GB2344590A

GB2344590A - Poliovirus having mutations in the Protease 2A gene

Info

Publication number: GB2344590A
Application number: GB9917024A
Authority: GB
Inventors: Andrew Joseph Macadam; Philip David Minor
Original assignee: BTG International Ltd
Current assignee: BTG International Ltd
Priority date: 1998-08-06
Filing date: 1999-07-20
Publication date: 2000-06-14
Anticipated expiration: 2019-07-20
Also published as: GB2344590B; GB9917024D0; WO2000007622A1; GB9817082D0

Abstract

A vaccine comprising a poliovirus having a mutation in the protease 2A gene. The mutation is associated with reduced temperature sensitivity but the degree of attenuation of the virus is unaffected. The mutations must not change residues 20 (His), 38 (Asp) or 109 (Cys) which crate the active site of the protease. The vaccine may be live or inactivated.

Description

2344590 ATTENUATED POLIOVIRUS VAMNE

Background of the invention

1. Field of the invention

This invention relates to a vaccine containing an attenuated poliovirus.

2. Description of the related art

The live attenuated poliovirus vaccines developed by Sabin in the 1950's have found great use throughout the world. Vaccine strains derived from each of the three poliovirus serotypes, known as Sabin types 1, 2 and 3, were prepared by passage of wild-type viruses in cell cultures and whole animals until attenuated strains were obtained. These attenuated viruses are substantially less able to cause poliomyelitis in humans than the original wild-type strains. They are administered orally and replicate in the gut to induce a protective immune response.

Although these vaccines are generally regarded as safe, their use is associated with a small incidence of paralysis in vaccinees. This is most often associated with serotypes 2 and 3 and rarely, if ever, with serotype 1. There is therefore a requirement for improved serotype 2 and scrotype 3 vaccines which would be comparable in safety to the excellent serotype I strain. Consequently, it is desirable to find genetic alterations to the Sabin strain which preserve their attenuation and reduce the risk that they will revert to wild type.

In prior PCT application WO 98/41619, alterations were made in the 5'-non coding region. Their effect on attenuation was screened first by measuring temperature sensitivity. It had been previously shown that particular changes in domain V of the 5'-non-coding region (NCR) that weakened its RNA secondary structure resulted in increased temperature sensitivity of viral growth in BGM or L20B cells. That is, when the virus is genetically altered in such a way in domain V it can no longer grow at elevated temperatures and grows less well at normal temperatures (around 37'Q than the strain from which it was derived. It is therefore more temperature-sensitive. In the said prior PCT application, it was confirmed that this increase in temperature sensitivity correlated with an increase in attenuation (=decrease in virulence). In that application, the temperature sensitivity was represented by the parameter "T", being the temperature at which the number of plaque-forming units (pfu) was reduced by a power of 10 (=logio pfu reduced by 1) from the number of plaque-forming units obtained at 35'C. The lower the value of T, the greater the temperature sensitivity and the greater the degree of attenuation.

In view of the success of the WHO global polio eradication programme it is likely that the world will soon be declared free of wild poliovirus. At that time the laboratory stocks of wild polioviruses present in research and diagnostic laboratories and vaccine production facilities will present a serious threat and will have to be handled under stringent containment conditions. Such conditions will severely hinder the production of inactivated poliovirus vaccines, which are currently composed of wild strains because of their superior growth characteristics. Release of wild polioviruses from such facilities into the community has been documented on at least two occasions. One solution to this problem would be to use attenuated strains for the production of inactivated poliovirus vaccines, if they could be made to grow better.

Summary of the inveption it has now surprisingly been found that alterations in a completely different region of the poliovirus genome result in reduction of temperature sensitivity without any reduction in the degree of attenuation. In other words, this is contrary to what was expected from experience in domain V.

Further, because the attenuated viruses grow better at the normal temperature for viral growth (around 37'C) than the strains from which they are derived, notably Sabin strains, they could be used to produce inactivated poliovirus vaccines.

