Gly; in this case the pseudorevertant in which the parental Lys was replaced with Arg was more probable than the true revertant. The revertant of ts20 that was accompanied by a change of Glu-200 to Lys might provide an explanation as to why pseudorevertants, whether same site or second site, have not been seen more often. The finding that revertant populations from ts20 derived from two independent plaques were accompanied by the exogenous change of Glu-200 to Lys suggests that these two plaques were clonally related; that is, that reversion in this case first occurred in a variant in which (an innocuous) change at position 200 had already occurred, and this revertant was then amplified in the population. RNA vi-
ET AL.
ruses are highly evolved for rapid growth; any change that enables a virus to grow even slightly faster will be positively selected for during virus evolution. Thus, any change from this sequence, especially one which causes conformational changes in the protein rendering them inactive at 40”, will be more or less disabling. A true revertant, once it arises, will in general outgrow the mutant, even at the permissive temperature, and be amplified in the population. If pseudorevertants (that Iead to plaque formation at 40”) do occur, they may not possess this replication advantage leading to amplification, and thus would represent a small fraction of t.s+revertants in a population. The isolation of true revertants has made the interpretation of which lesions are responsible for temperature sensitivity quite straightforward. Variants and cloning artifacts. In this study, a number of variants, mostly containing silent mutations, have been cloned. This suggests that, given the high rate of RNA mutation because of lack of proofreading functions by RNA replicases, silent changes occur often and persist in the RNA population. Changes that seriously impair the ability of the virus to replicate are quickly screened out during growth and are unlikely to be isolated. This means that Sindbis genome RNA consists of a population of molecules with a defined average sequence but in which each individual RNA molecule can possesschanges from this sequence. Those changes which confer only slight selection advantage or disadvantage can be fixed in the population during plaque purification. In addition, as culture conditions are changed, the selective pressures operating may be different, leading to selection of variants (Strauss and Strauss, 1980). We also isolated two clones that contained deletions which may be due to the cloning procedure. The deletion of a nucleotide in these clones, whether due to a cloning artifact or to the cloning of a defective variant in the population, would lead to a serious misinterpretation of the coding potential of the RNA if only a single clone were being examined, and points out the need to confirm sequence results from
GLYCOPROTEIN
E2 MUTANT
cloned copies of RNA by sequencing more than one clone or by other methods. ACKNOWLEDGMENTS We acknowledge the expert technical assistance of Ms. Edith Lenches who maintained the virus stocks and isolated the revertants studied here. One of us (B.H.L.) acknowledges a travel grant from the Norwegian Research Council for Science and Humanities in connection with a sabbatical stay at Caltech. This work was supported in part by Grants AI 10793 and AI 20612 from the National Institutes of Health and by Grant PCM 8316856 from the National Science Foundation, and by Biomedical Research Support Grant RR 07003. REFERENCES ARIAS, C., BELL, J. R., LENCHES, E. M., STRAUSS, E. G., and STRAUSS, J. H. (1983). Sequence analysis of two mutants of Sindbis virus defective in the intracellular transport of their glycoproteins. J. MoL Bid 168,87-102. BELL, J. R., HUNKAPILLER, M. W., HOOD, L. E., and STRAUSS, J. H. (19’78). Amino-terminal sequence analysis of the structural proteins of Sindbis virus. Prw. Natl. Acad Sk USA 75,2722-2726. BROWN, D. T. (1980). The assembly of alphaviruses. In “The Togaviruses: Biology, Structure, Replication” (R. W. Schlesinger, ed.), Chap. 17, pp. 4’73-501. Academic Press, New York. BURGE, B. W., and PFEFFERKORN, E. R. (1966a). Isolation and characterization of conditional-lethal mutants of Sindbis virus. Virology 30,204-213. BURGE, B. W., and PFEFFERKORN, E. R. (1966b). Complementation between temperature-sensitive mutants of Sindbis virus. Virology 30,214-223. COLE, G. A., SCHMALJOHN, A. L., and DALRYMPLE, J. M. (1982). Protection against lethal Sindbis virus infection with specific