- Email: [email protected]
Multiple mutations involved in the phenotype of a temperature-sensitive small-plaque mutant of poliovirus
VIROLOGY
157, 75-82 (1987)
Multiple
Mutations
Involved in the Phenotype of a Temperature-Sensitive Small-Plaque Mutant of Poliovirus
C. BELLOCQ,’ Unit4 de Virologie Mokkulaire,
K. M. KEAN,’ 0. FICHOT, M. GIRARD, AND H. AGUT UA CNRS 545, lnstitut Pasteur, 25 rue du Dr. Roux, 75724 Paris, Cedex 15, France
Received June 79, 1986; accepted
October 2 1. 1986
A temperature-sensitive small-plaque mutant of poliovirus type 1, ts247, has been analyzed previously. Several mutations were detected in the P3 region of the genome by analysis of proteins and by Tl oligonucleotide mapping of viral RNA. We have now studied spontaneous reversion of ts247 to the wild-type phenotype. This was found to be a two-step event, reversion to a ts+ phenotype (revertant R247-51) being distinct from acquisition of normal plaque size (revertant R247-12). The mutation responsible for the ts phenotype of ts247, implicated also in virus aggregation and heat lability, could not be detected by biochemical studies. Analysis of homotypic recombinants obtained by crossing ts247 with a guanidine-resistant derivative of a temperature-sensitive replicase mutant mapped this mutation to the Pl region or to the 5’ end of the P2 region of the genome. The small-plaque phenotype of ts247 and R247-51 was correlated with an abnormality in polypeptide 3C (protease); direct sequencing of viral RNA revealed a U to C change at nucleotide 5658, which altered an isoleucine to threonine in the protease of ts247 and R247-51 but not of R247-12. Two other mutations were present in the region of the genome coding for polypeptide 3D of ts247 and of both classes of revertants. They thus seemed to play no role in the phenotype of ts247. One mutation, an A to G change at nucleotide 7135, was silent at the protein level, whereas the other, an A to G change at nucleotide 6264, o 1987 Academic PUSS, hc. determined a major amino acid change from glutamate to glycine in the viral replicase.
peptides can be found in infected cells. As an example, polypeptide 3CD, the precursor to 3C and 3D, can follow an alternative cleavage pathway to give polypeptides 3C’ and 3D’ (see Fig. 1). The functions and mechanism of action of many viral polypeptides remain unknown. Genetic studies of poliovirus have previously proved of limited use (for a review, see Cooper, 1977) because until recently conditional mutants were not well defined and were difficult to map. However, biochemical methods now available allow new approaches to genetic analysis, as exemplified by the recent studies of guanidine resistance (Pincus et a/., 1986) and of attenuation of neurovirulence (Omata et a/., 1985; Almond et a/., 1985). We have previously described the isolation and characterization of several temperature-sensitive mutants of poliovirus type 1 (Agut et al., 1981; Bellocq et al., 1984). One of these mutants, ts247, has a complex phenotype: it shows a 3-log inhibition of plaquing efficiency and no significant RNA synthesis at 39”, produces small plaques at 37” compared to wild-type virus, and its virions are very heat labile and readily aggregate, as demonstrated by electron microscopy. Two mutations were detected in the ts247 genome by Tl oligonucleotide mapping, one in the region coding for the protease 3C and the second in the region coding for the replicase 30 (protein nomenclature is according to Ruecker-t and Wimmer, 1984). Polypeptides 3C, 30, 3CD, and 3C’ showed abnormal migration on SDS-
INTRODUCTION Poliovirus has been intensively studied since its identification as the agent responsible for poliomyelitis (Landsteiner and Popper, 1908). The three-dimensional structure of the virion has recently been determined (Hogle eta/., 1985). The viral genome is a monocistronic mRNA, approximately 7500 nucleotides long, of known sequence (Kitamura et al., 1981; Racaniello and Baltimore, 1981). All identified virus-encoded polypeptides are generated by multiple proteolytic cleavages of a single translation product, and they have been mapped onto the viral RNA (Kitamura et a/., 1981; Pallansch et a/., 1984). From 5’to 3’the coding region of the genome can be divided into three parts, Pl, P2, and P3, corresponding to the major initial cleavage products of the polyprotein. Pl codes for the capsid polypeptides VPl, VP2, VP3, and VP4, while P2 and P3 code for nonstructural proteins and for 3B, the small protein covalently linked to the 5’ end of the genomic RNA. It is generally accepted that P3 codes for four final products, including one of the viral proteases (3C) and the viral replicase (3D) whose genes are located adjacent to one another. Precursors and alternative cleavage products are abundant; up to 27 viral-encoded poly’ Present address: DBpartement de Microbiologic. GenBve. 1205 Geneva, Switzerland. ’ To whom reprint requests should be addressed.
Universit6
de
75
0042-6822187
$3.00
Copynght 0 1987 by Academic Press. Inc. All nghts of reproduction in any form reserved
76
BELLOCQ
polyacrylamide gel electrophoresis and/or isoelectrofocusing (Bellocq et al., 1984). In order to correlate the functional defects of ts247 with the mutations detected biochemically, we have selected and analyzed revertants of ts247 and homotypic recombinants obtained by crossing ts247 with a temperature-sensitive replicase mutant (Agut et a/., 1984). As is shown, neither the mutation in polypeptide 3C nor that in polypeptide 3D is responsible for the temperature-sensitive phenotype of ts247. An unidentified mutation in the 5’half of the genome is implicated in this phenotypic trait, while the mutations already detected are responsible for the small-plaque phenotype.
Recombinants between ts247 and ts035Gr, a guanidine-resistant derivative of a temperature-sensitive replicase mutant, were obtained using the infectious center method (McCahon and Slade, 1981) as described previously (Agut et a/., 1984).
Phenotypic analysis of viruses Virus yields after infection of HeLa cells and heat stability of virus stocks were determined as previously described (Agut eta/., 1981). Briefly, to determine heat inactivation, virus stocks diluted in PBS to about lo* PFU/ml were incubated at 45’ for 10 min, immediately diluted in PBS at 4’, and titrated by plaque assay at 37”. Titers were compared with those of control samples kept at 4’.
MATERIALS AND METHODS Virus strains and cells The temperature-sensitive poliovirus mutant ts247 was derived from the type 1 Mahoney strain by mutagenesis with nitrous acid as described (Agut et al., 1981). Unless otherwise indicated, viruses were grown and plaque assayed at 37” on the Hopkins strain of HeLa cells. Biochemical analyses were carried out using these cells or the ATCC strain of HeLa cells.
Analysis of viral polypeptides HeLa cells were infected at a multiplicity of 100 PFU/ cell and incubated at 37” for 4 hr. The cells were next incubated in methionine-free medium for 30 min and then labeled for 1 hr with 30 &i/ml of [35S]methionine (3000 Ci/mmol, Amersham) before being lysed. Preparation of cytoplasmic extracts was as previously described (Bellocq et al., 1984). Samples of the extracts (1 O6 cpm) were denatured in 2% SDS, 5% 2-mercaptoethanol at 100” for 2 min, and analyzed by electrophoresis on a lo-18% SDSpolyacrylamide gel (Hanecak et al., 1982). Cytoplasmic extracts were also subjected to two-dimensional polyacrylamide gel electrophoresis, using isoelectric focusing (IEF) from pH 3.5 to pH 10 in the first dimension and SDS-PAGE in the second (O’Farell, 1975; Lowe et a/., 1981; Wiegers and Dernick, 1981). Prior to IEF, each sample was incubated for 15 min at 37” in the presence of 3.5% (v/v) 2-mercaptoethanol, 9 M urea, and 0.1 mg/ml RNase A as previously described (Bellocq et a/., 1984).
Isolation of revertants and recombinants Spontaneous first-generation revertants of ts247 were isolated from randomly picked plaques obtained after infection of HeLa cells with ts247 at 39” (nonpermissive temperature). Each isolate was plaque purified twice at 39” before virus stocks were made at 37”. These revertants gave small plaques at 39”. Variants with large-plaque phenotype could be isolated after infection of HeLa cells with first-generation revertants at 39’, and were assumed to be second-generation revertants of ts247. These second-generation revertants were also plaque purified twice at 39” before virus stocks were prepared at 37”.
5110
5372 lsq3fl
59.87
.
. *
.
3AB 3A 13Jl
I
ET AL.
6431
7370
. 3CD
3c
* 3C’
3D .
30’
FIG. 1. Schematic diagram of the major products of proteolytic processing in the P3 region of the poliovirus genome. Polypeptides derive from the primary cleavage product, P3 (BABCD) by cleavage at glutamine-glycine (A) or tyrosine-glycine (0) sites. Nucleotide positions on the genome are indicated on the broad upper line (after Pallansch er al., 1984).