Moreover, selection pressures against attenuating mutations during virus growth would be reduced, so that vaccine strains incorporating these changes could be grown without selection of neurovirulent revertants, thus increasing the safety of live attenuated vaccine preparations.

The invention is derived from the idea of investigating the effect on attenuation of genetic mutations in part of the coding region. (The term "mutation" as used herein, comprises variation, addition or deletion.) The region investigated is that encoding protease 2A. In A. J. Macadam et aL, EMBO J. 13, 924-927 (1994) the inventors and their colleagues reported experiments in which mutations were made to the domain V (domain VI as it was then called) of the 5'-NCR region of poliovirus and the viruses grown at 39'C in BGM cells. The effect of these changes was to increase temperature sensitivity. Thus, logio [p.f u. at 35'C/p.f.u. at 39'C] increased in the Leon Lansing (LL) strain from 0. 1 to between 2 and 4, depending on the precise mutation. When the plaques of these strains containing the mutations, grown at 39"C or 39.3'C, were picked to select so-called revertants (viruses which might have reverted to the 5'-NCR of their parent), it was found that the "revertants" were less temperature-sensitive than the viruses from which they were derived, despite retaining the original 5'-NCR mutations. It was shown that these selected "revertants" had undergone mutation in the protease 2A gene. The same point mutations, but solely in the protease 2A gene, were then introduced into Sabin 2 polioviruses, which also resulted in loss of temperature sensitivity. According to experience with mutations in the 5'-NCR, loss in temperature sensitivity should correlate with a loss of attenuation, thus indicating that these mutants would be of no interest for making a vaccine.

However, according to the present invention, a wide variety of mutations in the protease 2A gene lead to a loss of temperature sensitivity without decrease in attenuation. It is not yet clear precisely which (loss of) ftinction of the protease 2A gene correlates with the decrease in temperature-sensitivity. While not wishing to be bound by any theory, the inventors believe that the Sabin strain viruses have a defect in their translation which is associated with temperature sensitivity and that the effect of the mutations in the protease 2A gene is to suppress this defect, thus enhancing translational efficiency at higher temperatures. Further, one function of the protease 2A gene is to cleave a host cell protein, which thereby releases a factor reportedly capable of acting in trans on translation initiation. However, the factor may remain partially associated with the protease. Mutation of the protease 2A gene may cause this factor to disassociate from the protease and act to increase translational efficiency. Whatever may be the mechanistic explanation, the loss of temperature sensitivity provides a way of testing whether a particular mutation will be effective.

Mutations in the active site of the virus which abolish its protease activity are lethal to the virus and are therefore excluded. The protease active site is bounded by amino acids 20 (His), 38 (Asp) and 109 (Cys). Mutations in these amino acids should therefore be avoided.

The reduced temperature sensitivity with which the mutation in the protease 2A gene is associated, is conveniently defined by an increase in the number of p.f.u.

when the virus is grown at 39'C divided by the number of p.f.u. when the virus is grown at 35'C, under otherwise identical conditions. The logarithmic growth ratio in BGM cells:

[p.f.u. of the virus grown at 35-C] 10910 [p.f.u. of the virus grown at 39C] will generally be a positive fraction of 1, i.e. above 0 and less than 1.

The reduction in temperature sensitivity differs quantitatively between cells.

The effect has been observed in other monkey cells and in human cells, but the BGM cells have been selected in order to provide a standard reference point.

The invention accordingly provides a poliovirus having such a mutation in the protease 2A gene for use in a vaccine against wild type poliovirus and the use of such a mutated poliovirus in the preparation of the vaccine.

Further the invention includes the vaccine wherein the poliovirus may be in an inactivated or live form, and which may also contain an adjuvant or any other ingredient of a type conventional in the formulation of such a vaccine.

