Search

We modified the haemagglutinin-neuraminidase (HN) glycoprotein of the virulent Newcastle disease virus (NDV) strain Banjarmasin/010/10 (Ban/010) by adding C-terminal extensions similar to those found in certain avirulent NDV strains. Extension of the 571 aa wt Ban/010 HN protein to 577 and 616 aa by removal of one or two translational stop codons moderately reduced HN function and viral pathogenicity in 1-day-old and 3-week-old chickens. Substantially greater reductions were achieved by altering the 616 aa form by introducing a R596C mutation or by replacing the C-terminal extension with that of avirulent strain Ulster, which naturally contains the amino acid 596C. These results showed that extension of the C terminus of HN reduces NDV pathogenicity, and that this effect is substantially increased by the presence of 596C. These results indicate that this attenuating mechanism in avirulent strains such as Ulster can be applied directly to a highly virulent strain recently in circulation.

Virulent and moderately virulent strains of Newcastle disease virus (NDV), representing avian paramyxovirus serotype 1 (APMV-1), cause respiratory and neurological disease in chickens and other species of birds. In contrast, APMV-2 is avirulent in chickens. We investigated the role of the fusion (F) and hemagglutinin-neuraminidase (HN) envelope glycoproteins in these contrasting phenotypes by designing chimeric viruses in which the F and HN glycoproteins or their ectodomains were exchanged individually or together between the moderately virulent, neurotropic NDV strain Beaudette C (BC) and the avirulent APMV-2 strain Yucaipa. When we attempted to exchange the complete F and HN glycoproteins individually and together between the two viruses, the only construct that could be recovered was recombinant APMV-2 strain Yucaipa (rAPMV-2), containing the NDV F glycoprotein in place of its own. This substitution of NDV F into APMV-2 was sufficient to confer the neurotropic, neuroinvasive, and neurovirulent phenotypes, in spite of all being at reduced levels compared to what was seen for NDV-BC. When the ectodomains of F and HN were exchanged individually and together, two constructs could be recovered: NDV, containing both the F and HN ectodomains of APMV-2; and APMV-2, containing both ectodomains of NDV. This supported the idea that homologous cytoplasmic tails and matched F and HN ectodomains are important for virus replication. Analysis of these viruses for replication in vitro, syncytium formation, mean embryo death time, intracerebral pathogenicity index, and replication and tropism in 1-day-old chicks and 2-week-old chickens showed that the two contrasting phenotypes of NDV and APMV-2 could largely be transferred between the two backbones by transfer of homotypic F and HN ectodomains. Further analysis provided evidence that the homologous stalk domain of NDV HN is essential for virus replication, while the globular head domain of NDV HN could be replaced with that of APMV-2 with only a minimal attenuating effect. These results demonstrate that the F and HN ectodomains together determine the cell fusion, tropism, and virulence phenotypes of NDV and APMV-2 and that the regions of HN that are critical to replication and the species-specific phenotypes include the cytoplasmic tail and stalk domain but not the globular head domain.

Newcastle disease virus (NDV) belongs to serotype 1 of the avian paramyxoviruses (APMV-1) and causes severe disease in chickens. Current live attenuated NDV vaccines are not fully satisfactory. An alternative is to use a viral vector vaccine that infects chickens but does not cause disease. APMV serotype 3 infects a wide variety of avian species but does not cause any apparent disease in chickens. In this study, we constructed a reverse-genetics system for recovery of infectious APMV-3 strain Netherlands from cloned cDNAs. Two recombinant viruses, rAPMV3-F and rAPMV3-HN, were generated expressing the NDV fusion (F) and hemagglutinin-neuraminidase (HN) proteins, respectively, from added genes. These viruses were used to immunize 2-week-old chickens by the oculonasal route in order to evaluate the contribution of each protein to the induction of NDV-specific neutralizing antibodies and protective immunity. Each virus induced high titers of NDV-specific hemagglutination inhibition and serum neutralizing antibodies, but the response to F protein was greater. Protective immunity was evaluated by challenging the immunized birds 21 days later with virulent NDV via the oculonasal, intramuscular, or intravenous route. With oculonasal or intramuscular challenge, all three recombinant viruses (rAPMV3, rAPMV3-F, and rAPMV3-HN) were protective, while all unvaccinated birds succumbed to death. These results indicated that rAPMV3 alone can provide cross-protection against NDV challenge. However, with intravenous challenge, birds immunized with rAPMV3 were not protected, whereas birds immunized with rAPMV3-F alone or in combination with rAPMV3-HN were completely protected, and birds immunized with rAPMV3-HN alone were partially protected. These results indicate that the NDV F and HN proteins are independent neutralization and protective antigens, but the contribution by F is greater. rAMPV3 represents an avirulent vaccine vector that can be used against NDV and other poultry pathogens.

The hemagglutinin-neuraminidase (HN) protein of Newcastle disease virus (NDV) is a multifunctional protein that plays a crucial role in virus infectivity. In this study, using the mesogenic strain Beaudette C (BC), we mutated three conserved amino acids thought to be part of the binding/catalytic active site in the HN protein. We also mutated five additional residues near the proposed active site that are nonconserved between BC and the avirulent strain LaSota. The eight recovered NDV HN mutants were assessed for effects on biological activities. While most of the mutations had surprisingly little effect, mutation at conserved residue Y526 reduced the neuraminidase, receptor binding, and fusion activities and attenuated viral virulence in eggs and young birds.

Paramyxoviruses such as human parainfluenza viruses that bear inserts encoding protective antigens of heterologous viruses can induce an effective immunity against the heterologous viruses in experimental animals. However, vectors based on common human pathogens would be expected to be restricted in replication in the adult human population due to high seroprevalence, an effect that would reduce vector immunogenicity. To address this issue, we evaluated Newcastle disease virus (NDV), an avian paramyxovirus that is serotypically distinct from common human pathogens, as a vaccine vector. Two strains were evaluated: the attenuated vaccine strain LaSota (NDV-LS) that replicates mostly in the chicken respiratory tract and the Beaudette C (NDV-BC) strain of intermediate virulence that produces mild systemic infection in chickens. A recombinant version of each virus was modified by the insertion, between the P and M genes, of a gene cassette encoding the human parainfluenza virus type 3 (HPIV3) hemagglutinin-neuraminidase (HN) protein, a test antigen with considerable historic data. The recombinant viruses were administered to African green monkeys (NDV-BC and NDV-LS) and rhesus monkeys (NDV-BC only) by combined intranasal and intratracheal routes at a dose of 10(6.5) PFU per site, with a second equivalent dose administered 28 days later. Little or no virus shedding was detected in nose-throat swabs or tracheal lavages following immunization with either strain. In a separate experiment, direct examination of lung tissue confirmed a highly attenuated, restricted pattern of replication by parental NDV-BC. The serum antibody response to the foreign HN protein induced by the first immunization with either NDV vector was somewhat less than that observed following a wild-type HPIV3 infection; however, the titer following the second dose exceeded that observed with HPIV3 infection, even though HPIV3 replicates much more efficiently than NDV in these animals. NDV appears to be a promising vector for the development of vaccines for humans; one application would be in controlling localized outbreaks of emerging pathogens.

Human metapneumovirus (HMPV) has recently been identified as a significant cause of serious respiratory tract disease in humans. In particular, the emerging information on the contribution of HMPV to pediatric respiratory tract disease suggests that it will be important to develop a vaccine against this virus for use in conjunction with those being developed for human respiratory syncytial virus and the human parainfluenza viruses. A recently described reverse genetic system (S. Biacchesi, M. H. Skiadopoulos, K. C. Tran, B. R. Murphy, P. L. Collins, and U. J. Buchholz, Virology 321:247-259, 2004) was used to generate recombinant HMPVs (rHMPVs) that lack the G gene, the SH gene, or both. The DeltaSH, DeltaG, and DeltaSH/G deletion mutants were readily recovered and were found to replicate efficiently during multicycle growth in cell culture. Thus, the SH and G proteins are not essential for growth in cell culture. Apart from the absence of the deleted protein(s), the virions produced by the gene deletion mutants were similar by protein yield and gel electrophoresis protein profile to wild-type HMPV. When administered intranasally to hamsters, the DeltaG and DeltaSH/G mutants replicated in both the upper and lower respiratory tracts, showing that HMPV containing F as the sole viral surface protein is competent for replication in vivo. However, both viruses were at least 40-fold and 600-fold restricted in replication in the lower and upper respiratory tract, respectively, compared to wild-type rHMPV. They also induced high titers of HMPV-neutralizing serum antibodies and conferred complete protection against replication of wild-type HMPV challenge virus in the lungs. Surprisingly, G is dispensable for protection, and the DeltaG and DeltaSH/G viruses represent promising vaccine candidates. In contrast, DeltaSH replicated somewhat more efficiently in hamster lungs compared to wild-type rHMPV (20-fold increase on day 5 postinfection). This indicates that SH is completely dispensable in vivo and that its deletion does not confer an attenuating effect, at least in this rodent model.