A ts PLURIMUTANT
77
OF POLIOVIRUS
TABLE 1 PHENOTYPIC PROPERTIES 0~ts247
AND ITS REVERTANTS
Virus yieldc
Plaque sizeb Virus
Titerat 39” Titerat 37”
wt ts247 R247-51
R247-12
Heat lability of virionsB
Aggregation virions
of 39”
37”
>4
3.2X 10'
1 5x1o-4 1
1.5 5 1.5
+ -
3 1.5 1.5
1
1.5
-
3
>4
B Log drop in titer at 37”, after incubation at 45” for 10 min. b Diameters of plaques (mm) 48 hr after infection of HeLa cells. ’ HeLa cells were infected for 8 hr with 20 PFU per cell and broken open by freeze-thawing, at 37”.
RNA fingerprinting
and sequencing
Viral RNA was purified as described by Lee et al. (1979). The ts247 virions were purified by isopycnic centrifugation through cesium chloride, whereas revertants and recombinants were purified through sucrose gradients in the presence of 0.5% SDS. RNase Tl oligonucleotide mapping of viral RNA (De Wachter and Fiers, 1972) was carried out as previously described (Bellocq et a/., 1984). Briefly, RNA was digested with RNase Tl (PL Biochemicals) and simultaneously 5’ labeled with [3’P]ATP by T4 polynucleotide kinase (PL Biochemicals). Tl -resistant oligonucleotides were subsequently separated by two-dimensional polyacrylamide gel electrophoresis. Viral RNA was sequenced by the dideoxynucleotide chain termination method (Qu et a/., 1983), using synthetic oligonucleotides as primers for reverse transcription (Dreano et a/., 1985). RESULTS Phenotypic
characterization
of ts247 revertants
Spontaneous first-generation revertants of ts247, characterized by their ability to form plaques at 39’, were isolated with a frequency of approximately 5 X 1Oe4. All 15 of the revertants analyzed (typified by R247-51) exhibited the same phenotype (Table 1). Their titer was the same at 39 as at 37”. In addition, their virions were as heat stable as those of wild-type virus, and could readily be purified on SDS-sucrose gradients. The ts247 cannot be purified by this method, like the Sabin type 1 strain, nor on sarkosyl-sucrose gradients. SDS or sarkosyl inactivation experiments have shown that this characteristic was not linked with inactivation by detergents, but was probably due to aggregation of virions which strongly increased their sedimentation rate. Aggregation of ts247 particles was readily observed by electron microscopy, after purification by isopycnic centrifugation through cesium
39”
37”
6.3x
1.3x
lo8
10' 1.3X106
5.0x 3.8x
lo8 10’
1.6x 10’ 6.1 X lOa
and the viral titer (PFWml) was determined
chloride, whereas cesium chloride-purified R247-51 showed completely dissociated particles (data not shown). The first-generation revertants of ts247 were different from wild-type virus as shown by the fact that they produced smaller plaques at 39 than at 37“, and they produced smaller plaques than wild-type virus at both temperatures. This could be related to the virus yield from HeLa cells infected for 8 hr. The yield of R247-5 1 was about 0.5 log lower at 39 than at 37’, whereas that of wild-type virus was increased by about 0.3 log, and that of ts247 reduced by 2 logs (Table 1). Large-plaque variants of R247-5 1 could be isolated at 39” with a frequency of about 5 X 1Op3. These viruses were assumed to be second-generation revertants of ts247. The three viruses tested (typified by R247-12) behaved phenotypically as wild-type virus (Table 1). Thus it seemed that the complex phenotype of ts247 could be segregated in two: temperature sensitivity, heat lability, and aggregation of the virions reverted concomitantly and separately from the small-plaque phenotype. The reversion frequencies observed for the two sets of properties were compatible with the implication of two point mutations. Analysis
of viral polypeptides
Two alterations of ts247 proteins have already been detected, one affecting the protease (3C) and the other the replicase (3D) (Bellocq et a/., 1984). The polypeptides of the first- and second-generation revertants were analyzed for the presence of these modifications by SDS-polyacrylamide gel electrophoresis (Fig. 2) and by two-dimensional gel electrophoresis (Fig. 3). Surprisingly, the electrophoretic pattern of R247-51 polypeptides was identical to that of ts247 concerning the polypeptides encoded by the P3 region of the genome. Polypeptide 3C migrated slowly on SDS-PAGE, whereas 30 migrated rapidly, and both were present
BELLOCQ
78 B
1234
ET AL.
The small-plaque phenotype of ts247 and first-generation revertants (R247-51) could be correlated with the presence of an altered 3C. -3CD -3D -3C’
FIG. 2. SDS-polyacrylamide gel analysis of viral polypeptides. Cytoplasmic extracts (10’ cpm) from infected HeLa cells were analyzed on a 1O-l 8% linear gradient SDS-polyacrylamide gel. The positions of wild-type viral polypeptides and molecular weight markers run in parallel are indicated. (A) Cells were infected with R247-51 (lane l), ts247 (lane 2) or wild-type virus (lane 3). Purified virions (lane 4) and mock-infected cells (lane 5) were included as controls. (B) Cells were infected with wild-type virus (lane 2) ts247 (lane 3) or R247-12 (lane 4). A mock-infected cell extract (lane 1) was run in parallel.
in reduced amounts when compared to the wild-type virus pattern (Fig. 2A, compare lanes 1 and 2 with lane 3). Also, polypeptides 3D, 3CD, and 3C’ from ts247 and R247-51 showed identical pl values (Fig. 3B), different from those of the corresponding wild-type polypeptides Fig. 3A). It should be noted that ts247 repeatedly showed an extremely weak VP2 band (Figs. 2A, lane 2, and 2B, lane 3). In contrast, the second-generation revertants gave a unique protein profile. On SDS-PAGE (Fig. 2B, lane 4) polypeptide 3C migrated identically to the wild-type equivalent, while polypeptide 3D migrated like the ts247 homologous polypeptide, but its amount was significantly increased compared to ts247 (compare lanes 3 and 4 of Fig. 2B). Polypeptides 3CD and 3C’ of ts247 migrated like those of wild-type virus, whereas the R247-12 counterparts migrated more rapidly (compare Fig. 2B, lane 4, with lanes 2 and 3). Figure 3 allowed the p/values of R247-12 polypeptides to be compared to those of ts247 (Fig. 3C) and wild-type virus (Fig. 3D). The p/values of 3CD, 3D, and 3C’were similar to those of the ts247 equivalents, and clearly distinct from those of the corresponding wild-type polypeptides. The fast migration rate of polypeptides 3CD and 3C’ of R24712 was clearly visible, particularly in Fig. 30 because of the difference in pl values of these polypeptides between R247-12 and wild-type virus. It thus appeared that the temperature sensitivity, heat lability, and aggregation of ts247 were not related to the alterations observed in polypeptides 3C and 3D.
Tl -resistant oligonucleotide
mapping
In order to confirm these results, the RNA from ts247 and from revertants R247-51 and R247-12 was subjected to RNase Tl mapping (Fig. 4). The first-generation revertants (R247-51) showed the same abnormalities as ts247: a charge change in Tl oligonucleotide No. 10 (numbered according to Lee er al., 1979) which maps at nucleotide 5644, in the region of the genome coding for polypeptide 3C; and the disappearance of oligonucleotide No. 8 located at nucleotide 7125, in the sequence coding for polypeptide 3D (compare Figs. 4A, 48, and 4C). Oligonucleotide 8 remained absent in the case of the second-generation revertants, whereas oligonucleotide 10 had regained its normal charge (compare Figs. 4A and 4D). The modification of Tl oligonucleotide No. 10 correlated with the alteration of polypeptide 3C, suggesting the likely location of the mutation responsible for the small-plaque phenotype of ts247 and R247-5 1. On the other hand, the mutation indicated by the alteration of oligonucleotide No. 8, which is located at nucleotide 7125, could not be implicated in abnormal pl values
FIG. 3. Two-dimensional PAGE analysis of viral polypeptides. Mixtures of infected cell extracts: ts247 + wild-type (A); ts247 + R24751 (B); ts247 + R247-12 (C); wild-type + R247-12 (D) were subjected to IEF followed by SDS-PAGE (Bellocq ef a/., 1984). Second-dimension electrophoresis was on a 1O-l 8% (A,B) or a 15% (CD) SDSpolyacrylamide gel.