Where patent law permits, the invention also includes a method of vaccinating a human subject against wild type poliovirus, which comprises administering to the subject a prophylactically effective dose of the vaccine.

It is expected that many of the mutations introduced into the 2A gene will genetically stabilise attenuating determinants in domain V of the 5'-NCR against reversion to the corresponding poliovirus having wild-type domain V sequence. In particular, mutations in the 2A genes of attenuated poliovirus strains may result in more efficient viral growth in culture and more stable attenuation phenotypes. Such strains would be very useful in the production of both safe live- attenuated and inactivated poliovirus vaccines, the latter particularly at a time when virulent, wild type strains have been eradicated. These 2A mutations may also genetically stabilise attenuating mutations in the 5-NCRs of live vaccine strains during growth in the human alimentary tract and thus reduce reversion frequency in vivo, thereby further improving vaccine safety.

When wild-type strains have been eradicated, they will be unavailable as controls for assays of temperature sensitivity of polioviruses. The vaccine of the invention wherein the poliovirus is of serotype 2, e.g. is Sabin 2 modified as described, is substantially non-temperature sensitive, yet is attenuated. It may therefore serve for use as a control. The term "vaccine", as used herein, covers such formulations.

Description of the preferred embodiments

The protease 2A has 149 amino acids and is well conserved among attenuated polioviruses. The amino acid sequences for the Sabin 1, 2 and 3 strains and four others are set out in A. J. Macadam et a[ (1994) cited above. Mutations at 20 of these positions have been found to decrease temperature sensitivity. Seven mutations at the same position, introduced in vitro, have all been found to decrease temperature sensitivity. It is thus likely that the current list is not exhaustive, though equally it is unlikely that all possible changes to 2A would decrease temperature sensitivity.

Further, as previously mentioned, mutations in the region encoding the active site of the protease enzyme that abolish proteolytic activity are known to be lethal to the virus. This site is controlled by a catalytic triad of amino acids 20 (His), 38 (Asp) and 109 (Cys). None of these should be changed, although it is permissible to make a 64silent" nucleotide variation which does not change the amino acid.

Since temperature sensitivity can be measured relatively easily, it is within the competence of the ordinary person in the art to discover by this means which mutations are suitable.

Generally stated the mutant should have a logarithmic growth ratio in BGM log, [p.f.u. of the virus grown at 35-C] 0 [p.f.u. of the virus grown at 390Ci which is a positive fraction of 1, preferably less than 0.7, more preferably less than 0.6. however, it is unlikely to be less than 0.01; more usually it is not less than 0.1.

The invention is preferably practised on a strain which is already attenuated, preferably a Sabin 1, 2 or 3 strain. Such a strain may be unmodified or, more preferably, modified in its 5'-NCR region as described in the above- mentioned prior PCT application. That patent application recommends avoiding a base-pair mismatch, i.e. non Watson-Crick base pairing, in domain V and, especially, proposes a U-A base pair at positions 472-537 in a type 3 poliovirus such as Sabin 3 or 469- 534 in a type 1 or 2 poliovirus such as Sabin I or 2. These modifications can be adopted in the present invention.

Mutations in the protease 2A gene can be introduced by any of the standard methods of introducing point mutations in DNA. Viruses can be recovered by making an RNA transcript of the DNA using e.g. T7 poIymerase, and transfecting cell monolayers. BGM cells can be infected with the virus and grown at 35'C and 39'C under agar to detect the difference in the virus' ability to grow at these temperatures and thus determine its temperature sensitivity. The logarithmic growth ratio can then be determined.

The poliovirus vaccines of the invention may be formulated in any appropriate conventional way. When preparing an inactivated vaccine, partially purified virus preparations are conveniently treated with any of the agents conventionally used for inactivation, e.g. formalin or P-propiolactone.

The live vaccines of the invention may be formulated with any diluent or carrier which is physiologically suitable, e.g. comprising M902, sucrose and phosphate buffer.