The G glycoprotein of human respiratory syncytial virus (RSV) was identified previously as the viral attachment protein. Although we and others recently showed that G is not essential for replication in vitro, it does affect the efficiency of replication in a cell type-dependent fashion and is required for efficient replication in vivo. The ectodomain of G is composed of two heavily glycosylated domains with mucin-like characteristics that are separated by a short central region that is relatively devoid of glycosylation sites. This central region contains a 13-amino acid segment that is conserved in the same form among RSV isolates and is overlapped by a second segment containing four cysteine residues whose spacings are conserved in the same form and which create a cystine noose. The conserved nature of the cystine noose and flanking 13-amino acid segment suggested that this region likely was important for attachment activity. To test this hypothesis, we constructed recombinant RSVs from which the region containing the cysteine residues was deleted together with part or all of the conserved 13-amino acid segment. Surprisingly, each deletion had little or no effect on the intracellular synthesis and processing of the G protein, the kinetics or efficiency of virus replication in vitro, or sensitivity to neutralization by soluble heparin in vitro. In addition, neither deletion had any discernible effect on the ability of RSV to infect the upper respiratory tract of mice and both resulted in a 3- to 10-fold reduction in the lower respiratory tract. Thus, although the G protein is necessary for efficient virus replication in vivo, this activity does not require the central conserved cystine noose region.

The surface glycoproteins of viruses can play important roles in viral attachment, entry, and morphogenesis. Here, we investigated the role of the attachment G glycoprotein of human respiratory syncytial virus (RSV) in viral infection. RSV G is produced both as a complete, transmembrane form and as an N-terminally truncated form that is secreted. Using reverse genetics, we created mutant recombinant RSVs (rRSV) that do not express G (DeltaG) or express either the secreted or the membrane-bound form of G only (sG and mG, respectively). In Vero cells, the DeltaG virus formed plaques and grew as efficiently as wild-type rRSV and mG. In contrast, DeltaG replicated less efficiently and did not form distinct plaques in HEp-2 cells. This defect was primarily at the level of the initiation of infection, with only a minor additional effect at the level of packaging. Replication of DeltaG in the respiratory tract of mice was very highly restricted, indicating that G is important in vivo. Although the G protein expressed by the sG virus was confirmed to be secreted, this virus grew at least as efficiently as wild-type in HEp-2 cells and was only moderately attenuated in vivo. Thus, the G protein was important for efficient replication in HEp-2 cells and in vivo, but this function could be supplied in large part by the secreted form and thus does not require the cytoplasmic and transmembrane domains. Amino acids 184-198 have been identified as the major heparin-binding domain of the G protein and were implicated in mediating binding to cells [S. A. Feldman et al., 1999, J. Virol. 73, 6610-6617]. Heparin-like glycosaminoglycans also appeared to be important for infection in vitro by direct clinical isolates of RSV. Deletion of amino acids 187-197 from rRSV did not reduce its sensitivity to neutralization in vitro by incubation with soluble heparin, did not reduce its efficiency of growth in vitro, and resulted in only a modest reduction in vivo. Thus, the putative heparin-binding domain is not the sole determinant of heparin sensitivity and is not a critical functional domain.

This study examines the contribution of the fusion (F) and hemagglutinin-neuraminidase (HN) glycoprotein genes of bovine parainfluenza virus type 3 (BPIV3) to its restricted replication in the respiratory tract of nonhuman primates. A chimeric recombinant human parainfluenza type 3 virus (HPIV3) containing BPIV3 F and HN glycoprotein genes in place of its own and the reciprocal recombinant consisting of BPIV3 bearing the HPIV3 F and HN genes (rBPIV3-F(H)HN(H)) were generated to assess the effect of glycoprotein substitution on replication of HPIV3 and BPIV3 in the upper and lower respiratory tract of rhesus monkeys. The chimeric viruses were readily recovered and replicated in simian LLC-MK2 cells to a level comparable to that of their parental viruses, suggesting that the heterologous glycoproteins were compatible with the PIV3 internal proteins. HPIV3 bearing the BPIV3 F and HN genes was restricted in replication in rhesus monkeys to a level similar to that of its BPIV3 parent virus, indicating that the glycoprotein genes of BPIV3 are major determinants of its host range restriction of replication in rhesus monkeys. rBPIV3-F(H)HN(H) replicated in rhesus monkeys to a level intermediate between that of HPIV3 and BPIV3. This observation indicates that the F and HN genes make a significant contribution to the overall attenuation of BPIV3 for rhesus monkeys. Furthermore, it shows that BPIV3 sequences outside the F and HN region also contribute to the attenuation phenotype in primates, a finding consistent with the previous demonstration that the nucleoprotein coding sequence of BPIV3 is a determinant of its attenuation for primates. Despite its restricted replication in the respiratory tract of rhesus monkeys, rBPIV3-F(H)HN(H) conferred a level of protection against challenge with HPIV3 that was indistinguishable from that induced by previous infection with wild-type HPIV3. The usefulness of rBPIV3-F(H)HN(H) as a vaccine candidate against HPIV3 and as a vector for other viral antigens is discussed.

Mutant recombinant respiratory syncytial viruses (RSV) which cannot express the NS1 and M2-2 proteins, designated rA2DeltaNS1 and rA2DeltaM2-2, respectively, were evaluated as live-attenuated RSV vaccines. The rA2DeltaNS1 virus contains a large deletion that should have the advantageous property of genetic stability during replication in vitro and in vivo. In vitro, rA2DeltaNS1 replicated approximately 10-fold less well than wild-type recombinant RSV (rA2), while rA2DeltaM2-2 had delayed growth kinetics but reached a final titer similar to that of rA2. Each virus was administered to the respiratory tracts of RSV-seronegative chimpanzees to assess replication, immunogenicity, and protective efficacy. The rA2DeltaNS1 and rA2DeltaM2-2 viruses were 2,200- to 55,000-fold restricted in replication in the upper and lower respiratory tracts but induced a level of RSV-neutralizing antibody in serum that was only slightly reduced compared to the level induced by wild-type RSV. The replication of wild-type RSV in immunized chimpanzees after challenge was reduced more than 10,000-fold at each site. Importantly, rA2DeltaNS1 and rA2DeltaM2-2 were 10-fold more restricted in replication in the upper respiratory tract than was the cpts248/404 virus, a vaccine candidate that retained mild reactogenicity in the upper respiratory tracts of 1-month-old infants. Thus, either rA2DeltaNS1 or rA2DeltaM2-2 might be appropriately attenuated for this age group, which is the major target population for an RSV vaccine. In addition, these results show that neither NS1 nor M2-2 is essential for RSV replication in vivo, although each is important for efficient replication.

We sought to develop a live attenuated parainfluenza virus type 2 (PIV2) vaccine strain for use in infants and young children, using reverse genetic techniques that previously were used to rapidly produce a live attenuated PIV1 vaccine candidate. The PIV1 vaccine candidate, designated rPIV3-1cp45, was generated by substituting the full-length HN and F proteins of PIV1 for those of PIV3 in the attenuated cp45 PIV3 vaccine candidate (T. Tao et al., J. Virol. 72:2955-2961, 1998; M. H. Skiadopoulos et al., Vaccine 18:503-510, 1999). However, using the same strategy, we failed to recover recombinant chimeric PIV3-PIV2 isolate carrying the full-length PIV2 glycoproteins in a wild-type PIV3 backbone. Viable PIV3-PIV2 chimeras were recovered when chimeric HN and F open reading frames (ORFs) rather than complete PIV2 F and HN ORFs were used to construct the full-length cDNA. The recovered viruses, designated rPIV3-2CT, in which the PIV2 ectodomain and transmembrane domain were fused to the PIV3 cytoplasmic domain, and rPIV3-2TM, in which the PIV2 ectodomain was fused to the PIV3 transmembrane and cytoplasmic tail domain, possessed similar in vitro and in vivo phenotypes. Thus, it appeared that only the cytoplasmic tail of the HN or F glycoprotein of PIV3 was required for successful recovery of PIV3-PIV2 chimeras. Although rPIV3-2CT and rPIV3-2TM replicated efficiently in vitro, they were moderately to highly attenuated for replication in the respiratory tracts of hamsters, African green monkeys (AGMs), and chimpanzees. This unexpected finding indicated that chimerization of the HN and F proteins of PIV2 and PIV3 itself specified an attenuation phenotype in vivo. Despite this attenuation, these viruses were highly immunogenic and protective against challenge with wild-type PIV2 in hamsters and AGMs, and they represent promising candidates for clinical evaluation as a vaccine against PIV2. These chimeric viruses were further attenuated by the addition of 12 mutations of PIV3cp45 which lie outside of the HN and F genes. The attenuating effects of these mutations were additive with that of the chimerization, and thus inclusion of all or some of the cp45 mutations provides a means to further attenuate the PIV3-PIV2 chimeric vaccine candidates if necessary.