A ts PLURIMUTANT
FIG. 4. RNase Tl fingerprints of viral RNA. (A) Wild-type; (B) ts247; (C) R247-51; (D) R247-12. Oligonucleotide No. 10 is indicated by a black arrow, and oligonucleotide No. 8 (or its absence) by a hollow arrow.
and migration on SDS-PAGE of polypeptides 3D, 3CD, and 3C’, since 3C’ is not coded by this region of the genome. The concomitant alteration of 3D, 3CD, and 3C’ suggested a mutation between nucleotides 5987 and 6431, the region of the genome coding simultaneously for these three polypeptides (Fig. 1). Nucleotide
sequence
determination
To locate and identify the mutations in the P3 region of the genome, we sequenced the RNA of ts247 in the regions suggested above as carrying mutations. Three mutations were identified (Table 2) one at nucleotide 5658 in the region coding for the protease 3C, and two (one of which is silent at the protein level) in the region coding for the replicase 3D, at nucleotides 6264 and 7135. R247-51 carried these same three mutations, whereas R247-12 had reverted to the wild-type sequence at nucleotide 5658, but had kept the two mutations in the part of the genome coding for polypeptide 3D. These three mutations are sufficient to explain the abnormalities observed in the proteins and RNA of ts247 and its revertants. Nucleotide 5658 falls within Tl oligonucleotide No. 10, and a change from U to C at this position will alter the charge of the oligonucleotide. The resultant change
79
OF POLIOVIRUS
from isoleucine to threonine in polypeptide 3C does not alter the overall charge of the protein, but modifies its hydrophilicity. This could affect the migration of the polypeptide on SDS-PAGE, but not its isoelectric point, as seen. Nucleotide 6264 does not lie within any of the large Tl oligonucleotides readily identified by fingerprinting (Kitamura et al., 1981), which would explain the failure of this method to detect the mutation. The glutamate to glycine change at this point simultaneously changes the charge and hydrophilicity of 3CD, 3D, and 3C’. The presence of this mutation is consistent with the charge changes observed by two-dimensional electrophoresis. SDS-PAGE migrations can be explained if the mutation at nucleotide 6264 results in faster migration of 3D than normal, whereas the mutation at nucleotide 5658 results in slower migration of 3C than normal. If the two mutations compensate each other, polypeptides which carry both alterations would migrate normally (3CD and 3C’of ts247 and R247-51); in contrast, those which only carry the modification resulting from the mutation at nucleotide 6264 (e.g., 3CD and 3c’of R24712) would migrate faster than normal, as observed. Nucleotide 7135 falls within Tl oligonucleotide No. 8. An A to G change at this position introduces an additional Tl RNase cut and hence the disappearance of this oligonucleotide. This mutation does not modify the amino acid sequence, and therefore has no effect on protein migration.
TABLE 2 LOCATIONOF MUTATIONSIN THE P3 REGIONOFTHE GENOMESOF ts247 AND ITSTwo CLASSESOF REVERTANTS Nucleotide number: Polypeptide: Amino acid number in the polypeptide: Sequence ts247
5658 Protease (3C)
6264 Replicase (3D)
7135 Replicase (3D)
74
93
383
ACA Thr
AAG LYS AAG LYS AAG LYS AAA LYS
R247-51
ACA Thr
R247-12
AUA Ile
Wild-type
AUA Ile
GGA GIY GGA GIY GGA GIY GAA Glu
0
f4.3
0
+1.4
-3.0
0
Charge change of amino acid Hydrophilicity change of amino acide a Calculated
after Hopp and Woods (1981)
80
BELLOCQ
Analysis of recombinants None of the mutations detected by Tl fingerprinting or polypeptide analysis could be responsible for the ts phenotype of ts247. The fact that the ts character seemed to be linked to thermolability and virion aggregation could implicate a mutation in the Pl region of the genome, which codes for the capsid proteins. To verify this hypothesis, the phenotypes of ts+ recombinants obtained from the cross ts035Gr X ts247 were analyzed. The ts035Gr is a guanidine-resistant derivative of a temperature-sensitive replicase (3D) mutant of poliovirus. The advantage of this approach is that the ts247 ts mutation can be located with reference to an independent marker, resistance to guanidine. There is now convincing evidence that guanidine resistance maps to the middle of the poliovirus genome, within the region coding for polypeptide 2C (Pincus et a/., 1986). Hence, if the ts mutation of ts247 is located 3’ of the guanidine-resistance marker, like the ts mutation of ts035, recombinants generated by a single crossing over between the two ts mutations would be either all guanidine resistant or all guanidine sensitive, depending on the position of the ts mutations relative to each other. If, on the other hand, the ts mutation of ts247 is located 5’ of the guanidine resistance, both guanidine-sensitive and guanidine-resistant recombinants would be generated by single crossing-over events. Twelve ts+ recombinants were picked and characterized. They could be classified into three groups (Table 3) typified by RCOOl (4/l 2) which is guanidine sensitive, and RC003 (2/12) and RC008 (6112) which are
TABLE 3 PHENOTYPICPROPERTIESOF RECOMNNANTSBETWEEN ts035Gr AND ts247
Virus Parents ts035Gr ts247 Recombinants RCOOl RC003 RC008
Titerat 39” Titerat 37”
Heat lability of virionsB
Resistance to guanidineb
Plaque size at 37”’
2x1o-3 5x 1o-4
+
+ -
3 1.5
-
1.5 1.5 3
1 1 1
+ +
a Virus was considered as heat labile if it exhibited a drop in titer of more than 2 logs after heating for 10 min at 45”. * Virus was considered as resistant if it exhibited the same titer in the absence and in the presence of 1 mM guanidine-HCI. c Measured as diameter of plaques in mm 48 hr after infection of HeLa cells.
ET AL.
guanidine resistant. RCOOl and RC003 have the same phenotypic characteristics as R247-51, whereas RC008 resembles R247-12. Preliminary analysis showed that, as in the case of revertants, small-plaque phenotype correlated with the presence of the mutation at nucleotide 5658 (Agut et a/., in preparation). Thus the recombinants obtained carried the 5’ end of the ts035 genome, linked to the 3’ end of the ts247 genome. A comparison of RCOOl and RC003 showed that exclusion of the two ts mutations was independent of acquisition of the guanidine-resistance mutation from ts035. Assuming that most recombinants were generated by a single crossing over, these results demonstrate that the ts mutation of ts247 is separated from that of ts035 by the mutation responsible for guanidine resistance. Consequently, we concluded that the ts247 mutation is located in Pl or in the 5’ end of P2.
DISCUSSION The main advantage of chemical mutagenesis of viruses over the presently employed techniques of sitedirected mutagenesis of viral genomes is the relative ease with which conditional lethal mutant viruses can be obtained, for example, by screening for a readily detectable phenotype such as temperature sensitivity. However, several mutations can accumulate in a single genome, and it may be extremely laborious to map the mutation(s) implicated in a given defect. A temperaturesensitive mutant, ts247, was obtained after such chemical mutagenesis of poliovirus type 1, and preliminary analyses suggested that multiple mutations were responsible for its complex phenotype. Partial sequencing of the ts247 genome has allowed us to identify three mutations in the P3 region. One of these mutations located at nucleotide 5658 results in an isoleucine to threonine change in the viral protease, polypeptide 3C. A second mutation, located at nucleotide 6264, changes a glutamate to glycine in the replicase, polypeptide 3D. The third mutation, also located in the region of the genome coding for polypeptide 3D (nucleotide 7135), is silent at the protein level. The presence of these three mutations could readily be ascertained from Tl fingerprints of the viral RNA and/or from viral polypeptide migration patterns. Studies of spontaneous ts+ revertants of ts247 have allowed us to evaluate the correlation between these three mutations and viral phenotype. A comparison between the first- and second-generation revertants of ts247 showed that the presence of an altered protease, due to the mutation at nucleotide 5658, correlated with a small-plaque phenotype. In ts247 and its first-generation revertants (R247-5 l), the altered migration and decreased amount of polypeptide 3C were associated
A ts PLURIMUTANT
with a decreased amount and a converse alteration of migration of 3D. We first thought that the mutation at nucleotide 5658 resulted in cleavage of 3CD at an abnormal site to give a larger 3C and smaller 30 than normal. However, altered SDS-PAGE migration of 3D was dissociated from that of 3C in the case of the second-generation revertants (R247-12). The decreased amount of 30, on the other hand, was associated with the alteration of 3C. It thus appears that the isoleucine to threonine change at residue 74 of the viral protease decreases the amount of functional replicase, probably through impaired protease activity. That this is a major factor responsible for the small plaque size of ts247 and R247-51 has been confirmed by molecular cloning of the protease mutation (Kean er al., in preparation). At first view, the two mutations detected in the replicase gene have no effect on virus growth. Secondgeneration revertants (R247-12) of ts247, which carried both these mutations, behaved phenotypically as wildtype virus. It seems surprising that the viral replicase can apparently tolerate a relatively major change from glutamate to glycine, resulting from an A to G change at nucleotide 6264. However, it has been observed (Dreano et a/., 1985) that different stocks of poliovirus type 1 (Mahoney strain) differ at the preceding codon, having either a threonine or an isoleucine at amino acid 92 in the replicase. The amino acids at these two positions are not conserved in different picornaviruses (Argos et al., 1984). It thus appears that conservation of the amino acid sequence in this region of polypeptide 3D is not important for its activity. It would be interesting to investigate to what extent the virus can support changes in this region of its genome, using site-directed mutagenesis of cloned viral cDNA. None of the mutations identified in the P3 region of the genome could be held responsible for the temperature-sensitive character of ts247 since the first-generation revertants were ts+ but had retained all three mutations. Such reversion of ts247 to a ts+ phenotype (R247-51) was accompanied by the loss of heat lability and aggregation of the virions. The high reversion frequency observed implied that temperature sensitivity, thermolability, and aggregation of ts247 arose concomitantly from the presence of a single point mutation. The nature of these phenotypic characteristics suggested that this mutation lay in the region of the genome coding for the capsid proteins. Analysis of homotypic recombinants formed by genetic recombination between ts247 and a guanidine-resistant ts replicase mutant (ts035Gr) mapped the ts mutation in ts247 5’ of the region of the genome coding for polypeptide 2C. To date, we have been unable to detect a biochemical marker that could help to map this mutation more precisely.