Doses will be similar to those which are conventional in poliovirus vaccines.

The following Examples illustrate the invention. All temperatures are in degrees Celsius. Numbering used to describe the 5-non-coding region is that of serotype 3, unless otherwise stated. Note in particular that "484" in serotype 3 numbering is "481" in serotype 2.

EXAMPLEI

Virus assays The BGM cell line is derived from African Green Monkey kidney cells. BGM cells were grown in monolayers maintained in modified Eagle's medium supplemented with 5% fetal calf serum and buffered with 15 mM HEPES and 0. 02% bicarbonate.

Temperature sensitivity was assayed by plaque formation on BGM cells at different temperatures. Cell sheets in six-well plates were inoculated and then overlayed with 1% agar in MEM supplemented with 0.5% fetal calf serum. The concentration of fetal calf serum used was chosen to maintain the particular cell sheet at elevated temperatures and does not affect the ts phenotype. Plates were incubated in sealed plastic boxes submerged in water baths whose temperatures fluctuated less than 0.01'. Temperatures were determined using a calibrated electronic thermometer with a platinum probe. After 3 days, cell sheets were stained with naphthalene black and plaques counted. All viruses were assayed at least twice and control viruses with known phenotypes were always included for validation.

Selection of revertants The parental viruses from which revertants were selected were serotype 2 viruses deriving the Y-non-coding region (NCR) nucleotides of the genome from the Sabin type 2 vaccine strain, P2/Sabin, or they were site-directed mutants of Leon Lansing. This is a recombinant virus deriving its 5'-NCR (about 750 bases) from the serotype 3 strain P3/Leon and the remainder of its genorne from the serotype 2 strain Lansing (Table 1). Construction of infectious clones of P2/Sabin and P2/117 (an isolate from a vaccine-associated case of poliomyelitis) was as described by S. R.

Pollard et al., J. Virology 63, 4949-4951 (1989). Thus, the viruses were propagated in Hep2C cells as described by P. D. Minor, J. Virology 34, 73-84 (1980). Virus was purified by sucrose gradient centrifugation. Virion RNA was extracted and cloned by using the RNA -cDNA cloning hybrid method. Colonies were screened by colony hybridisation with 32p-labelled nick-translated DNA from an infectious cDNA of a type 3 poliovirus. Plasmids isolated from the positive colonies were characterised by restriction enzyme analysis to identify those containing cDNA covering the entire genome. A T7 promoter was inserted into the plasmid before the 5'-end of the viral genome and the cDNA was transcribed to RNA. Transfection of Hep2C cells with the RNA transcripts enabled the viruses to be recovered. PS/I 175' and P I 17/S5' are reciprocal recombinants thus constructed from the clones of P2/Sabin and P2/117.

PS/I 175' has the first 491 nucleotides of P2/117 (which are in the 5'NCR). The remainder of PS/1175' has the P2/Sabin sequence. PI 17/S5' has the first 491 nucleotides of P2/Sabin and the remainder have the P2/117 sequence. They were prepared as described by A. J. Macadam et aL, Virology 181, 451-458 cited above (199 1). In the 5'-NCR the two viruses differ only at two positions, one of which is in domain V at nucleotide 481 of serotype 2 numbering which is equivalent to nucleotide 484 of serotype 3 numbering. Hereinafter serotype 3 numbering is used unless otherwise stated. P117/S5', like Sabin-2 has A at 484, PS/1175'has Gat 484.

Site-directed mutants having changes in domain V of the 5'-NCR were constructed from Leon-Lansing and recovered as described by A. J. Macadam et aL, Virology 189, 415-422 (1992). The mutants prepared are shown in Table I. The positions of mutations are of the Leon sequence: see Fig. 1 of A J. Macadam et aL, (1994) cited above for the structure of domain V on which the nomenclature is based.