Recombinant parainfluenza virus 3 (rPIV3) is being developed as a vector to express foreign genes as a bivalent or multivalent live attenuated virus vaccine. In the present study, we examined the effect of inserted foreign sequence on virus replication in vitro and in vivo, focusing on the parameter of insert length. In one type of construct, foreign sequence of increasing length was flanked by PIV3 transcription signals and inserted as an additional gene unit (GU insert) between the HN and L genes, so that one additional mRNA would be made. In a second type of construct, foreign sequence was inserted into the downstream NCR (NCR insert) of the HN gene, so that the number of encoded mRNAs remained unchanged. In each case, the foreign sequence was designed to lack any significant open reading frame, which permitted an evaluation of the effect of insert length on replication independent of an effect of an expressed protein. The GU or NCR insert sizes ranged from 168 nucleotides (nt) to 3918 nt. rPIV3s containing GU insertions of up to 3918 nt in length, the largest size tested, were viable and replicated efficiently at permissive temperatures in vitro, but a reduction in plaque size was seen at 39 degrees C and 40 degrees C. The rPIV3 with a 3918-nt GU insertion was restricted in replication in the upper (fivefold) and lower (25-fold) respiratory tracts of hamsters. Although a 1908-nt GU insertion did not significantly modify replication of wild-type PIV3 in vitro or in vivo, its introduction significantly augmented the level of temperature sensitivity (ts) and attenuation (att) specified by three mutations in the L protein of a cold-passaged attenuated PIV3 vaccine virus. rPIV3s bearing a 3126- or 3894-nt NCR insertion exhibited in vitro and in vivo phenotypes like those of the rPIV3s bearing similar-sized GU insertions. These findings indicate that rPIV3s whose genome length has been increased by more than 3000 nt by either a GU or an NCR insertion exhibit an unexpected host-range phenotype, that is, efficient replication in vitro but restricted replication in hamsters, especially in the lower respiratory tract. Furthermore, these effects were greatly enhanced when the rPIV3 backbone contained other ts or att mutations. The implications of these findings for the use of single-stranded, negative-sense RNA viruses as vectors for vaccines are discussed., (Copyright 2000 Academic Press.)

The nucleocapsid (N) protein of bovine respiratory syncytial virus (BRSV) is a multifunctional protein that plays a central role in transcription and replication of viral genomic RNA. To investigate the domains and specific residues involved in different N activities, we generated a total of 27 deletion and 12 point mutants of the N protein. These mutants were characterized using an intracellular BRSV-CAT minigenome replication system for the ability to (1) direct minigenome RNA synthesis, (2) direct minigenome encapsidation, and (3) form a complex with the phosphoprotein (P). The mutations tested were defective in synthesis of RNA from the BRSV-CAT minigenome template with the exception of the following: a deletion involving the first N-terminal amino acid and mutations involving conservative substitution at the second amino acid and at certain internal cysteine residues. Micrococcal nuclease enzyme protection assays showed that mutations involving amino acids 1-364 of the 391-amino-acid N protein prevented minigenome encapsidation. Thus the BRSV N protein has a C-terminal, 27-amino-acid tail that is not required for encapsidation. Interestingly, two of the mutations that ablated encapsidation did not greatly affect RNA synthesis; the mutant involving deletion of the N-terminal amino acid and the mutant involving a substitution at position 2. This finding indicates that the formation of a nucleocapsid sufficient to protect the RNA from nuclease is not required for template function. Coimmunoprecipitation of N and P using N- or P-specific antiserum revealed two regions of the N protein that are important for association with the P protein: a central portion of 244-290 amino acids and a C-terminal portion of 338-364 amino acids., (Copyright 2000 Academic Press.)

We recently developed a system for the generation of infectious bovine respiratory syncytial virus (BRSV) from cDNA. Here, we report the recovery of fully viable chimeric recombinant BRSVs (rBRSVs) that carry human respiratory syncytial virus (HRSV) glycoproteins in place of their BRSV counterparts, thus combining the replication machinery of BRSV with the major antigenic determinants of HRSV. A cDNA encoding the BRSV antigenome was modified so that the complete G and F genes, including the gene start and gene end signals, were replaced by their HRSV A2 counterparts. Alternatively, the BRSV F gene alone was replaced by that of HRSV Long. Each antigenomic cDNA directed the successful recovery of recombinant virus, yielding rBRSV/A2 and rBRSV/LongF, respectively. The HRSV G and F proteins or the HRSV F in combination with BRSV G were expressed efficiently in cells infected with the appropriate chimeric virus and were efficiently incorporated into recombinant virions. Whereas BRSV and HRSV grew more efficiently in bovine and human cells, respectively, the chimeric rBRSV/A2 exhibited intermediate growth characteristics in a human cell line and grew better than either parent in a bovine line. The cytopathology induced by the chimera more closely resembled that of BRSV. BRSV was confirmed to be highly restricted for replication in the respiratory tract of chimpanzees, a host that is highly permissive for HRSV. Interestingly, the rBRSV/A2 chimeric virus was somewhat more competent than BRSV for replication in chimpanzees but remained highly restricted compared to HRSV. This showed that the substitution of the G and F glycoproteins alone was not sufficient to induce efficient replication in chimpanzees. Thus, the F and G proteins contribute to the host range restriction of BRSV but are not the major determinants of this phenotype. Although rBRSV/A2 expresses the major neutralization and protective antigens of HRSV, chimpanzees infected with this chimeric virus were not significantly protected against subsequent challenge with wild-type HRSV. This suggests that the growth restriction of rBRSV/A2 was too great to provide adequate antigen expression and that the capacity of this chimeric vaccine candidate for replication in primates will need to be increased by the importation of additional HRSV genes.

Human respiratory syncytial virus (RSV) exists as two antigenic subgroups, A and B, both of which should be represented in a vaccine. The F and G glycoproteins are the major neutralization and protective antigens, and the G protein in particular is highly divergent between the subgroups. The existing system for reverse genetics is based on the A2 strain of RSV subgroup A, and most efforts to develop a live attenuated RSV vaccine have focused on strain A2 or other subgroup A viruses. In the present study, the development of a live attenuated subgroup B component was expedited by the replacement of the F and G glycoproteins of recombinant A2 virus with their counterparts from the RSV subgroup B strain B1. This gene replacement was initially done for wild-type (wt) recombinant A2 virus to create a wt AB chimeric virus and then for a series of A2 derivatives which contain various combinations of A2-derived attenuating mutations located in genes other than F and G. The wt AB virus replicated in cell culture with an efficiency which was comparable to that of the wt A2 and B1 parents. AB viruses containing temperature-sensitive mutations in the A2 background exhibited levels of temperature sensitivity in vitro which were similar to those of A2 viruses bearing the same mutations. In chimpanzees, the replication of the wt AB chimera was intermediate between that of the A2 and B1 wt viruses and was accompanied by moderate rhinorrhea, as previously seen in this species. An AB chimeric virus, rABcp248/404/1030, which was constructed to contain a mixture of attenuating mutations derived from two different biologically attenuated A2 viruses, was highly attenuated in both the upper and lower respiratory tracts of chimpanzees. This attenuated AB chimeric virus was immunogenic and conferred a high level of resistance on chimpanzees to challenge with wt AB virus. The rABcp248/404/1030 chimeric virus is a promising vaccine candidate for RSV subgroup B and will be evaluated next in humans. Furthermore, these results suggest that additional attenuating mutations derived from strain A2 can be inserted into the A2 background of the recombinant chimeric AB virus as necessary to modify the attenuation phenotype in a reasonably predictable manner to achieve an optimal balance between attenuation and immunogenicity in a virus bearing the subgroup B antigenic determinants.