81
OF POLIOVIRUS
The observed phenotype of ts247 is due to a cumulative effect of multiple mutations. This experimental model is imperfectly characterized as yet, but is of great interest in view of the search for safer attenuated poliovirus vaccines based on the idea of accumulating mutations to increase attenuation and prevent reversion to neurovirulence. It is of note that mutations involving a relatively major amino acid change, such as the glutamate to glycine change at residue 93 in the viral replicase, can apparently be without functional effect. Conversely, it has already been shown in the case of picornaviruses that mutations with no effect on protein sequence, notably mutations in the noncoding regions of the genome, can have profound phenotypic effects on the virus (Almond et a/., 1985; Sarnow et al., 1986; Racaniello, personal communication). As will be shown elsewhere (manuscript in preparation) the mutation at nucleotide 7135 of ts247, albeit silent at the protein level, is implicated in a temperature-dependent sensitivity of viral RNA replication to actinomycin D.
ACKNOWLEDGMENTS This work was supported in part by INSERM Grant CA 40501, by NATO Grant RG 175/80, and by MRT Programme Biotechnologies.
REFERENCES AGUT, H., BELLOCQ, C., VAN DER WERF, S., and GIRARD, M. (1984). Recombination and rescue between temperature-sensitive mutants of poliovirus type 1. Virology 139,393-402. AGUT. H., MATSUKURA.T., BELLOCQ, C., DREANO, M., NICOLAS, J. C., and GIRARD, M. (1981). Isolation and preliminary characterization of temperature-sensitive mutants of poliovirus type 1. Ann. Viral. (Inst. Pasteur) 132E, 445-460. ALMOND, I. W., WESTROP, G. D., CANN, A. I., STANWAY, G., EVANS, D. M. A., MINOR, P. D., and SCHILD, G. C. (1985). Attenuation and reversion to neurovirulence of the Sabin polio type 3 vaccine. ln “Vaccines 85,” pp. 271-277, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. ARGO& P., KAMER, G., NICKLIN, M. J. H., and WIMMER, E. (1984). Similarity in gene organization and homology between proteins of the animal picornaviruses and a plant comovirus suggest common ancestry of these virus families. Nucleic Acids Res. 12, 72517276. BELLOCQ, C., AGUT, H., VAN DER WERF, S., and GIR~RD, M. (1984). Biochemical characterization of poliovirus type 1 temperaturesensitive mutants. Virology 139, 403-407. COOPER,P. D. (1977). Genetics of picornaviruses. In “Comprehensive Virology” (H. FraenkeCConrat and R. R. Wagner, Ed%), Vol. 9, pp. 133-207. Plenum, New York. DE WACHTER, R., and FIERS,W. (1972). Preparative two-dimensional polyacrylamide gel electrophoresis of 3ZP-labelled RNA. Anal. Biochem. 49, 184-l 97. DREANO, M., BELLOCCI,C., FICHOT, 0.. VAN DERWERF, S., and GIRARD, M. (1985). Genetic variations in the Mahoney strain of poliovirus type 1. Ann. Vifol. (Inst. Pasteur) 136E, 10 l-l 14. HANECAK,R., SEMLER,B. L., ANDERSON,C. W., and WIMMER, E. (1982). Proteolytic processing of poliovirus polypeptide: antibodies to polypeptide P3-7c inhibits cleavage at glutamine-glycine amino acid pairs. Proc. Natl. Acad. Sci. USA 79, 3973-3977.
82
BELLOCQ
HOGLE,J. M., CHOW, M., and FILMAND, D. J. (1985). Three-dimensional structure of poliovirus at 2.9 A resolution. Science 229, 13581365. HOPP, T. H., and WOODS, K. R. (1981). Prediction of protein antigenic determinants from amino acid sequences. Proc. Nat/. Acad. Sci. USA 78,3824-3828. KITAMURA,N., SEMLER, B. L., ROTHBERG,P. G., LARSEN,G. R., ADLER, C. J., DORNER, A. J., EMINI, E. A., HANECAK, R., LEE, J. J., VAN DER WERF, S., ANDERSON,C. W., and WIMMER, E. (1981). Primary structure, gene organization and polypeptide expression of poliovirus RNA. Nature (London) 291, 547-553. LANDSTEINER,K., and POPPER, E. (1908). Mikroskopische praparate von einem menschlichen und zwei Aftenrijckenmarken. Wein. K/in. Wochenschr. 21, 1830. LEE, Y. F., KITAMURA, N., NOMOTO, A., and WIMMER, E. (1979). Sequence studies of poliovirus RNA. IV. Nucleotide sequence complexities of poliovirus type 1, type 2 and two type 1 defective interfering particles RNAs, and fingerprint of the poliovirus type 3 genome. J. Gen. Viral. 44, 311-322. LOWE, P. A., KING, A. M. Q., MCCAHON, D., BROWN, F., and NEWMAN, J. W. I. (1981). Temperature-sensitive RNA polymerase mutants of a picornavirus. Proc. Nat/ Acad. Sci. USA 78, 4448-4452. MCCAHON, D., and SLADE, W. R. (1981). A sensitive method for the detection and isolation of recombinants of foot-and-mouth disease virus. J. Gen. Viral. 53, 333-342. O’FARRELL, P. H. (1975). High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021. OMATA, T., KOHARA, M., ASE, S., ITOH, H., KOMATSU, T., ARITA, M., SEMLER, B. L., WIMMER, E., KUGE, S., KAMEDA, A., and NOMOTO, A.
ET AL. (1985). Construction of recombinant viruses between Mahoney and Sabin strains of type 1 Poliovirus and their biological characteristics. In “Vaccines 85,” pp. 279-283. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. PALLANSCH, M. A., KEW, 0. M., SEMLER, B. L., OMILIANOWSKI,D. R., ANDERSON,C. W., WIMMER, E., and RUECKERT,R. R. (1984). Protein processing map of poliovirus. J. Viral. 49, 873-880. PINCUS, S. E., DIAMOND, D. C., EMINI, E. A., and WIMMER, E. (1986). Guanidine-selected mutants of poliovirus: Mapping of point mutations to polypeptide 2C. J. Viral. 57, 638-646. Qu, L. H., MICHOT, B., and BACHELLERIE, J. P. (1983). Improved methods for structure probing in large RNAs: A rapid “heterologous” sequencing approach is coupled to the direct mapping of nuclease accessible site. Application to the 5’terminal domain of eukaryotic 28s rRNA. Nucleic Acids Res. 11, 5903-5920. RACANIELLO,V. R., and BALTIMORE, D. (1981). Molecular cloning of poliovirus cDNA and determination of the complete nucleotide sequence of the viral genome. Proc. Nat/. Acad. Sci. USA 78,48874891. RUECKERT,R. R., and WIMMER, E. (1984). Systematic nomenclature of picornavirus proteins. J. Viral. 50, 957-959. SARNOW,P., BERNSTEIN,H. D., and BALTIMORE,D. (1986). A poliovirus temperature-sensitive RNA synthesis mutant located in a noncoding region of the genome. Proc. Natl. Acad. Sci. USA 83, 571575. WIEGERS, K. J., and DERNICK,R. (1981). Poliovirus-specific polypeptides in infected HeLa cells analysed by isoelectric focusing and 2Danalysis. J. Gen. Viral. 52, 6 l-69.