This structure assumes that many bases are paired. Thus 471 (U) is paired with 538 (A), 472 (U) with 537 (G) etc. Referring to Table 1, the mutants are described by showing the bases or base-pairs which have replaced those present in Leon- Lansing.

A signifies a deletion. By way of illustration, mutant LL 471/538 AA had an A at 471, replacing U in Leon-Lansing, retaining A at 538 as in Leon-Lansing.

LL472/537UG had a U at 472, rather than a C as in Leon-Lansing, retaining G at 537 as in Leon-Lansing; mutants LL A472 and LL,&483 had single base deletions at 472 and 483 respectively; mutant LL479/532UC had substitutions at both 479 and 532, creating a mismatched UC base-pair and LL514A had a single mutation U-A at 514.

As a result of these 5'-NCR sequences, growth in BGM cells of all parental viruses was sensitive to elevated temperatures, as illustrated by the ratios of numbers of plaques formed at 35' and 39' (Table 1). In contrast, growth of the strains in which the predicted structure of the 5'-NCR was not disrupted, such as P2/117, PS/I 175' and Leon-Lansing, was similar at 3 9' and 3 5'.

The Sabin 2 virus itself differs from P3/Leon-Lansing in that it has A at 484 in poliovirus type 3 numbering used for P3/Leon-Lansing. This mutation is a strong determinant of temperature sensitivity. PI 17/S5' also has A at 484. These viruses were already temperature-sensitive without making any other change in domain V of the NCR. Five revertants of reduced temperature sensitivity were obtained by selection of revertants from these strains and are shown in the top boxed row of Table 1.

The ts figures quoted in parenthesis in Table I are from an assay where the higher temperature of 39.3'C was used in order better to show the phenotype of the parent virus.

Table I

5'-NCR MUTATIONS 2A CODING CHANGE Logio [pfu at 35'C/pf -at 39'C], BGM cells None 2.2 Sabin 2 (484 A) 8 Ala - Val 0.1 or 19 Tyr - His 0.4 P I 17/S 5' 79 Thr - Ala 0.1 (484 A) 96 His - Tyr 0.6 122 Ile - Val 0.2 None 1.7 LL471/538 AA 65 Glu - Lys 0.5 None 1.8 LL471/538 GA 122 Ile - Val 0.7 None 2.0 Glu - Gly 0.3 33 Ile - Val 0.2 LL472/537 UG 93 Tyr - His 0.9 106 Pro - Ser 0.5 134 Ser - Tbr 0.5 None 1.0 LL472/537 UA 30 Ala - Pro 0 None 4.0 LLA 472 80 Phe - Leu 0.5 17 Cys - Tyr 0.7 None 0.9 (2.0 at 39.30C) LL473/537 UG 45 Lys - Glu 0.1 (0.5 at 39.30C) 79 Thr - Ala 0.2 (0.4 at 39.30C) None 2.8 LL479/532 UC 82 Tyr - His 0.2 None 2.8 LL482/529 AC 25 Glu - Gly 0.3 Tyr - Cys 0 None 4.0 LLA 483 23 Thr - Ile 0.9 None 2.5 LLA 483/528 48 Gly - Asp 0.2 None 3.0 LL514A 10 Tyr - Cys 0.4 19 Tyr - Cys 1.5 None 1.5 LL514C 65 Glu -Val 0.2 I I Table 11

Temperature sensitivity phenotypes of parent, revertant and in vitro reconstructed strains Parent Strain of Loglo [pfu@35'/ Revertant 1, Loglo [pfu@35'/ In vitro recon- Loglo [pfu@35'/ revertent pfu@39-] strain pfu@39'] structed strain pfu@39'] P2/117 (484G) 0.2 PS/I 175' (484G) 0.3 P2/Sabin (484A) 2.2 96 His - Tyr 0.6 S2/2A-l 0.5 P I 17/S5' (484A) 2.3 19 Tyr - His 0.4 SMA-2 0.3 P I 17/S5' (484A) 2.3 8 Ala - Val 0.1 SMA-3 0.1, contains the sarne changes in the 2A gene as the revertant, but is based on P2/Sabin.