Parainfluenza virus type 1 (PIV1) is a major cause of croup in infants and young children, and a vaccine is needed to prevent the serious disease caused by this virus. In the present study, a live attenuated PIV1 vaccine candidate was generated by modification of the extensively-studied PIV3 cold-passaged (cp) cp45 vaccine candidate using the techniques of reverse genetics. The HN and F glycoproteins of the PIV3 cp45 candidate vaccine virus were replaced with those of PIV1. This created a live attenuated PIV1 vaccine candidate, termed rPIV3-1 cp45, which contained the attenuated background of the PIV3 cp45 vaccine virus together with the HN and F protective antigens of PIV1. Three of the 15 mutations of cp45 lie within the HN and F genes, and those in the F gene are attenuating. Thus, some attenuation might be lost by the HN and F glycoprotein replacement. To address this issue we also constructed a derivative of PIV3 cp45, designated rPIV3 cp45 (F(wt)HN(wt)), that possessed wild type PIV3 HN and F glycoproteins but retained the 12 other cp45 mutations. rPIV3 cp45 (F(wt)HN(wt)) replicated in the respiratory tract of hamsters to a level three- to four-fold higher than rPIV3 cp45, indicating that loss of the two attenuating mutations in the cp45 F gene effected a slight reduction in the overall attenuation of cp45 for hamsters. However, the chimeric rPIV3-1 cp45 virus was about 5-fold more restricted in replication in hamsters than rPIV3 cp45 and about 15- to 20-fold more restricted than rPIV3 cp45 (F(wt)HN(wt)). This suggests that two components contribute to the attenuation of the new chimeric rPIV3-1 cp45 PIV1 vaccine candidate: one being the 12 cp45 mutations, which provide most of the observed attenuation, and the other resulting from the introduction of the heterologous PIV1 HN and F proteins into PIV3 (i.e., a chimerization effect). rPIV3-1 cp45 was observed to be immunogenic and protective against challenge with wild type PIV1 in hamsters. This virus shows sufficient promise that it should be evaluated further as a candidate live attenuated vaccine strain for preventing severe lower respiratory tract PIV1 disease in infants and young children.

The M2 mRNA of human respiratory syncytial virus (RSV) contains two overlapping ORFs, encoding the transcription antitermination protein (M2-1) and the 90-aa M2-2 protein of unknown function. Viable recombinant RSV was recovered in which expression of M2-2 was ablated, identifying it as an accessory factor dispensable for growth in vitro. Virus lacking M2-2 grew less efficiently than did the wild-type parent in vitro, with titers that were reduced 1, 000-fold during the initial 2-5 days and 10-fold by days 7-8. Compared with wild-type virus, the intracellular accumulation of RNA by M2-2 knockout virus was reduced 3- to 4-fold or more for genomic RNA and increased 2- to 4-fold or more for mRNA. Synthesis of the F and G glycoproteins, the major RSV neutralization and protective antigens, was increased in proportion with that of mRNA. In cells infected with wild-type RSV, mRNA accumulation increased dramatically up to approximately 12-15 hr after infection and then leveled off, whereas accumulation continued to increase in cells infected with the M2-2 knockout viruses. These findings suggest that M2-2 mediates a regulatory "switch" from transcription to RNA replication, one that provides an initial high level of mRNA synthesis followed by a shift in the RNA synthetic program in favor of genomic RNA for virion assembly. With regard to vaccine development, the M2-2 knockout has a highly desirable phenotype in which virus growth is attenuated while gene expression is concomitantly increased.

Respiratory syncytial virus (RSV) can be recovered from plasmids that separately encode antigenomic RNA and the N, P, L, and M2-1 proteins of the nucleocapsid. However, in a recent study the inclusion of a separate M2-1 expression plasmid was found to be unnecessary (H. Jin, D. Clarke, H. Zhou, X. Cheng, K. Coelingh, M. Bryant, and S. Li, Virology 1998, 251, 206-214). This suggested that the M2-1 protein, which is a transcription antitermination factor, is not required to reconstitute the minimum unit of infectivity, namely a nucleocapsid fully functional for viral transcription and RNA replication. Here we show that the antigenomic plasmid is remarkably efficient as a substitute for an M2-1 expression plasmid in supporting processive transcription by an RSV minigenome. Thus, the simple expedient of omitting an expression plasmid is invalid for evaluating recovery requirements. The issue of the requirement of M2-1 for the recovery of infectious RSV is discussed.

M2-1 protein of human respiratory syncytial virus (RSV) is a transcription antitermination factor that is important for the efficient synthesis of full-length mRNAs as well as for the synthesis of polycistronic readthrough mRNAs, which are characteristic of nonsegmented negative-strand RNA viruses. The contributions of these effects to RSV sequential transcription were investigated with minigenomes which contained one to five genes which were either foreign marker genes or authentic RSV genes. When evaluated on a promoter-proximal gene, the effect of M2-1 on the synthesis of full-length mRNA was much greater for a long (1,212- or 1,780-nucleotide) gene (up to a 615-fold increase) than for a short (274-nucleotide) gene (less than a 2-fold increase). This was independent of whether the gene contained non-RSV or RSV-specific sequence. Once the polymerase had terminated prematurely, it was unable to reinitiate at a downstream gene. These studies also confirmed that M2-1 enhances the synthesis of polycistronic mRNAs and that the magnitude of this effect varied greatly among different naturally occurring gene junctions. The synthesis of polycistronic mRNAs, which presumably involves antitermination at the gene-end signal, required a higher level of M2-1 than did the synthesis of the corresponding monocistronic mRNAs. M2-1 did not have a comparable antitermination effect at the junction between the leader region and the first gene. In a minigenome containing the NS1 and NS2 genes in their authentic sequence context, synthesis of full-length NS1 and NS2 mRNAs in the absence of M2-1 was remarkably high (36 and 57%, respectively, of the maximum levels observed in the presence of M2-1). In contrast, synthesis of mRNA from additional downstream genes was highly dependent on M2-1. Thus, RSV has the potential for two transcription programs: one in the absence of M2-1, in which only the NS1 and NS2 genes are transcribed, and one in the presence of M2-1, in which sequential transcription of the complete genome occurs. The dependence on M2-1 for transcription was greater for a gene in the fifth position from the promoter than for one in the third position. This indicates that under conditions where M2-1 is limiting, its concentration affects the gradient of transcription. Although M2-1 was found to have profound effects on transcription, it had no effect on replication of any minigenome tested, suggesting that it is not an active participant in RNA replication or regulation of RNA replication. Finally, since a permissive RSV infection is marked by a gradual increase in the intracellular accumulation of viral proteins including M2-1, we examined the relative abundances of various mRNAs during RSV infection for evidence of temporal regulation of transcription. None was found, implying that the availability of M2-1 during a permissive infection is sufficient at all times such that its concentration does not mediate temporal regulation of gene transcription.