monoclonal antibodies. In “Viral Diseases in South-east Asia and the Western Pacific” (J. S. Mackenzie, ed.), pp. 54-545. Academic Press, Orlando, Fla./Sydney. DALGARNO, L., RICE, C. M., and STRAUSS, J. H. (1983). Ross River virus 26s RNA: Complete nucleotide sequence and deduced sequence of the encoded structural proteins. Virology 129,170-187. DALRYMPLE, J., SCHLESINGER, S., and RUSSELL, P. K. (1976). Antigenic characterization of two Sindbis envelope glycoproteins separated by isoelectric focusing. Virology 69,93-103. DRAKE, G. R. (1970). “The Molecular Basis of Mutation,” pp. 146-159. Holden-Day, San Francisco. DURBIN, R. K., and STOLLAR, V. (1984). A mutant of Sindbis virus with a host-dependent defect in mat-
OF SINDBIS
VIRUS
19
uration associated with hyperglycosylation of E2. Virobgy 135,331~344. GAROFF, H., FRISCHAUF, A.-M., SIMONS, K., LEHRACH, H., and DELIUS, H. (1980). Nucleotide sequence of cDNA coding for Semliki Forest virus membrane glycoproteins. Nature (Lund& 288,235-241. HAHN, C. S., STRAUSS,E. G., and STRAUSS,J. H. (1985). Sequence analysis of three Sindbis virus mutants temperature-sensitive in the capsid protein autoprotease. Proc. Natl. Ad Sci USA 82,4648-4652. HUTH, A., RAPOPORT,T. A., and KAARIAINEN, L. (1984). Envelope proteins of Semliki Forest virus synthesized in Xenopus oocytes are transported to the cell surface. EMBO J. 3,767-771. KYTE, J., and DOOLITTLE, R. F. (1982). A simple method for displaying the hydropathic character of a protein. J. Mot Biol 157,105-132. MANIATIS, T., FRITSCH, E. F., and SAMBROOK,J. (1982). “Molecular Cloning, a Laboratory Manual.” Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. MAXAM, A. M., and GILBERT, W. (1980). Sequencing end-labeled DNA with base-specific chemical cleavages. In “Methods in Enzymology” (L. Grossman and K. Moldave, eds.), Vol. 65, pp. 499-560. Academic Press, Orlando, Fla. MAYNE, J. T., BELL, J. R., STRAUSS,E. G., and STRAUSS, J. H. (1985). Pattern of gylcosylation of Sindbis virus envelope proteins synthesized in hamster and chicken cells. Virology 142, 121-133. MAYNE, J. T., RICE, C. M., STRAUSS, E. G., HUNKAPILLER, M. W., and STRAUSS, J. H. (1984). Biochemical studies of the maturation of the small Sindbis virus glycoprotein E3. Virology 134,338-357. OKAYAMA, H., and BERG, P. (1982). High-efficiency cloning of full-length cDNA. Mol. CeU Biol. 2,161170. Ou, J.-H., STRAUSS, E. G., and STRAUSS, J. H. (1981). Comparative studies of the 3’ terminal sequences of several alphavirus RNAs. Virology 109,281-289. Ou, J.-H., TRENT, D. W., and STRAUSS, J. H. (1982). The 3’ non-coding regions of alphavirus RNAs contain repeating sequences. J. Mol. Biol 156,719-730. RICE, C. M., FRANKE, C. A., STRAUSS,J. H., and HRUBY, D. E. (1985). Expression of Sindbis virus structural proteins via recombinant vaccinia virus: Synthesis, processing, and incorporation into mature Sindbis virions. J. Vird 56,227-239. RICE, C. M., and STRAUSS, J. H. (1981). Nucleotide sequence of the 26s mRNA of Sindbis virus and deduced sequence of the encoded virus structural proteins. Proc. Natl Ad Sci USA 78,2062-2066. SANGER, F., and COULSON,A. R. (1978). The use of thin acrylamide gels for DNA sequencing. FEBS Lett 87,107-110. SANGER, F., NICKLEN, S., and COULSON,R. (1977). DNA sequencing with chain-terminating inhibitors. Proc Natl Amd. Sci. USA ?4,5463-5467.
20
LINDQVIST
STRAUSS, E. G., LENCHES, E. M., and STRAUSS, J. H. (1976). Mutants of Sindbis virus. I. Isolation and partial characterization of 89 new temperaturesensitive mutants. virology 74,154-168. STRAUSS, E. G., RICE, C. M., and STRAUSS,J. H. (1984). Complete nucleotide sequence of the genomic RNA of Sindbis virus. Vidogy 133, 92-110. STRAUSS, E. G., and STRAUSS, J. H. (1980). Mutants of alphaviruses: Genetics and physiology. In “The Togaviruses: Biology, Structure, Replication” (R. W. Schlesinger, ed.), Chap. 14, pp. 393-426. Academic Press, New York. STRAUSS, E. G., and STRAUSS, J. H. (1986). Structure
ET AL. and replication of the alphavirus genome. In “The Togaviridae and Flaviviridae” (S. Schlesinger and M. J. Schlesinger, eds.), Chap. 3, pp. 35-90. Plenum, New York. WIRTH, D. F., KATZ, F., SMALL, B., and LODISH, H. (19’77). How a single Sindbis virus mRNA directs the synthesis of one soluble protein and two integral membrane glycoproteins. Cell 10,253-263. ZIMMERN, D., and KAESBERG, P. (1978). 3’-Terminal nucleotide sequence of encephalomyocarditis virus RNA determined by reverse transcriptase and chain-terminating inhibitors. Proc. Nati Auw! Sk. USA 75,4257-4261.