157, 75-82 (1987)
Multiple
Mutations
Involved in the Phenotype of a Temperature-Sensitive Small-Plaque Mutant of Poliovirus
C. BELLOCQ,’ Unit4 de Virologie Mokkulaire,
K. M. KEAN,’ 0. FICHOT, M. GIRARD, AND H. AGUT UA CNRS 545, lnstitut Pasteur, 25 rue du Dr. Roux, 75724 Paris, Cedex 15, France
Received June 79, 1986; accepted
October 2 1. 1986
A temperature-sensitive small-plaque mutant of poliovirus type 1, ts247, has been analyzed previously. Several mutations were detected in the P3 region of the genome by analysis of proteins and by Tl oligonucleotide mapping of viral RNA. We have now studied spontaneous reversion of ts247 to the wild-type phenotype. This was found to be a two-step event, reversion to a ts+ phenotype (revertant R247-51) being distinct from acquisition of normal plaque size (revertant R247-12). The mutation responsible for the ts phenotype of ts247, implicated also in virus aggregation and heat lability, could not be detected by biochemical studies. Analysis of homotypic recombinants obtained by crossing ts247 with a guanidine-resistant derivative of a temperature-sensitive replicase mutant mapped this mutation to the Pl region or to the 5’ end of the P2 region of the genome. The small-plaque phenotype of ts247 and R247-51 was correlated with an abnormality in polypeptide 3C (protease); direct sequencing of viral RNA revealed a U to C change at nucleotide 5658, which altered an isoleucine to threonine in the protease of ts247 and R247-51 but not of R247-12. Two other mutations were present in the region of the genome coding for polypeptide 3D of ts247 and of both classes of revertants. They thus seemed to play no role in the phenotype of ts247. One mutation, an A to G change at nucleotide 7135, was silent at the protein level, whereas the other, an A to G change at nucleotide 6264, o 1987 Academic PUSS, hc. determined a major amino acid change from glutamate to glycine in the viral replicase.
peptides can be found in infected cells. As an example, polypeptide 3CD, the precursor to 3C and 3D, can follow an alternative cleavage pathway to give polypeptides 3C’ and 3D’ (see Fig. 1). The functions and mechanism of action of many viral polypeptides remain unknown. Genetic studies of poliovirus have previously proved of limited use (for a review, see Cooper, 1977) because until recently conditional mutants were not well defined and were difficult to map. However, biochemical methods now available allow new approaches to genetic analysis, as exemplified by the recent studies of guanidine resistance (Pincus et a/., 1986) and of attenuation of neurovirulence (Omata et a/., 1985; Almond et a/., 1985). We have previously described the isolation and characterization of several temperature-sensitive mutants of poliovirus type 1 (Agut et al., 1981; Bellocq et al., 1984). One of these mutants, ts247, has a complex phenotype: it shows a 3-log inhibition of plaquing efficiency and no significant RNA synthesis at 39”, produces small plaques at 37” compared to wild-type virus, and its virions are very heat labile and readily aggregate, as demonstrated by electron microscopy. Two mutations were detected in the ts247 genome by Tl oligonucleotide mapping, one in the region coding for the protease 3C and the second in the region coding for the replicase 30 (protein nomenclature is according to Ruecker-t and Wimmer, 1984). Polypeptides 3C, 30, 3CD, and 3C’ showed abnormal migration on SDS-
INTRODUCTION Poliovirus has been intensively studied since its identification as the agent responsible for poliomyelitis (Landsteiner and Popper, 1908). The three-dimensional structure of the virion has recently been determined (Hogle eta/., 1985). The viral genome is a monocistronic mRNA, approximately 7500 nucleotides long, of known sequence (Kitamura et al., 1981; Racaniello and Baltimore, 1981). All identified virus-encoded polypeptides are generated by multiple proteolytic cleavages of a single translation product, and they have been mapped onto the viral RNA (Kitamura et a/., 1981; Pallansch et a/., 1984). From 5’to 3’the coding region of the genome can be divided into three parts, Pl, P2, and P3, corresponding to the major initial cleavage products of the polyprotein. Pl codes for the capsid polypeptides VPl, VP2, VP3, and VP4, while P2 and P3 code for nonstructural proteins and for 3B, the small protein covalently linked to the 5’ end of the genomic RNA. It is generally accepted that P3 codes for four final products, including one of the viral proteases (3C) and the viral replicase (3D) whose genes are located adjacent to one another. Precursors and alternative cleavage products are abundant; up to 27 viral-encoded poly’ Present address: DBpartement de Microbiologic. GenBve. 1205 Geneva, Switzerland. ’ To whom reprint requests should be addressed.
Universit6
de
75
0042-6822187
$3.00
Copynght 0 1987 by Academic Press. Inc. All nghts of reproduction in any form reserved
76
BELLOCQ
polyacrylamide gel electrophoresis and/or isoelectrofocusing (Bellocq et al., 1984). In order to correlate the functional defects of ts247 with the mutations detected biochemically, we have selected and analyzed revertants of ts247 and homotypic recombinants obtained by crossing ts247 with a temperature-sensitive replicase mutant (Agut et a/., 1984). As is shown, neither the mutation in polypeptide 3C nor that in polypeptide 3D is responsible for the temperature-sensitive phenotype of ts247. An unidentified mutation in the 5’half of the genome is implicated in this phenotypic trait, while the mutations already detected are responsible for the small-plaque phenotype.
Recombinants between ts247 and ts035Gr, a guanidine-resistant derivative of a temperature-sensitive replicase mutant, were obtained using the infectious center method (McCahon and Slade, 1981) as described previously (Agut et a/., 1984).
Phenotypic analysis of viruses Virus yields after infection of HeLa cells and heat stability of virus stocks were determined as previously described (Agut eta/., 1981). Briefly, to determine heat inactivation, virus stocks diluted in PBS to about lo* PFU/ml were incubated at 45’ for 10 min, immediately diluted in PBS at 4’, and titrated by plaque assay at 37”. Titers were compared with those of control samples kept at 4’.
MATERIALS AND METHODS Virus strains and cells The temperature-sensitive poliovirus mutant ts247 was derived from the type 1 Mahoney strain by mutagenesis with nitrous acid as described (Agut et al., 1981). Unless otherwise indicated, viruses were grown and plaque assayed at 37” on the Hopkins strain of HeLa cells. Biochemical analyses were carried out using these cells or the ATCC strain of HeLa cells.
Analysis of viral polypeptides HeLa cells were infected at a multiplicity of 100 PFU/ cell and incubated at 37” for 4 hr. The cells were next incubated in methionine-free medium for 30 min and then labeled for 1 hr with 30 &i/ml of [35S]methionine (3000 Ci/mmol, Amersham) before being lysed. Preparation of cytoplasmic extracts was as previously described (Bellocq et al., 1984). Samples of the extracts (1 O6 cpm) were denatured in 2% SDS, 5% 2-mercaptoethanol at 100” for 2 min, and analyzed by electrophoresis on a lo-18% SDSpolyacrylamide gel (Hanecak et al., 1982). Cytoplasmic extracts were also subjected to two-dimensional polyacrylamide gel electrophoresis, using isoelectric focusing (IEF) from pH 3.5 to pH 10 in the first dimension and SDS-PAGE in the second (O’Farell, 1975; Lowe et a/., 1981; Wiegers and Dernick, 1981). Prior to IEF, each sample was incubated for 15 min at 37” in the presence of 3.5% (v/v) 2-mercaptoethanol, 9 M urea, and 0.1 mg/ml RNase A as previously described (Bellocq et a/., 1984).
Isolation of revertants and recombinants Spontaneous first-generation revertants of ts247 were isolated from randomly picked plaques obtained after infection of HeLa cells with ts247 at 39” (nonpermissive temperature). Each isolate was plaque purified twice at 39” before virus stocks were made at 37”. These revertants gave small plaques at 39”. Variants with large-plaque phenotype could be isolated after infection of HeLa cells with first-generation revertants at 39’, and were assumed to be second-generation revertants of ts247. These second-generation revertants were also plaque purified twice at 39” before virus stocks were prepared at 37”.
5110
5372 lsq3fl
59.87
.
. *
.
3AB 3A 13Jl
I
ET AL.
6431
7370
. 3CD
3c
* 3C’
3D .
30’
FIG. 1. Schematic diagram of the major products of proteolytic processing in the P3 region of the poliovirus genome. Polypeptides derive from the primary cleavage product, P3 (BABCD) by cleavage at glutamine-glycine (A) or tyrosine-glycine (0) sites. Nucleotide positions on the genome are indicated on the broad upper line (after Pallansch er al., 1984).