The NS2 and SH genes of respiratory syncytial virus (RSV) have been separately deleted from a recombinant wild-type RSV strain, A2 (M. N. Teng and P. L. Collins, J. Virol. 73:466-473, 1998; A. Bukreyev et al., J. Virol. 71:8973-8982, 1997; and this study). The resulting viruses, designated rA2DeltaNS2 and rA2DeltaSH, were administered to chimpanzees to evaluate their levels of attenuation and immunogenicity. Recombinant virus rA2DeltaNS2 replicated to moderate levels in the upper respiratory tract, was highly attenuated in the lower respiratory tract, and induced significant resistance to challenge with wild-type RSV. The replication of rA2DeltaSH virus was only moderately reduced in the lower, but not the upper, respiratory tract. However, chimpanzees infected with either virus developed significantly less rhinorrhea than those infected with wild-type RSV. These findings demonstrate that a recombinant RSV mutant lacking either the NS2 or SH gene is attenuated and indicate that these deletions may be useful as attenuating mutations in new, live recombinant RSV vaccine candidates for both pediatric and elderly populations. The DeltaSH mutation was incorporated into a recombinant form of the cpts248/404 vaccine candidate, was evaluated for safety in seronegative chimpanzees, and can now be evaluated as a vaccine for humans.

cpts530/1009 is a live-attenuated, temperature-sensitive (ts) RSV vaccine candidate that was shown previously to be attenuated for seronegative humans. It was generated by two rounds of chemical mutagenesis: first, a partially attenuated, cold-passaged (cp), non-ts RSV mutant (cpRSV) was mutagenized to yield the ts derivative cpts530, and then cpts530 was mutagenized to yield cpts530/1009, which is more ts. Previous nucleotide (nt) sequence analysis of cpts530 showed that it has a single nt change compared to cpRSV that results in an amino acid substitution at residue 521 in the L protein. Reverse genetics confirmed that this mutation is responsible for the ts phenotype of cpts530. Here, determination of the complete 15,222-nt sequence of cpts530/ 1009 identified a single change compared to cpts530, namely a point mutation at nt 12002, which results in a methionine-tovaline substitution at amino acid 1169 in the L protein. The contribution of the 1009 mutation to the level of temperature sensitivity and attenuation exhibited by cpts530/1009 was evaluated by its introduction alone or with the 530 and cp mutations into the full-length cDNA clone of wild-type (wt) RSV. Subsequent analysis of infectious viruses recovered from the mutant cDNAs indicated that (i) the 1009 mutation indeed was a ts mutation and the level of temperature sensitivity specified by the 1009 mutation was less than that specified by the 530 mutation, (ii) the 530 and 1009 mutations each contributed to attenuation in the upper respiratory tract of mice and their effects were additive, (iii) viruses bearing the 1009 mutation were more attenuated in the lower respiratory tract of mice than viruses bearing the 530 mutation and (iv) the combination of the 530 and 1009 mutations in the cpRSV background resulted in the same level of temperature sensitivity and attenuation in mice as that observed for the biologically-derived cpts530/1009 mutant. These data show that the genetic basis of the attenuation and temperature sensitivity of the cpts530/1009 candidate vaccine virus is the sum of the contributions of seven identified amino acid substitutions, i.e. the 5 cpRSV mutations, the 530 mutation and the 1009 mutation.

Respiratory syncytial virus (RSV) cpts530/1030 is an attenuated, temperature-sensitive subgroup A vaccine candidate derived previously from cold-passaged RSV (cpRSV) by two sequential rounds of chemical mutagenesis and biological selection. Here, cpts530/1030 was shown to be highly attenuated in the upper and lower respiratory tracts of seronegative chimpanzees. However, evaluation in seropositive children showed that it retains sufficient replicative capacity and virulence to preclude its direct use as a live attenuated vaccine. Nucleotide sequence analysis of the genome of cpts530/1030 showed that it had acquired two nucleotide substitutions (compared to its cpts530 parent), both of which were in the L gene: a silent mutation at nucleotide position 8821 (amino acid 108) and a missense mutation at nucleotide position 12458 resulting in a tyrosine-to-asparagine change at amino acid 1321, herein referred to as the 1030 mutation. It also contained the previously identified 530 missense mutation at nucleotide 10060 in the L gene. The genetic basis of attenuation of cpts530/1030 was defined by the introduction of the 530 and 1030 mutations into a cDNA clone of cpRSV, from which recombinant RSV was derived and analyzed to determine the contribution of each mutation to the temperature sensitivity (ts) and attenuation (att) phenotypes of cpts530/1030. The 530 mutation, derived from cpts530, was previously shown to be responsible for the ts and att phenotypes of that virus. In the present study, the 1030 mutation was shown to be responsible for the increased temperature sensitivity of cpts530/1030. In addition, the 1030 mutation was shown to be responsible for the increased level of attenuation of cpts530/1030 in the upper and lower respiratory tracts of mice. The 530 and 1030 mutations were additive in their effects on the ts and att phenotypes. It was possible to introduce the 1030 mutation, but not the 530 mutation, into an attenuated vaccine candidate with residual reactogenicity in very young infants, namely, cpts248/404, by use of reverse genetics. The inability to introduce the 530 mutation into the cpts248/404 virus was shown to be due to its incompatibility with the 248 missense mutation at the level of L protein function. The resulting rA2cp248/404/1030 mutant virus was more temperature sensitive and more attenuated than the cpts248/404 parent virus, making it a promising new RSV vaccine candidate created by use of reverse genetics to improve upon an existing vaccine virus.

The last two genes of respiratory syncytial virus (RSV), M2 and L, overlap by 68 nucleotides, an arrangement which has counterparts in a number of nonsegmented negative-strand RNA viruses. Thus, the gene-end (GE) signal of M2 lies downstream of the L gene-start (GS) signal, separated by 45 nucleotides. Since RSV transcription ostensibly is sequential and unidirectional from a single promoter within the 3' leader region, it was unclear how the polymerase accesses the L GS signal. Furthermore, it was previously shown that 90% of transcripts which are initiated at the L GS signal are polyadenylated and terminated at the M2 GE signal, yielding a short, truncated L mRNA as the major transcription product of the L gene. Despite these apparent down-regulatory features, we show that the accumulation of full-length L mRNA during RSV infection is only sixfold less than that of its upstream neighbor, M2. We used cDNA-encoded genome analogs in an intracellular transcription assay to investigate the mechanism of transcription of the overlapped genes. Expression of L was found to be dependent on sequential transcription from the 3' end of the genome. Apart from the L GS signal, the only other strict requirement for initiation at L was the M2 GE signal. This implies that the polymerase accesses the L GS signal only following arrival at the M2 GE signal. Thus, polymerase which terminates at the M2 GE signal presumably scans upstream to initiate at the L GS signal. This also would provide a mechanism whereby polymerase which terminates prematurely during transcription of L could recycle from the M2 GE signal to the L GS signal, thereby accounting for the unexpectedly high level of synthesis of full-length L mRNA. The sequence and spacing between the two signals were not critical. Furthermore, the polymerase also was capable of efficiently transcribing from an L GS signal placed downstream of the M2 GE signal, implying that the overlapping arrangement is not obligatory. When copies of the L GS signal were placed concurrently upstream and downstream of the M2 GE signal, both were utilized. This finding indicates that a polymerase situated at a GE signal is capable of scanning for a GS signal in either the upstream or downstream direction and thereafter initiating transcription.

The complete nucleotide sequence of a functional clone of the large polymerase (L) gene of bovine respiratory syncytial virus (BRSV) strain A51908 was determined by analysis of cloned cDNAs obtained from genomic and mRNAs. The BRSV L gene is 6573 nt in length and the derived polypeptide has 2162 aa. Alignment of the sequences of the BRSV L gene, and its encoded protein, with sequences of the L gene and protein of human respiratory syncytial virus strain A2 showed 77% identity at the nucleotide level and 84% identity at the amino acid level. By comparison, the L gene and protein of avian pneumovirus showed only 50% identity at the nucleotide level and 64% identity at the amino acid level. A minigenome was constructed to encode a BRSV vRNA analogue containing the gene for chloramphenicol acetyltransferase (CAT) under the control of putative BRSV transcription motifs and flanked by the BRSV genomic termini. Transfection of plasmids encoding the BRSV minigenome, nucleocapsid protein (N), phosphoprotein (P) and L protein, each under the control of T7 promoter, into cells infected with a vaccinia virus recombinant expressing the T7 RNA polymerase gave rise to CAT activity and progeny with the minigenome. This result indicates that the N, P and L proteins are necessary and sufficient for transcription and replication of the BRSV minigenome and are functional. Further, inclusion of small amounts of the M2 protein along with the N, P and L proteins greatly augmented minigenome transcription.

Respiratory syncytial virus (RSV) cpts248/404 is a live-attenuated, temperature-sensitive (ts) vaccine candidate derived from cole-passaged cpRSV by two rounds of chemical mutagenesis and biological selection. Previous sequence analysis showed that these two steps introduced three single nucleotide substitutions into the cpRSV parent. Two of these occurred with the coding sequence for the L protein, and each resulted in a single amino acid substitution: Gin-831-Leu (248 mutation) and Asp-1183-Glu (404-L mutation). The third mutation resulted in a nucleotide substitution at position 9 of the c/s-acting gene start signal of the M2 gene (404-M2 mutation). In the present study, the genetic basis of attenuation of cpts248/404 was defined by the introduction of each of these mutations (singly or in combination) into a full-length cDNA clone of cpRSV. Recombinant RSV derived from each mutant cDNA was analyzed to determine the contribution of each mutation to the ts and attenuation phenotypes of the virus. This analysis showed that the 248 mutation specifies a significant reduction of plaque formation at 38 degrees and is responsible for an intermediate level of attenuation in mice. In contrast, the 404-L mutation did not contribute to the ts or attenuation phenotype alone or in combination with other mutations and is thus an incidental change. unexpectedly, the 404-M2 mutation alone specified complete restriction of plaque formation at 37 degrees C an a high level of attenuation in mice. This indicates that the level of temperature sensitivity and attenuation of cpts248/404 can be attributed primarily to the 404-M2 mutation. Thus the cpts248/404 virus contains a set of ts and non-ts attenuating mutations, which likely accounts for its genetic stability. The recombinant version of this virus, rA2cp248/404, was phenotypically indistinguishable from cpts248/404 and represents a background into which additional mutations can be introduced as needed to obtain the desired level of attenuation for successful immunization of the very young human infant.