Revertants of the temperature-sensitive viruses were selected by picking plaques fanned at 39'C (or in one instance 39.3'C) on BGM cells and their phenotypes are shown in Table 1. It can be seen that all were markedly less temperature-sensitive (ts) in their growth than the viruses from which they derived.

All viruses grew to similar titres and with similar plaque morphology at 35'C, except for parental viruses with deletions, which had smaller plaques at this temperature.

Genetic basis of reversion The sequences of the entire 5'-NCRs of the genornes of revertants, here named P2/Sabin 96 His-Tyr, P117/S5' 19 Tyr His and P117/S5' 8 Ala - Val, were determined and found to be identical to the parental viruses. The sequence of the region of the genome encompassing domain V was determined for the other revertant viruses shown in Table I and also shown to be identical to that of the parent from which they derived. Further sequencing of the PI and P2 genes of P2/Sabin 96 His - Tyr revealed a mutation in the region of the genome encoding 2A and subsequently mutations were found in the 2A genes of all the revertant viruses studied here, as shown in Table 1. Mutations were identified by sequencing of genomic RNA or PCR products through the entire 2A gene.

Experiments in which the viral proteins were labelled with 35S methionine and the lysates of infected cells analysed by SDS-PAGE led to the conclusion that the 2A mutations present in P2/Sabin 96 His - Tyr and other revertant viruses apparently suppressed the temperature-sensitive defect by enhancing translational efficiency at elevated temperatures. These experiments have been described in more detail by A. J. Macadam et al., (1994) cited above.

Constructions in vitro of mutant strains These were done by replacing the BstER-SnaBl region (2892-4455) of the P2/Sabin clone with BstEH-SnaBI-digested PCR fragments from the same region of DNA generated from the viral RNA of the revertants. The 2A encoding regions of revertants P2/Sabin 96 His - Tyr, P I 17/SS' 19 Tyr - His and P I 17/S5' 8 Ala - Val were amplified by PCR and inserted into the infectious cDNA clone of P2/Sabin. The resulting clones were designated S2/2A-1, -2 or -3. PCR-derived regions of these - I I - clones were sequenced in full and found to be identical to the parental virus (P2/Sabin or PI 17/S5') except at the previously identified position in the 2A gene. The cDNA was transcribed to RNA using polymerase. Viruses were recovered by transfection of the T7 transcripts and the genomic RNA was sequenced to verify the mutant genotypes. As shown in Table II, the recovered viruses with reconstructed genotypes (S/2A-1, -2 and -3) were identical in phenotype to the revertant viruses from which they derived their 2A genes, thus demonstrating that the mutations in the 2A region were responsible for the suppression of the temperature-sensitive phenotype attributable to the mutations in the 5'-NCR.

Neurovirulence Neurovirulence tests were carried out as described in WHO Technical Report N' 800, 46-48 (1990). Thus, monkeys weighing at least 1.5 kg were inoculated with the virus at a dose of 10'-'-10'-' TCID5o/0.1 ml into the lumbar region of the central nervous system. Monkeys were observed for 17-22 days for symptoms suggestive of poliovirus infection. The monkeys which died were autopsied, histologically examined and the lesions scored as follows:

1. Cellular infiltration only (this is not sufficient for the monkey to be considered as positive).

2. Cellular infiltration with minimal neuronal damage.

3. Cellular infiltration with extensive neuronal damage.

4. Massive neuronal damage with or without cellular infiltration.

Severity scores were based on hernisection readings of the lumbar (L), cervical (C) and brain (B) histological sections. The Lesion Score (LS) for each positive monkey was calculated as follows:

sum of the L scores [ sum of the C scores I F sum of the B scores LS=[ No. of hemi sections No. of hemi sec tions] ' [ No. of hemi sec tions 3 A mean Lesion Score was calculated for each group of positive monkeys.

