Attenuated temperature-sensitive respiratory syncytial virus mutants generated by cold adaptation.

Two strains of respiratory syncytial virus (RSV), RSV 2B and RSV 3A (representing subgroup B and A virus respectively) were cold-adapted by passaging in Vero cells for up to 42 weeks at successively lower temperatures down to 20 degrees C. Successful cold adaptation of the virus population was dependent on the amount of time the cultures were maintained at the various low temperatures, as well as on the strain of virus used. Temperature-sensitive (TS) mutants appeared in the cold passaged virus populations; however, the majority of the virus variants remained predominantly non-TS. Four RSV 2B and three RSV 3A TS mutants were selected for further characterization. These seven TS mutants retained their fusion phenotype and two major neutralizing antibody epitopes, and displayed varying levels of temperature sensitivity. Six of the seven mutants had a cold-adapted (CA) phenotype. All of the RSV 2B mutants were highly attenuated in cotton rats and two of the mutants elicited relatively high levels of neutralizing antibody and were able to protect rats against virus challenge. The RSV 3A TS mutants grew well in the nose but poorly in the cotton rat lungs, as did the parental 3A virus. All 3A mutants elicited high titers of neutralizing antibody and provided complete protection against virus challenge. These mutants showed varying levels of temperature sensitivity in vitro and attenuation in vivo and represent potential vaccine candidates.

Identification and characterization of a new base substitution in the vaccine strain of Sabin 3 poliovirus.

The complete RNA sequence of Sabin 3 (LED3) used in vaccine in the United States has been determined. The LED3 Sabin 3 sequence contains the attenuating mutations at bases 472 and 2034 but differs from that published by Stanway et al. (Nucleic Acids Res., 11, 5629-5643, 1983) at two other base positions, 2493 and 6061. The change at base 6061 is silent and does not affect amino acid composition. The other base, a C at position 2493, is contained in the viral capsid protein VP1 and predicts a new Sabin 3 specific amino acid change of a threonine instead of an isoleucine at amino acid 6 of the protein [corrected]. Reversion of this base to that present in the pathogenic progenitor strain, Leon, is observed to occur after replication of vaccine virus in the gut of primary vaccines and in nervous tissue of neurovirulence test monkeys. Passage conditions have been identified that lead to the reversion of base 2493 as well as the reversion of the attenuated base to the parental base (Leon) at position 472 in the 5' noncoding region. The observation that these two bases delta position are found to revert during passage suggests that there is a selective advantage for virus containing the parental bases at these positions.

Attenuated poliovirus strain as a live vector: expression of regions of rotavirus outer capsid protein VP7 by using recombinant Sabin 3 viruses.

The ability to express heterologous antigens from attenuated poliovirus strains suggests the potential for use as live vectored vaccines. Full- or partial-length sequences of the gene encoding rotavirus major outer capsid protein VP7 were cloned into the open reading frame of a full-length cDNA copy of poliovirus Sabin type 3. They were inserted either at the 5' end or immediately after the capsid protein coding region, at the junction between precursors P1 and P2. A protease cleavage site for 3C protease was introduced 3' to the foreign sequences to enable proteolytic processing of the antigen from the poliovirus polyprotein. Infectious viruses were generated from several of the DNA constructs, and the presence of the foreign gene sequences was confirmed by reverse transcription of the viral RNA and PCR amplification. Viruses with inserts of about 300 bases maintained the foreign sequences during passage in Vero cells. Viruses carrying larger sequences were unstable, and deletions were generated within the foreign sequences. Expression of the VP7 polypeptides was demonstrated by immunoprecipitation with specific antiserum of labeled proteins from cells infected with Sabin 3 recombinant viruses. Comparative studies of RNA synthesis showed similar kinetics for Sabin 3 and the Sabin 3/VP7 recombinants. One-step growth curves showed that production of recombinant viruses was slower than that of Sabin 3 and that the final titers were 1 to 1.5 logs lower. Accumulation of VP7-containing precursors in infected cells suggests that slow cleavage at the engineered 3C protease site may be a limiting step in the growth of these recombinant Sabin polioviruses and may influence the permissible size of foreign sequence to be inserted.

Significance of a newly identified attenuating mutation in Sabin 3 oral poliovirus vaccine.

Sequence analysis of poliovirus vaccine RNA has resulted in the identification of a new point mutation at position 2493 in Sabin 3. cDNA-derived Sabin 3 virus whose genome includes the new mutation has phenotypic properties expected for a vaccine strain. Virus engineered to contain a single base substitution at 2493 has lost the small plaque phenotype, and low neurovirulence characteristics normally associated with Sabin poliovirus strains. The significance of these observations and a potential relationship to the frequency of vaccine-associated poliomyelitis are described.

A mutation present in the amino terminus of Sabin 3 poliovirus VP1 protein is attenuating.