The recent recovery of human parainfluenza virus type 3 (PIV3) from cDNA, together with the availability of a promising, highly characterized live attenuated PIV3 vaccine virus, suggested a novel strategy for the rapid development of comparable recombinant vaccine viruses for human PIV1 and PIV2. The strategy, illustrated here for PIV1, is to create chimeric viruses in which the two protective antigens, the hemagglutinin-neuraminidase (HN) and fusion (F) envelope glycoproteins, of an attenuated PIV3 variant are replaced by those of PIV1 or PIV2. As a first step, this has been achieved by using recombinant wild-type (wt) PIV3 as the recipient for PIV1 HN and F, engineered so that each PIV1 open reading frame is flanked by the existing PIV3 nontranslated regions and transcription signals. This yielded a viable chimeric recombinant virus, designated rPIV3-1, that encodes the PIV1 HN and F glycoproteins in the background of the wt PIV3 internal proteins. There were three noteworthy findings. First, in contrast to recently reported glycoprotein replacement chimeras of vesicular somatitis virus or measles virus, the PIV3-1 chimera replicates in LLC-MK2 cells and in the respiratory tract of hamsters as efficiently as its PIV1 and PIV3 parents. This is remarkable because the HN and F glycoproteins share only 43 and 47%, respectively, overall amino acid sequence identity between serotypes. In particular, the cytoplasmic tails share only 9 to 11% identity, suggesting that their presumed role in virion morphogenesis does not involve sequence-specific contacts. Second, rPIV3-1 was found to possess biological properties derived from each of its parent viruses. Specifically, it requires trypsin for efficient plaque formation in tissue culture, like its PIV1 parent but unlike PIV3. On the other hand, it causes an extensive cytopathic effect (CPE) in LLC-MK2 cultures which resembles that of its PIV3 parent but differs from that of its noncytopathic PIV1 parent. This latter finding indicates that the genetic basis for the CPE of PIV3 in tissue culture lies outside regions encoding the HN or F glycoprotein. Third, it should now be possible to rapidly develop a live attenuated PIV1 vaccine by the staged introduction of known, characterized attenuating mutations present in a live attenuated PIV3 vaccine candidate into the PIV3-1 cDNA followed by recovery of attenuated derivatives of rPIV3-1.

The small hydrophobic protein SH of human respiratory syncytial virus (RSV) is a short transmembrane surface protein of unknown function. A full-length cDNA of RSV strain A2 (subgroup A) antigenomic RNA was modified such that the entire SH gene, including the transcription signals and the complete mRNA-encoding sequence, was deleted and replaced by a synthetic intergenic region. This reduced the length of the antigenome by 398 nucleotides and ablated expression of 1 of the 10 RSV mRNAs. Recombinant virus containing this engineered deletion was recovered, and the absence of the SH gene was confirmed by reverse transcription in conjunction with PCR. Northern blot analysis of intracellular RNAs and gel electrophoresis of labeled intracellular proteins confirmed the lack of expression of the SH mRNA and protein. The absence of the SH gene did not noticeably affect RNA replication, but two effects on transcription were noted. First, synthesis of the G, F, and M2 mRNAs was increased, presumably due to their being one position closer to the promoter in the gene order. Second, transcription of genes downstream of the engineered site exhibited a steeper gradient of polarity. On monolayers of HEp-2 cells, the SH-minus virus produced syncytia which were at least equivalent in size to those of the wild type and produced plaques which were 70% larger. Furthermore, the SH-minus virus grew somewhat better (up to 12.6-fold) than wild-type recombinant RSV in certain cell lines. While the function of the SH protein remains to be determined, it seems to be completely dispensable for growth in tissue culture and fusion function. When inoculated intranasally into mice, the SH-minus virus resembled the wild-type recombinant virus in its efficiency of replication in the lungs, whereas it replicated 10-fold less efficiently in the upper respiratory tract. In mice, the SH-minus and wild-type recombinant viruses were similarly immunogenic and effective in inducing resistance to virus challenge.

A popular model for RNA synthesis by nonsegmented negative-strand RNA viruses is that transcription and RNA replication are executed by the same polymerase complex and that there is a dynamic balance between the two processes that is mediated by the nucleocapsid N protein. According to this model, transcription occurs until sufficient soluble N protein accumulates to initiate encapsidation of the nascent RNA product, which somehow switches the polymerase into a readthrough replicative mode. This model was examined for respiratory syncytial virus (RSV) using a reconstituted transcription and RNA replication system that involves a minireplicon and viral proteins that are expressed intracellularly from transfected plasmids. Preliminary experiments showed that reconstituted RNA replication was highly productive, such that on average each molecule of plasmid-supplied minigenome that became encapsidated was amplified 10- to 50-fold. N protein was increased on its own or in concert with the phosphoprotein P and in the presence or absence of the M2 ORF1 transcription elongation factor. The maximum level of N and P protein expression achieved from plasmids equalled or exceeded that obtained in RSV-infected cells. Increased levels of N protein stimulated RNA replication. This is consistent with the idea that RNA replication is dependent on the availability of N protein for encapsidation, which is one postulate of the model. The M2 ORF1 protein had no detectable effect on RNA replication under the various conditions of expression of N and P, which confirmed and extended previous results. However, there was no evidence of a significant switch in positive-sense RNA synthesis from transcription (synthesis of mRNAs) to RNA replication (synthesis of antigenome). The synthesis of positive-sense antigenome and mRNA appeared to occur at a fixed ratio, with mRNA being by far the more abundant product., (Copyright 1997 Academic Press.)

cpts530, a candidate live-virus vaccine, is an attenuated strain of human respiratory syncytial virus (RSV). It was derived by subjecting a cold-passaged (cp) strain of RSV to a single round of chemical mutagenesis. cpts530 is a temperature-sensitive (ts) mutant that is attenuated in mice and chimpanzees, and its ts phenotype exhibits a high level of stability during replication in both species. In the present study, the complete nucleotide sequence of cpts530 RSV was determined. The five mutations known to be present in the parent cpRSV were retained in its cpts530 derivative, and one additional nucleotide change was identified at nucleotide (nt) 10060, which resulted in a phenylalanine-to-leucine change at amino acid 521 in the large polymerase (L) protein. To determine if this single amino acid substitution was indeed responsible for the ts phenotype of cpts530, it was introduced alone or in combination with the cp mutations into the full-length cDNA clone of the wild-type A2 RSV. Analysis of infectious viruses recovered from mutant cDNAs indicated that this single mutation specified complete restriction of plaque formation of recombinant cp530 in HEp-2 cell monolayer cultures at 40 degrees C, and the level of temperature sensitivity was not influenced by the presence of the five cpRSV mutations. These findings identify the phenylalanine-to-leucine change at amino acid 521 in the L protein as the mutation that specifies the ts phenotype of cpts530. Furthermore, these findings illustrate the feasibility of using the cDNA-based recovery system to analyze and construct defined attenuated vaccine viruses.

Infectious human respiratory syncytial virus (RSV) was produced by the intracellular coexpression of five plasmid-borne cDNAs. One cDNA encoded a complete positive-sense version of the RSV genome (corresponding to the replicative intermediate RNA or antigenome), and each of the other four encoded a separate RSV protein, namely, the major nucleocapsid N protein, the nucleocapsid P phosphoprotein, the major polymerase L protein, or the protein from the 5' proximal open reading frame of the M2 mRNA [M2(ORF1)]. RSV was not produced if any of the five plasmids was omitted. The requirement for the M2(ORF1) protein is consistent with its recent identification as a transcription elongation factor and confirms its importance for RSV gene expression. It should thus be possible to introduce defined changes into infectious RSV. This should be useful for basic studies of RSV molecular biology and pathogenesis; in addition, there are immediate applications to the development of live attenuated vaccine strains bearing predetermined defined attenuating mutations.