Nucleotide 484A of P2/Sabin (481A in poliovirus type 2 numbering) is the major determinant of attenuation and temperature-sensitivity in this virus, A. J.

Macadam et al., (1991) cited above. An A to G mutation at this nucleotide 484 results in reversion of temperature sensitivity and the attenuated character of the virus, i.e. it reverts to virulence. Since substitutions in the 2A protease gene were found to suppress the temperature sensitivity of phenotype due to the 484A, their effect on attenuation was tested. Remarkably, neurovirulence assays showed S2/2A-1 and S2/2A-2 to be indistinguishable from P2/Sabin in terms of clinical signs and mean histological lesion scores (Table 111), despite having lost the ts; phenotype as a result of their substitutions in the 2A gene. In contrast, inoculation with PS/1 175', which has a G at nucleotide 484, resulted in paralysis of three out of four animals and histological lesion scores that were significantly higher than those obtained with the Sabin 2 vaccine strain or the 2A gene mutants.

Table M

Neurovirulence of Sabin 2 and non-temperature-sensitive mutants EXAMPLE2

This Example describes mutagenesis of P2/Sabin at the codons for amino acid residues 79 and 80 of the protease 2A gene.

A clone of Sabin 2 was made in which the ApaI site at position 3522 was removed by the introduction of a silent base change. The modification had no effect on the virus phenotype. This means that the now unique ApaI site at 3617 and the SnaBI site at 4455 of Sabin 2 could be used to replace the nucleic acid between these sites. Part of the 2A gene of Sabin 2 was amplified using primers with degenerate Virus Clinical Mean Lesion Score Phenotype Score (range) P2/Sabin 0/4 0.21 (0.14-0.27) attenuated/ts (484A) PS/I 175' 3/4 1.25 (0.93-1.83) virulent/non-ts (484G) S2/2A-1 0/4 0.22 (0.12-0.37 attenuated/non-ts (96 His - Tyr) S2/2A-2 0/4 0.34 (0.23-0.51) attenuated/non-ts (19 Tyr - His) bases spanning the 79h and 80ffi codons of 2A such that any amino acid could be introduced at position 79 from the primer sequence NNC/G or four different amino acids introduced at position 80 from the primer sequence NTC (where N is any base). These mixed population PCR products were digested and ligated into the ApaI and SnaBI unique sites of the Sabin 2 clone. RNA transcripts were made from the selected clones and transfected into Hep-2C cells (grown as described by P. D. Minor et aL (1980) cited above.

Table IV

VIRUS (80' 2A codon) Logio [pfu at 390C/pfu at OC] Sabin 2 (UUC) 2.3 Sabin 2/484G (UUC) 0.1 Phe - Val 0.9 Phe - Leu 0.5 Phe - Ile 0.6 Table V

VIRUS (79m 2A codon) Log I o [pfu at 39'C/pfu at 'C] Sabin 2 (ACC) 2.9 Sabin 2/484G (ACC) 0.4 79 Thr - Glu 1.7 79 Thr - Arg 0.6 79 Thr - Trp 0.7 79 Thr - Leu 0.4 79 Thr - His 0.6 79 Silent (ACG) 2.4 79 Thr - Ala (a) (GCG) 0.9 79 Thr - Ala (b) (GCQ 0.6 79 Thr - Ser (a) (UGQ 1.0 79 Thr - Ser (b) (AGQ 1.0 48 1 G in Sabin 2 numbering A change of threonine to alanine at residue 79 was found in non-ts- revertants of Sabin 2 and LL473/536 UG. A change of phenylalanine to leucine at residue 80 was found in a non-ts-revertant of LLA472. Viruses with changes of phenylalanine to valine, leucine and isoleucine at residue 80 were constructed. Ten viruses were also made with eight different amino acid changes at residue 79. Of these, S2/79 Silent had a different codon for theonine to that found in Sabin 2. S2/79 Thr - Ala (a) and Thr - Ala (b) had non synonymous codons for alanine, as did S2/Thr - Ser (a) and Thr - Ser (b) for serine. Viruses were also obtained with either glutamic acid arginine, tryptophan, leucine or histidine at residue 79.