The attenuated phenotype of Sabin 3 poliovirus compared with its neurovirulent progenitor strain has been largely accounted for by mutations in the genome at positions 472 and 2034 (G. D. Westrop, K. A. Wareham, D. M. A. Evans, G. Dunn, P. D. Minor, D. I. Magrath, F. Taffs, S. Marsden, M. A. Skinner, G. C. Schild, and J. W. Almond, J. Virol. 63:1338-1344, 1989). By sequencing vaccine virus RNA, we recently identified another Sabin 3-specific mutation at position 2493 (U----C), which predicts an Ile----Thr change at the sixth residue of VP1 (C. Weeks-Levy, J. M. Tatem, S. J. DiMichele, W. Waterfield, A. F. Georgiu, and S. J. Mento, Virology 185:934-937, 1991). Viruses generated by using cDNAs which represent the vaccine sequence (LED3) and a derivative (VR318) possessing a single base change to the wild-type nucleotide (U) at 2493 were used to determine the impact of the 2493 mutation on virus phenotype. The VP1 proteins of LED3 and VR318 viruses were distinguishable by denaturing electrophoretic analysis. LED3 produced smaller plaques in Vero cells than VR318 virus did. Neurovirulence testing of these cDNA-derived viruses in monkeys demonstrated that the 2493 mutation in LED3 virus is attenuating.

Characterization of Vero cells.

A cell line used in the production of biologicals should be free of infectious agents, and 'described with respect to cytogenetic characteristics and tumorigenicity'. Vero, a continuous cell line derived from a normal African green monkey kidney, was examined for the presence of retroviruses and for tumorigenic potential. We were unable to detect the presence of retroviruses by reverse transcriptase assay, electron microscopy or hybridization of cellular genomic DNA with Mason-Pfizer monkey virus DNA probes. In addition, passage 156 Vero cells did not form progressively growing tumors in nude mice or grow with high efficiency in soft agarose.

Oral poliovirus vaccine in the United States: molecular characterization of Sabin type 3 after replication in the gut of vaccinees.

Derivatives of Sabin 3 shed from recipients of oral poliovirus vaccine in the United States (U.S.) were examined for genetic changes identified in strains excreted by vaccinees in the United Kingdom [U.K.; Evans et al., 1985; Cammack et al., 1988, Macadam et al., 1989]. Among the eight primary vaccinees studied, the duration of excretion and molecular evolution of type 3 strains varied greatly. The period of virus excretion after vaccination ranged from as few as 2 days to as many as 36 days. Nucleotide sequence analysis of viral RNAs extracted from shed virus indicated that only fifty percent of the vaccinees exclusively excreted strains in which the attenuating mutation at nucleotide 472 in the 5' noncoding region of the genome had reverted from uracil (U) to cytosine (C), the nucleotide found in neurovirulent strains. Compared to the wild-type Leon strain, the low activity of stool isolate KW4 in a complete monkey neurovirulence test demonstrated that presence of C at 472 does not render a type 3 strain pathogenic. Conversely, an isolate was identified which efficiently replicated in monkey nervous tissue and maintained the attenuated U at 472. Oligonucleotide fingerprinting and sequence analysis of viral RNAs from stool isolates indicated that one vaccinee (KW) eventually excreted intertypic recombinant strains consistent with those reported in the U.K. studies. Unique to this study, one vaccinee (KS) excreted nonrecombinant virus possessing U at 472 for up to 21 days. The significance of the KS strain profile in relation to differences in the U.S. vaccine compared to the vaccine distributed in the U.K. and other countries is discussed.

Characterization of recombinant polioviruses expressing regions of rotavirus VP4, hepatitis B surface antigen, and herpes simplex virus type 2 glycoprotein D.

Recombinant polioviruses expressing antigens from rotavirus, herpes simplex virus type 2, and hepatitis B virus were generated. Fusion of the heterologous polypeptides to the amino terminus of the poliovirus polyprotein did not prevent myristylation of VP0, suggesting a novel mechanism of myristylation for these recombinant viruses. The effects of the parental genetic background, different foreign sequences, and different insert sizes on growth characteristics were compared. Both the size and the nature of the heterologous sequence appeared to be factors influencing the growth and stability of recombinant polioviruses. All of the recombinants showed a temperature-sensitive phenotype, regardless of the genetic background (attenuated or wild type) from which they were derived. Preliminary studies with transgenic mice carrying the poliovirus receptor gene are discussed.

Characterization of mouse lines transgenic with the human poliovirus receptor gene.

Two mouse lines transgenic with the human poliovirus receptor gene (PVR), TGM-PRG-1 and TGM-PRG-3, were characterized to determine whether transgene copy number and PVR expression levels influence susceptibility to poliovirus. The mouse lines have been bred for more than 10 generations and the transgene was stably transmitted to progeny as determined by Southern blot hybridization and restriction fragment length polymorphism. The transgene copy number is 10 in the TGM-PRG-3 mouse line and one in the TGM-PRG-1 mouse line. Abundance of PVR RNA is on average three-fold higher in TGM-PRG-3 relative to TGM-PRG-1 tissues, and the abundance of the receptor molecule is three-fold higher in TGM-PRG-3 central nervous system tissues compared to TGM-PRG-1 tissues as determined by Western blot analysis. When TGM-PRG-1 and TGM-PRG-3 mice were inoculated intracranially with a neurovirulent type III poliovirus strain, they developed clinical symptoms and CNS lesions characteristic of human poliomyelitis. These results indicate that the PVR gene is expressed as a functional receptor in the CNS of both mouse lines rendering the mice susceptible to poliovirus infection. Even though the two mouse lines have different copy numbers of the transgene and different levels of PVR RNA and protein, they are similar in their susceptibility to poliovirus.