Previously, a cDNA was constructed so that transcription by T7 RNA polymerase yielded a approximately 1-kb negative-sense analog of genomic RNA of human respiratory syncytial virus (RSV) containing the gene for chloramphenicol acetyltransferase (CAT) under the control of putative RSV transcription motifs and flanked by the RSV genomic termini. When transfected into RSV-infected cells, this minigenome was "rescued," as evidenced by high levels of CAT expression and the production of transmissible particles which propagated and expressed high levels of CAT expression during serial passage (P.L. Collins, M. A. Mink, and D. S. Stec, Proc. Natl. Acad. Sci. USA, 88:9663-9667, 1991). Here, this cDNA, together with a second one designed to yield an exact-copy positive-sense RSV-CAT RNA antigenome, were each modified to contain a self-cleaving hammerhead ribozyme for the generation of a nearly exact 3' end. Each cDNA was transfected into cells infected with a vaccinia virus recombinant expressing T7 RNA polymerase, together with plasmids encoding the RSV N, P, and L proteins, each under the control of a T7 promoter. When the plasmid-supplied template was the mini-antigenome, the minigenome was produced. When the plasmid-supplied template was the minigenome, the products were mini-antigenome, subgenomic polyadenylated mRNA and progeny minigenome. Identification of progeny minigenome made from the plasmid-supplied minigenome template indicates that the full RSV RNA replication cycle occurred. RNA synthesis required all three RSV proteins, N, P, and L, and was ablated completely by the substitution of Asn for Asp at position 989 in the L protein. Thus, the N, P, and L proteins were sufficient for the synthesis of correct minigenome and antigenome, but this was not the case for subgenomic mRNA, indicating that the requirements for RNA replication and transcription are not identical. Complementation with N, P, and L alone yielded an mRNA pattern containing a large fraction of molecules of incomplete, heterogeneous size. In contrast, complementation with RSV (supplying all of the RSV gene products) yielded a single discrete mRNA band. Superinfection with RSV of cells staging N/P/L-based RNA synthesis yielded the single discrete mRNA species. Some additional factor supplied by RSV superinfection appeared to be involved in transcription, the most obvious possibility being one or more additional RSV gene products.

In previous studies, a mutant (cp-RSV) of the RSV A2 strain derived from 52 serial cold passages in bovine embryonic tissue culture was highly attenuated in seropositive adults and children but caused upper respiratory tract disease in seronegative infants. We investigated the genetic basis for this attenuation phenotype by comparing the complete genomic RNA sequence of this virus with the published sequence of strain A2 as well as with that of its unattenuated wild-type parent (HEK-7) virus. RNA was extracted from virions grown in tissue culture, reverse transcribed, amplified by the polymerase chain reaction (PCR), cloned, and sequenced. Changes from the published A2 wild-type sequence were confirmed on independently derived cDNA clones and by direct sequencing of PCR fragments. The HEK-7 parent virus was then analyzed at these positions by direct sequencing of PCR fragments. Fifteen nucleotide differences between the published A2 wild-type virus and cp-RSV were found. None appeared to involve cis-acting RNA sequences. Of the 15 nucleotide differences, only 1 occurred outside a translational open reading frame (ORF), and 2 which did occur within ORFs were silent at the amino acid level. The remaining 12 nucleotide differences encoded amino acid changes. All 3 of the mutations which were silent at the amino acid level, and 8 of the 12 which resulted in amino acid differences, were also present in the HEK-7 parent virus and therefore were not changes acquired during the cold passages. Thus, the remaining 4 nucleotide differences and the attendant 4 amino acid changes are associated with the attenuation phenotype of the cp-RSV. Two of the changes occur in the F gene and two in the L gene. These results confirm the previously described RSV genomic sequence, provide the first sequence of a live attenuated RSV vaccine strain, provide the first sequence of an RSV strain which has been evaluated in chimpanzees and humans, and indicate that attenuation in humans of a pneumovirus can be associated with a relatively small number of nucleotide and amino acid changes.

We have studied the immunobiology of respiratory syncytial virus (RSV), a major cause of respiratory tract morbidity in children. As part of these studies, it was previously found that immunization of BALB/c (H-2d) mice with a recombinant vaccinia virus (rVV) which encoded the M2 protein of RSV provided complete protection against infection with RSV. This protection was transient and associated with M2-specific CD8+ T-cell (TCD8+) responses. In this study, we used two approaches to demonstrate that expression of an H-2Kd-restricted nonameric peptide (Ser Tyr Ile Gly Ser Ile Asn Asn Ile) corresponding to M2 residues 82 to 90 is necessary and sufficient to induce protective TCD8+ responses. First, infection of mice with an rVV which encoded the peptide M2Met82-90 induced levels of primary pulmonary TCD8+ and resistance to RSV challenge equivalent to that induced by infection with an rVV which expressed the complete M2 protein. Second, elimination of peptide binding to Kd by the replacement of Tyr with Arg at amino acid position 83 of the full-length protein completely abrogated the ability of an rVV-expressing full-length M2 to induce either M2-specific TCD8+ responses or resistance to RSV infection. These findings demonstrate that the M2(82-90) peptide is the sole determinant of immunity induced in BALB/c mice by the M2 protein and that a remarkably high level of transient resistance to infection with pulmonary virus is associated with TCD8+ responses to a single determinant.

Respiratory syncytial virus (RSV) is the most common cause of viral bronchiolitis and pneumonia in children. The present study compares the level of attenuation, genetic stability and efficacy of three conditional-lethal temperature-sensitive (ts) mutants of the RSV A2 wild-type virus, designated ts-1, ts-1-NG1, and ts-4, in seronegative chimpanzees and also compares their efficacy with that of vaccinia virus recombinants that express the surface glycoproteins of RSV. Each of the ts mutants was highly attenuated in the lower respiratory tract, but still retained the capacity to induce significant rhinorrhoea. Each of the three ts mutants underwent partial reversion to a non-ts (ts+) phenotype during replication in a minority of the chimpanzees. The ts+ virus present in the upper respiratory tract of the chimpanzees did not spread to the lower respiratory tract and represented only a minority fraction of the virus present in the nasopharyngeal swab specimens. The ts mutants were highly immunogenic and provided resistance that effectively restricted RSV replication following virus challenge. In contrast, the vaccinia-RSV recombinants were less immunogenic. They protected the lungs of two of four chimpanzees challenged with RSV, but failed to protect the upper respiratory tract. The chimpanzee can serve as a model for the rapid evaluation of further attenuated live RSV vaccines.

Previous work has demonstrated that the small hydrophobic (SH) protein of human respiratory syncytial virus (RSV) A2 strain is a 64 amino acid integral membrane protein that accumulates intracellularly as an unglycosylated major species (SH0), a minor species truncated at the amino terminus and two N-glycosylated species one of which contains a further addition of polylactosamine. In this study, the membrane orientation of SH0 was mapped by trypsinization of intact RSV-infected cells followed by washout, lysis and immunoprecipitation of protected fragments with antisera specific for the protein termini. This showed that the C terminus is extracellular and the SH protein was not detectably palmitylated. Analysis of the SH protein by sedimentation on sucrose gradients showed that it rapidly assembles into a homo-oligomer that co-sediments with the F protein tetramer. Interestingly, all forms of the SH protein were found in the oligomeric fraction. Chemical cross-linking generated species which appeared to represent dimers, trimers, tetramers and pentamers as well as a minor species of 180K which might correspond to the oligomeric form detected by sucrose gradient sedimentation.

The attachment protein, G, of human respiratory syncytial virus (RSV) is an M(r) 84K to 90K species which has a high content of N-linked and O-linked carbohydrates. The unglycosylated form of this protein was expressed by inserting a full-length cDNA copy of the mRNA from the A2 strain of RSV into a prokaryotic expression vector under the control of the lambda PL promoter. Salmonella typhimurium cells transformed with the G-containing plasmid synthesized a protein of M(r) 40,000 that specifically reacted with polyclonal and two neutralizing monoclonal antibodies raised against the native RSV G glycoprotein. Recombinant G protein was purified by immunoaffinity chromatography using a neutralizing monoclonal antibody. Cotton rats immunized with the recombinant G protein produced serum antibodies to the G glycoprotein that neutralized RSV in vitro. The study demonstrates that the G protein of RSV can be expressed in bacteria and that at least one neutralizing epitope is not structurally dependent on carbohydrates.