All viruses were viable and temperature sensitivities were determined in BGM cells as a measure of the effect of 2A change. Sabin 2 and Sabin 2(484G) were included as control viruses for each separate assay. Results (Tables IV and V) indicated that all amino acid changes gave viruses less temperature sensitive than Sabin 2. Of the viruses with changes at residue80 of 2A, S2/80 Phe - Val was the most temperature sensitive, the other viruses, S2/80 Phe - Leu and S2/80 Phe - Ile were less so. Similarly, all the viruses with changes at residue 79 were less temperature sensitive than Sabin 2 with the exception of S2/79 Silent which was comparable to Sabin 2. S2/79 Silent and Sabin 2 are identical except that the 79 1h threonine codons are different. In addition the viruses with nonsynonymous codons for alanine [S2/79 Thr - Ala (a) and 79 Thr - Ala (b)] and serine [S2/79 Thr - Ser (a) and (b)] at residue 79 were also comparable to each other. Of the other viruses, S2/79 Thr - Glu was the most temperature sensitive and S2/79 Thr - Leu had a similar phenotype to Sabin 2 484G (481G in Sabin 2 numbering). Consistency between phenotypes of viruses with non synonymous codons confirms the role of amino acid rather than RNA sequence in the compensation of 5'-NCR disruptions by changes in the 2A gene. There is no obvious link between the type of amino acid and the effect on phenotype.

All subject matter described above as contained in a prior reference is herein incorporated by reference to the extent relied on herein.

Claims

1. A vaccine comprising a poliovirus having a mutation in the protease 2A gene, other than one which changes amino acid 20 (His), 38 (Asp) or 109 (Cys), said mutation being associated with a reduced temperature sensitivity as defined by a decrease in the ratio of the number of plaque-forming units when the virus is grown at 35'C to plaque-forming units when the virus is grown at 39'C in BGM cells.

2. A vaccine according to Claim 1, wherein the logarithmic growth ratio in BGM cells:

log, [p.f. u. of the virus grown at 3 5'C] 0 Vf u. of the virus grown at 3 9"Cl is less than 1.0.

3. A vaccine according to Claim 2, wherein the logarithmic growth ratio is 0.01 to 0.7.

4. A vaccine according to Claim 2, wherein the mutation is any one or more of the following amino acid changes, shown in the form [aa number][amino acid of parent]-[amino acid of mutation]:

8 Ala - Val 65 Glu - Val Tyr - Cys 70 Tyr - Cys 17 Cys - Tyr 79 Thr - Ala 19 Tyr - Cys 80 Phe - Leu 19 Tyr - His 82 Tyr - His 23 Thr - Ile 93 Tyr His Glu - Gly 96 His - Tyr Ala - Pro 106 Pro - Ser 33 Ile - Val 122 Ile -Val Lys - Glu 134 Ser - Thr 48 Gly - Asp Glu - Lys

5. A vaccine according to any preceding Claim, wherein the poliovirus is inactivated.

6. A vaccine according to Claim 1, 2, 3 or 4, wherein the poliovirus (before mutation in the protease 2A gene) is of the Sabin 1, 2 or 3 strain.

7. A vaccine according to Claim 1, 2, 3, 4 or 6, wherein the poliovirus is live.

8. A vaccine according to Claim 5, which ftirther comprises a carrier.

9. A poliovirus defined as in Claim 1, 2, 3, 4 or 6, for use in a vaccine against wild type poliovirus.

10. Use of a poliovirus defined in Claim 1, 2, 3, 4 or 6, in the preparation of a vaccine against wild type poliovirus.