A ts PLURIMUTANT
77
OF POLIOVIRUS
TABLE 1 PHENOTYPIC PROPERTIES 0~ts247
AND ITS REVERTANTS
Virus yieldc
Plaque sizeb Virus
Titerat 39” Titerat 37”
wt ts247 R247-51
R247-12
Heat lability of virionsB
Aggregation virions
of 39”
37”
>4
3.2X 10'
1 5x1o-4 1
1.5 5 1.5
+ -
3 1.5 1.5
1
1.5
-
3
>4
B Log drop in titer at 37”, after incubation at 45” for 10 min. b Diameters of plaques (mm) 48 hr after infection of HeLa cells. ’ HeLa cells were infected for 8 hr with 20 PFU per cell and broken open by freeze-thawing, at 37”.
RNA fingerprinting
and sequencing
Viral RNA was purified as described by Lee et al. (1979). The ts247 virions were purified by isopycnic centrifugation through cesium chloride, whereas revertants and recombinants were purified through sucrose gradients in the presence of 0.5% SDS. RNase Tl oligonucleotide mapping of viral RNA (De Wachter and Fiers, 1972) was carried out as previously described (Bellocq et a/., 1984). Briefly, RNA was digested with RNase Tl (PL Biochemicals) and simultaneously 5’ labeled with [3’P]ATP by T4 polynucleotide kinase (PL Biochemicals). Tl -resistant oligonucleotides were subsequently separated by two-dimensional polyacrylamide gel electrophoresis. Viral RNA was sequenced by the dideoxynucleotide chain termination method (Qu et a/., 1983), using synthetic oligonucleotides as primers for reverse transcription (Dreano et a/., 1985). RESULTS Phenotypic
characterization
of ts247 revertants
Spontaneous first-generation revertants of ts247, characterized by their ability to form plaques at 39’, were isolated with a frequency of approximately 5 X 1Oe4. All 15 of the revertants analyzed (typified by R247-51) exhibited the same phenotype (Table 1). Their titer was the same at 39 as at 37”. In addition, their virions were as heat stable as those of wild-type virus, and could readily be purified on SDS-sucrose gradients. The ts247 cannot be purified by this method, like the Sabin type 1 strain, nor on sarkosyl-sucrose gradients. SDS or sarkosyl inactivation experiments have shown that this characteristic was not linked with inactivation by detergents, but was probably due to aggregation of virions which strongly increased their sedimentation rate. Aggregation of ts247 particles was readily observed by electron microscopy, after purification by isopycnic centrifugation through cesium
39”
37”
6.3x
1.3x
lo8
10' 1.3X106
5.0x 3.8x
lo8 10’
1.6x 10’ 6.1 X lOa
and the viral titer (PFWml) was determined
chloride, whereas cesium chloride-purified R247-51 showed completely dissociated particles (data not shown). The first-generation revertants of ts247 were different from wild-type virus as shown by the fact that they produced smaller plaques at 39 than at 37“, and they produced smaller plaques than wild-type virus at both temperatures. This could be related to the virus yield from HeLa cells infected for 8 hr. The yield of R247-5 1 was about 0.5 log lower at 39 than at 37’, whereas that of wild-type virus was increased by about 0.3 log, and that of ts247 reduced by 2 logs (Table 1). Large-plaque variants of R247-5 1 could be isolated at 39” with a frequency of about 5 X 1Op3. These viruses were assumed to be second-generation revertants of ts247. The three viruses tested (typified by R247-12) behaved phenotypically as wild-type virus (Table 1). Thus it seemed that the complex phenotype of ts247 could be segregated in two: temperature sensitivity, heat lability, and aggregation of the virions reverted concomitantly and separately from the small-plaque phenotype. The reversion frequencies observed for the two sets of properties were compatible with the implication of two point mutations. Analysis
of viral polypeptides
Two alterations of ts247 proteins have already been detected, one affecting the protease (3C) and the other the replicase (3D) (Bellocq et a/., 1984). The polypeptides of the first- and second-generation revertants were analyzed for the presence of these modifications by SDS-polyacrylamide gel electrophoresis (Fig. 2) and by two-dimensional gel electrophoresis (Fig. 3). Surprisingly, the electrophoretic pattern of R247-51 polypeptides was identical to that of ts247 concerning the polypeptides encoded by the P3 region of the genome. Polypeptide 3C migrated slowly on SDS-PAGE, whereas 30 migrated rapidly, and both were present
BELLOCQ
78 B
1234
ET AL.
The small-plaque phenotype of ts247 and first-generation revertants (R247-51) could be correlated with the presence of an altered 3C. -3CD -3D -3C’
FIG. 2. SDS-polyacrylamide gel analysis of viral polypeptides. Cytoplasmic extracts (10’ cpm) from infected HeLa cells were analyzed on a 1O-l 8% linear gradient SDS-polyacrylamide gel. The positions of wild-type viral polypeptides and molecular weight markers run in parallel are indicated. (A) Cells were infected with R247-51 (lane l), ts247 (lane 2) or wild-type virus (lane 3). Purified virions (lane 4) and mock-infected cells (lane 5) were included as controls. (B) Cells were infected with wild-type virus (lane 2) ts247 (lane 3) or R247-12 (lane 4). A mock-infected cell extract (lane 1) was run in parallel.
in reduced amounts when compared to the wild-type virus pattern (Fig. 2A, compare lanes 1 and 2 with lane 3). Also, polypeptides 3D, 3CD, and 3C’ from ts247 and R247-51 showed identical pl values (Fig. 3B), different from those of the corresponding wild-type polypeptides Fig. 3A). It should be noted that ts247 repeatedly showed an extremely weak VP2 band (Figs. 2A, lane 2, and 2B, lane 3). In contrast, the second-generation revertants gave a unique protein profile. On SDS-PAGE (Fig. 2B, lane 4) polypeptide 3C migrated identically to the wild-type equivalent, while polypeptide 3D migrated like the ts247 homologous polypeptide, but its amount was significantly increased compared to ts247 (compare lanes 3 and 4 of Fig. 2B). Polypeptides 3CD and 3C’ of ts247 migrated like those of wild-type virus, whereas the R247-12 counterparts migrated more rapidly (compare Fig. 2B, lane 4, with lanes 2 and 3). Figure 3 allowed the p/values of R247-12 polypeptides to be compared to those of ts247 (Fig. 3C) and wild-type virus (Fig. 3D). The p/values of 3CD, 3D, and 3C’were similar to those of the ts247 equivalents, and clearly distinct from those of the corresponding wild-type polypeptides. The fast migration rate of polypeptides 3CD and 3C’ of R24712 was clearly visible, particularly in Fig. 30 because of the difference in pl values of these polypeptides between R247-12 and wild-type virus. It thus appeared that the temperature sensitivity, heat lability, and aggregation of ts247 were not related to the alterations observed in polypeptides 3C and 3D.
Tl -resistant oligonucleotide
mapping
In order to confirm these results, the RNA from ts247 and from revertants R247-51 and R247-12 was subjected to RNase Tl mapping (Fig. 4). The first-generation revertants (R247-51) showed the same abnormalities as ts247: a charge change in Tl oligonucleotide No. 10 (numbered according to Lee er al., 1979) which maps at nucleotide 5644, in the region of the genome coding for polypeptide 3C; and the disappearance of oligonucleotide No. 8 located at nucleotide 7125, in the sequence coding for polypeptide 3D (compare Figs. 4A, 48, and 4C). Oligonucleotide 8 remained absent in the case of the second-generation revertants, whereas oligonucleotide 10 had regained its normal charge (compare Figs. 4A and 4D). The modification of Tl oligonucleotide No. 10 correlated with the alteration of polypeptide 3C, suggesting the likely location of the mutation responsible for the small-plaque phenotype of ts247 and R247-5 1. On the other hand, the mutation indicated by the alteration of oligonucleotide No. 8, which is located at nucleotide 7125, could not be implicated in abnormal pl values
FIG. 3. Two-dimensional PAGE analysis of viral polypeptides. Mixtures of infected cell extracts: ts247 + wild-type (A); ts247 + R24751 (B); ts247 + R247-12 (C); wild-type + R247-12 (D) were subjected to IEF followed by SDS-PAGE (Bellocq ef a/., 1984). Second-dimension electrophoresis was on a 1O-l 8% (A,B) or a 15% (CD) SDSpolyacrylamide gel.
A ts PLURIMUTANT
FIG. 4. RNase Tl fingerprints of viral RNA. (A) Wild-type; (B) ts247; (C) R247-51; (D) R247-12. Oligonucleotide No. 10 is indicated by a black arrow, and oligonucleotide No. 8 (or its absence) by a hollow arrow.
and migration on SDS-PAGE of polypeptides 3D, 3CD, and 3C’, since 3C’ is not coded by this region of the genome. The concomitant alteration of 3D, 3CD, and 3C’ suggested a mutation between nucleotides 5987 and 6431, the region of the genome coding simultaneously for these three polypeptides (Fig. 1). Nucleotide
sequence
determination
To locate and identify the mutations in the P3 region of the genome, we sequenced the RNA of ts247 in the regions suggested above as carrying mutations. Three mutations were identified (Table 2) one at nucleotide 5658 in the region coding for the protease 3C, and two (one of which is silent at the protein level) in the region coding for the replicase 3D, at nucleotides 6264 and 7135. R247-51 carried these same three mutations, whereas R247-12 had reverted to the wild-type sequence at nucleotide 5658, but had kept the two mutations in the part of the genome coding for polypeptide 3D. These three mutations are sufficient to explain the abnormalities observed in the proteins and RNA of ts247 and its revertants. Nucleotide 5658 falls within Tl oligonucleotide No. 10, and a change from U to C at this position will alter the charge of the oligonucleotide. The resultant change
79
OF POLIOVIRUS
from isoleucine to threonine in polypeptide 3C does not alter the overall charge of the protein, but modifies its hydrophilicity. This could affect the migration of the polypeptide on SDS-PAGE, but not its isoelectric point, as seen. Nucleotide 6264 does not lie within any of the large Tl oligonucleotides readily identified by fingerprinting (Kitamura et al., 1981), which would explain the failure of this method to detect the mutation. The glutamate to glycine change at this point simultaneously changes the charge and hydrophilicity of 3CD, 3D, and 3C’. The presence of this mutation is consistent with the charge changes observed by two-dimensional electrophoresis. SDS-PAGE migrations can be explained if the mutation at nucleotide 6264 results in faster migration of 3D than normal, whereas the mutation at nucleotide 5658 results in slower migration of 3C than normal. If the two mutations compensate each other, polypeptides which carry both alterations would migrate normally (3CD and 3C’of ts247 and R247-51); in contrast, those which only carry the modification resulting from the mutation at nucleotide 6264 (e.g., 3CD and 3c’of R24712) would migrate faster than normal, as observed. Nucleotide 7135 falls within Tl oligonucleotide No. 8. An A to G change at this position introduces an additional Tl RNase cut and hence the disappearance of this oligonucleotide. This mutation does not modify the amino acid sequence, and therefore has no effect on protein migration.
TABLE 2 LOCATIONOF MUTATIONSIN THE P3 REGIONOFTHE GENOMESOF ts247 AND ITSTwo CLASSESOF REVERTANTS Nucleotide number: Polypeptide: Amino acid number in the polypeptide: Sequence ts247
5658 Protease (3C)
6264 Replicase (3D)
7135 Replicase (3D)
74
93
383
ACA Thr
AAG LYS AAG LYS AAG LYS AAA LYS
R247-51
ACA Thr
R247-12
AUA Ile
Wild-type
AUA Ile
GGA GIY GGA GIY GGA GIY GAA Glu
0
f4.3
0
+1.4
-3.0
0
Charge change of amino acid Hydrophilicity change of amino acide a Calculated
after Hopp and Woods (1981)
80
BELLOCQ
Analysis of recombinants None of the mutations detected by Tl fingerprinting or polypeptide analysis could be responsible for the ts phenotype of ts247. The fact that the ts character seemed to be linked to thermolability and virion aggregation could implicate a mutation in the Pl region of the genome, which codes for the capsid proteins. To verify this hypothesis, the phenotypes of ts+ recombinants obtained from the cross ts035Gr X ts247 were analyzed. The ts035Gr is a guanidine-resistant derivative of a temperature-sensitive replicase (3D) mutant of poliovirus. The advantage of this approach is that the ts247 ts mutation can be located with reference to an independent marker, resistance to guanidine. There is now convincing evidence that guanidine resistance maps to the middle of the poliovirus genome, within the region coding for polypeptide 2C (Pincus et a/., 1986). Hence, if the ts mutation of ts247 is located 3’ of the guanidine-resistance marker, like the ts mutation of ts035, recombinants generated by a single crossing over between the two ts mutations would be either all guanidine resistant or all guanidine sensitive, depending on the position of the ts mutations relative to each other. If, on the other hand, the ts mutation of ts247 is located 5’ of the guanidine resistance, both guanidine-sensitive and guanidine-resistant recombinants would be generated by single crossing-over events. Twelve ts+ recombinants were picked and characterized. They could be classified into three groups (Table 3) typified by RCOOl (4/l 2) which is guanidine sensitive, and RC003 (2/12) and RC008 (6112) which are
TABLE 3 PHENOTYPICPROPERTIESOF RECOMNNANTSBETWEEN ts035Gr AND ts247
Virus Parents ts035Gr ts247 Recombinants RCOOl RC003 RC008
Titerat 39” Titerat 37”
Heat lability of virionsB
Resistance to guanidineb
Plaque size at 37”’
2x1o-3 5x 1o-4
+
+ -
3 1.5
-
1.5 1.5 3
1 1 1
+ +
a Virus was considered as heat labile if it exhibited a drop in titer of more than 2 logs after heating for 10 min at 45”. * Virus was considered as resistant if it exhibited the same titer in the absence and in the presence of 1 mM guanidine-HCI. c Measured as diameter of plaques in mm 48 hr after infection of HeLa cells.
ET AL.
guanidine resistant. RCOOl and RC003 have the same phenotypic characteristics as R247-51, whereas RC008 resembles R247-12. Preliminary analysis showed that, as in the case of revertants, small-plaque phenotype correlated with the presence of the mutation at nucleotide 5658 (Agut et a/., in preparation). Thus the recombinants obtained carried the 5’ end of the ts035 genome, linked to the 3’ end of the ts247 genome. A comparison of RCOOl and RC003 showed that exclusion of the two ts mutations was independent of acquisition of the guanidine-resistance mutation from ts035. Assuming that most recombinants were generated by a single crossing over, these results demonstrate that the ts mutation of ts247 is separated from that of ts035 by the mutation responsible for guanidine resistance. Consequently, we concluded that the ts247 mutation is located in Pl or in the 5’ end of P2.
DISCUSSION The main advantage of chemical mutagenesis of viruses over the presently employed techniques of sitedirected mutagenesis of viral genomes is the relative ease with which conditional lethal mutant viruses can be obtained, for example, by screening for a readily detectable phenotype such as temperature sensitivity. However, several mutations can accumulate in a single genome, and it may be extremely laborious to map the mutation(s) implicated in a given defect. A temperaturesensitive mutant, ts247, was obtained after such chemical mutagenesis of poliovirus type 1, and preliminary analyses suggested that multiple mutations were responsible for its complex phenotype. Partial sequencing of the ts247 genome has allowed us to identify three mutations in the P3 region. One of these mutations located at nucleotide 5658 results in an isoleucine to threonine change in the viral protease, polypeptide 3C. A second mutation, located at nucleotide 6264, changes a glutamate to glycine in the replicase, polypeptide 3D. The third mutation, also located in the region of the genome coding for polypeptide 3D (nucleotide 7135), is silent at the protein level. The presence of these three mutations could readily be ascertained from Tl fingerprints of the viral RNA and/or from viral polypeptide migration patterns. Studies of spontaneous ts+ revertants of ts247 have allowed us to evaluate the correlation between these three mutations and viral phenotype. A comparison between the first- and second-generation revertants of ts247 showed that the presence of an altered protease, due to the mutation at nucleotide 5658, correlated with a small-plaque phenotype. In ts247 and its first-generation revertants (R247-5 l), the altered migration and decreased amount of polypeptide 3C were associated
A ts PLURIMUTANT
with a decreased amount and a converse alteration of migration of 3D. We first thought that the mutation at nucleotide 5658 resulted in cleavage of 3CD at an abnormal site to give a larger 3C and smaller 30 than normal. However, altered SDS-PAGE migration of 3D was dissociated from that of 3C in the case of the second-generation revertants (R247-12). The decreased amount of 30, on the other hand, was associated with the alteration of 3C. It thus appears that the isoleucine to threonine change at residue 74 of the viral protease decreases the amount of functional replicase, probably through impaired protease activity. That this is a major factor responsible for the small plaque size of ts247 and R247-51 has been confirmed by molecular cloning of the protease mutation (Kean er al., in preparation). At first view, the two mutations detected in the replicase gene have no effect on virus growth. Secondgeneration revertants (R247-12) of ts247, which carried both these mutations, behaved phenotypically as wildtype virus. It seems surprising that the viral replicase can apparently tolerate a relatively major change from glutamate to glycine, resulting from an A to G change at nucleotide 6264. However, it has been observed (Dreano et a/., 1985) that different stocks of poliovirus type 1 (Mahoney strain) differ at the preceding codon, having either a threonine or an isoleucine at amino acid 92 in the replicase. The amino acids at these two positions are not conserved in different picornaviruses (Argos et al., 1984). It thus appears that conservation of the amino acid sequence in this region of polypeptide 3D is not important for its activity. It would be interesting to investigate to what extent the virus can support changes in this region of its genome, using site-directed mutagenesis of cloned viral cDNA. None of the mutations identified in the P3 region of the genome could be held responsible for the temperature-sensitive character of ts247 since the first-generation revertants were ts+ but had retained all three mutations. Such reversion of ts247 to a ts+ phenotype (R247-51) was accompanied by the loss of heat lability and aggregation of the virions. The high reversion frequency observed implied that temperature sensitivity, thermolability, and aggregation of ts247 arose concomitantly from the presence of a single point mutation. The nature of these phenotypic characteristics suggested that this mutation lay in the region of the genome coding for the capsid proteins. Analysis of homotypic recombinants formed by genetic recombination between ts247 and a guanidine-resistant ts replicase mutant (ts035Gr) mapped the ts mutation in ts247 5’ of the region of the genome coding for polypeptide 2C. To date, we have been unable to detect a biochemical marker that could help to map this mutation more precisely.
81
OF POLIOVIRUS
The observed phenotype of ts247 is due to a cumulative effect of multiple mutations. This experimental model is imperfectly characterized as yet, but is of great interest in view of the search for safer attenuated poliovirus vaccines based on the idea of accumulating mutations to increase attenuation and prevent reversion to neurovirulence. It is of note that mutations involving a relatively major amino acid change, such as the glutamate to glycine change at residue 93 in the viral replicase, can apparently be without functional effect. Conversely, it has already been shown in the case of picornaviruses that mutations with no effect on protein sequence, notably mutations in the noncoding regions of the genome, can have profound phenotypic effects on the virus (Almond et a/., 1985; Sarnow et al., 1986; Racaniello, personal communication). As will be shown elsewhere (manuscript in preparation) the mutation at nucleotide 7135 of ts247, albeit silent at the protein level, is implicated in a temperature-dependent sensitivity of viral RNA replication to actinomycin D.
ACKNOWLEDGMENTS This work was supported in part by INSERM Grant CA 40501, by NATO Grant RG 175/80, and by MRT Programme Biotechnologies.
REFERENCES AGUT, H., BELLOCQ, C., VAN DER WERF, S., and GIRARD, M. (1984). Recombination and rescue between temperature-sensitive mutants of poliovirus type 1. Virology 139,393-402. AGUT. H., MATSUKURA.T., BELLOCQ, C., DREANO, M., NICOLAS, J. C., and GIRARD, M. (1981). Isolation and preliminary characterization of temperature-sensitive mutants of poliovirus type 1. Ann. Viral. (Inst. Pasteur) 132E, 445-460. ALMOND, I. W., WESTROP, G. D., CANN, A. I., STANWAY, G., EVANS, D. M. A., MINOR, P. D., and SCHILD, G. C. (1985). Attenuation and reversion to neurovirulence of the Sabin polio type 3 vaccine. ln “Vaccines 85,” pp. 271-277, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. ARGO& P., KAMER, G., NICKLIN, M. J. H., and WIMMER, E. (1984). Similarity in gene organization and homology between proteins of the animal picornaviruses and a plant comovirus suggest common ancestry of these virus families. Nucleic Acids Res. 12, 72517276. BELLOCQ, C., AGUT, H., VAN DER WERF, S., and GIR~RD, M. (1984). Biochemical characterization of poliovirus type 1 temperaturesensitive mutants. Virology 139, 403-407. COOPER,P. D. (1977). Genetics of picornaviruses. In “Comprehensive Virology” (H. FraenkeCConrat and R. R. Wagner, Ed%), Vol. 9, pp. 133-207. Plenum, New York. DE WACHTER, R., and FIERS,W. (1972). Preparative two-dimensional polyacrylamide gel electrophoresis of 3ZP-labelled RNA. Anal. Biochem. 49, 184-l 97. DREANO, M., BELLOCCI,C., FICHOT, 0.. VAN DERWERF, S., and GIRARD, M. (1985). Genetic variations in the Mahoney strain of poliovirus type 1. Ann. Vifol. (Inst. Pasteur) 136E, 10 l-l 14. HANECAK,R., SEMLER,B. L., ANDERSON,C. W., and WIMMER, E. (1982). Proteolytic processing of poliovirus polypeptide: antibodies to polypeptide P3-7c inhibits cleavage at glutamine-glycine amino acid pairs. Proc. Natl. Acad. Sci. USA 79, 3973-3977.
82
BELLOCQ
HOGLE,J. M., CHOW, M., and FILMAND, D. J. (1985). Three-dimensional structure of poliovirus at 2.9 A resolution. Science 229, 13581365. HOPP, T. H., and WOODS, K. R. (1981). Prediction of protein antigenic determinants from amino acid sequences. Proc. Nat/. Acad. Sci. USA 78,3824-3828. KITAMURA,N., SEMLER, B. L., ROTHBERG,P. G., LARSEN,G. R., ADLER, C. J., DORNER, A. J., EMINI, E. A., HANECAK, R., LEE, J. J., VAN DER WERF, S., ANDERSON,C. W., and WIMMER, E. (1981). Primary structure, gene organization and polypeptide expression of poliovirus RNA. Nature (London) 291, 547-553. LANDSTEINER,K., and POPPER, E. (1908). Mikroskopische praparate von einem menschlichen und zwei Aftenrijckenmarken. Wein. K/in. Wochenschr. 21, 1830. LEE, Y. F., KITAMURA, N., NOMOTO, A., and WIMMER, E. (1979). Sequence studies of poliovirus RNA. IV. Nucleotide sequence complexities of poliovirus type 1, type 2 and two type 1 defective interfering particles RNAs, and fingerprint of the poliovirus type 3 genome. J. Gen. Viral. 44, 311-322. LOWE, P. A., KING, A. M. Q., MCCAHON, D., BROWN, F., and NEWMAN, J. W. I. (1981). Temperature-sensitive RNA polymerase mutants of a picornavirus. Proc. Nat/ Acad. Sci. USA 78, 4448-4452. MCCAHON, D., and SLADE, W. R. (1981). A sensitive method for the detection and isolation of recombinants of foot-and-mouth disease virus. J. Gen. Viral. 53, 333-342. O’FARRELL, P. H. (1975). High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 250, 4007-4021. OMATA, T., KOHARA, M., ASE, S., ITOH, H., KOMATSU, T., ARITA, M., SEMLER, B. L., WIMMER, E., KUGE, S., KAMEDA, A., and NOMOTO, A.
ET AL. (1985). Construction of recombinant viruses between Mahoney and Sabin strains of type 1 Poliovirus and their biological characteristics. In “Vaccines 85,” pp. 279-283. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. PALLANSCH, M. A., KEW, 0. M., SEMLER, B. L., OMILIANOWSKI,D. R., ANDERSON,C. W., WIMMER, E., and RUECKERT,R. R. (1984). Protein processing map of poliovirus. J. Viral. 49, 873-880. PINCUS, S. E., DIAMOND, D. C., EMINI, E. A., and WIMMER, E. (1986). Guanidine-selected mutants of poliovirus: Mapping of point mutations to polypeptide 2C. J. Viral. 57, 638-646. Qu, L. H., MICHOT, B., and BACHELLERIE, J. P. (1983). Improved methods for structure probing in large RNAs: A rapid “heterologous” sequencing approach is coupled to the direct mapping of nuclease accessible site. Application to the 5’terminal domain of eukaryotic 28s rRNA. Nucleic Acids Res. 11, 5903-5920. RACANIELLO,V. R., and BALTIMORE, D. (1981). Molecular cloning of poliovirus cDNA and determination of the complete nucleotide sequence of the viral genome. Proc. Nat/. Acad. Sci. USA 78,48874891. RUECKERT,R. R., and WIMMER, E. (1984). Systematic nomenclature of picornavirus proteins. J. Viral. 50, 957-959. SARNOW,P., BERNSTEIN,H. D., and BALTIMORE,D. (1986). A poliovirus temperature-sensitive RNA synthesis mutant located in a noncoding region of the genome. Proc. Natl. Acad. Sci. USA 83, 571575. WIEGERS, K. J., and DERNICK,R. (1981). Poliovirus-specific polypeptides in infected HeLa cells analysed by isoelectric focusing and 2Danalysis. J. Gen. Viral. 52, 6 l-69.