Recombinant adenovirus type 4, 5, and 7 expressing the fusion glycoprotein (F) gene, the attachment glycoprotein (G) gene, or both F and G genes of respiratory syncytial virus (RSV) was constructed. Intratracheal immunization of dogs with Ad7F induced moderate titers of RSV-neutralizing antibodies. After booster immunization with Ad4F, the dogs developed high titers of RSV-specific antibody. Subsequently, three two-dose vaccination regimens, Ad4F/Ad5F, Ad7G/Ad4G, and Ad7FG/Ad4FG, were compared with Ad7F/Ad4F for immunogenicity and protective efficacy. The results indicated that Ad4F/Ad5F was equal or greater in immunogenicity to Ad7F/Ad4F, but Ad7G/Ad4G and Ad7FG/Ad4FG were less effective than Ad7F/Ad4F in inducing RSV-neutralizing antibody. All vaccination regimens completely protected the lungs of dogs from RSV infection. A chimpanzee was sequentially immunized orally with Ad7F, Ad4F, and Ad5F. A low-level antibody response to RSV was induced after the primary immunization, but no significant increases were observed after booster immunizations.

The post-translational maturation of the attachment G glycoprotein of human respiratory syncytial virus (RSV) was investigated. The G protein formed homo-oligomers which sedimented in sucrose gradients at the same rate as the fusion F protein tetramer. Oligomerization of the G protein was insensitive to carbonylcyanide m-chlorophenylhydrazine, showing that this step occurs in the endoplasmic reticulum prior to O-glycosylation which initiated in the trans-Golgi compartment. The sedimentation of the G protein oligomer was essentially unchanged by the subsequent addition of O-linked sugars. This indicated that their contribution to the M(r) of the G protein is less than that estimated by electrophoretic mobility. It also suggested that O-glycosylation is not an important determinant of G protein oligomerization and, by implication, of polypeptide folding. The G protein is palmitylated. In short labelling pulses, the G protein accumulated as two species of 48K and 50K which contained only N-linked sugars, whose difference in M(r) was due solely to an N-linked sugar, which both assembled into oligomers, but which differed in the rate of subsequent O-glycosylation. The G protein was not detectably O-glycosylated in the presence of monensin, confirming previous work. In the presence of brefeldin A (BFA), it accumulated as a partially O-glycosylated species (BFA-G) of 68K to 78K. But further analysis by chase incubations following BFA-washout, by lectin-binding, and by glycosidase treatment suggested that BFA-G was not a fully authentic processing intermediate. In particular, some of the O-linked side-chains of the BFA-G protein were found to be sialylated. Rather than being a normal step in processing, this sialylation probably was due to altered distribution or activity of sialyltransferases during BFA treatment and may have resulted in the premature termination of elongation of some of the O-linked side-chains. Thus, these studies (i) indicate that O-glycosylation of the G protein begins in the trans-Golgi compartment and (ii) suggest that O-glycosylation is completed in as a subsequent compartment, but this latter suggestion is complicated by the evidence that the BFA-G protein is not a fully authentic intermediate.(ABSTRACT TRUNCATED AT 400 WORDS)

It was previously demonstrated that the vaccinia virus recombinants expressing the respiratory syncytial virus (RSV) F, G, or M2 (also designated as 22K) protein (Vac-F, Vac-G, or Vac-M2, respectively) induced almost complete resistance to RSV challenge in BALB/c mice. In the present study, we sought to identify the humoral and/or cellular mediators of this resistance. Mice were immunized by infection with a single recombinant vaccinia virus and were subsequently given a monoclonal antibody directed against CD4+ or CD8+ T cells or gamma interferon (IFN-gamma) to cause depletion of effector T cells or IFN-gamma, respectively, at the time of RSV challenge (10 days after immunization). Mice immunized with Vac-F or Vac-G were completely or almost completely resistant to RSV challenge after depletion of both CD4+ and CD8+ T cells prior to challenge, indicating that these cells were not required at the time of virus challenge for expression of resistance to RSV infection induced by the recombinants. In contrast, the high level of protection of mice immunized with Vac-M2 was completely abrogated by depletion of CD8+ T cells, whereas depletion of CD4+ T cells or IFN-gamma resulted in intermediate levels of resistance. These results demonstrate that antibodies are sufficient to mediate the resistance to RSV induced by the F and G proteins, whereas the resistance induced by the M2 protein is mediated primarily by CD8+ T cells, with CD4+ T cells and IFN-gamma also contributing to resistance.

In studies conducted in the 1960s, children previously immunized with a formalin-inactivated respiratory syncytial virus (RSV) vaccine (FI-RSV) developed a greater incidence and severity of pulmonary disease during subsequent natural RSV infection than did controls. It was previously shown that cotton rats immunized with FI-RSV or immunoaffinity-purified fusion (F) glycoprotein developed enhanced pulmonary histopathology following intranasal challenge with RSV. In the present studies, various forms of immunization, including parenteral inoculation of an immunoaffinity-purified F glycoprotein or a chimeric FG glycoprotein produced in insect cells using a baculovirus vector (Bac-FG), intradermal infection with a vaccinia-F recombinant (Vac-F) or intranasal infection with an adenovirus-F recombinant (Ad-F) or RSV, were compared for immunogenicity, efficacy and ability to alter the host so that enhanced pulmonary histopathology developed during RSV infection 3 months after immunization. Immunization of cotton rats with F glycoprotein, Bac-FG, Vac-F, Ad-F or infection with RSV induced high levels of ELISA-F antibodies, but the antibodies induced by purified F glycoprotein of Bac-FG had low levels of neutralizing activity. Immunization with Vac-F or Ad-F, or infection with RSV induced a high level of resistance to pulmonary RSV replication, whereas animals immunized with Bac-FG or FI-RSV were only partially protected. Following RSV challenge, animals immunized with purified F glycoprotein or Bac-FG developed the highest levels of bronchiolar and alveolar histopathology, those immunized with FI-RSV had intermediate levels, and those immunized with Vac-F or RSV had histopathology scores at control levels. Ad-F immunized animals had elevated scores of bronchiolar but not alveolar histopathology; however, this finding was not reproducible. Passive transfer of pooled immune sera from animals infected with RSV or Vac-F and Vac-G was highly protective, whereas pooled sera from animals immunized with Bac-FG failed to protect the lungs against RSV challenge. Increased pulmonary histopathology was not observed in the passively immunized animals following RSV challenge, suggesting that the histopathology was mediated by RSV-specific T cells. These data indicate that subunit F glycoprotein or chimeric FG vaccines share with FI-RSV the properties of (i) induction of F antibodies with low neutralizing activity and (ii) enhancement of pulmonary histopathology during subsequent RSV infection. These observations confirm the need for caution in studies involving the administration of RSV subunit vaccines to seronegative humans.

The post-translational maturation of the fusion protein (F) of human respiratory syncytial virus was investigated. Chemical cross-linking experiments indicated that F forms homotetramers and provided evidence that the intermonomer contacts involve primarily the F1 subunit. Homooligomerization as measured by sedimentation in sucrose gradients was insensitive to carbonyl cyanide m-chlorophenylhydrazone, indicating that it occurs in the endoplasmic reticulum. Cleavage of the F0 precursor to yield the F1 and F2 subunits was blocked by monensin or brefeldin A, indicating that it takes place in distal cisternae of the trans Golgi compartment or in the more distal trans Golgi network. The F0 precursor was not detected at the cell surface in surface immunoprecipitation experiments, indicating that cleavage is intracellular. The appearance of the cleaved F1 protein at the cell surface was concurrent with that of the attachment glycoprotein (G); this and other information indicated that the type 2 membrane orientation of G is not obligatorily associated with a reduced transit rate. Examination of F maturation in the presence of tunicamycin provided evidence that its expression at the cell surface depends upon cleavage and not directly upon glycosylation.

A cDNA copy of the gene encoding the entire amino acid sequence of the fusion (F) protein of human respiratory syncytial virus (strain A2) was inserted into a bacterial expression vector containing the lambda PR promoter. Upon heat induction, Escherichia coli cells harboring the vector produced a 45-kDa peptide which reacted with rabbit polyclonal antiserum to the native F protein. Expression of the F gene resulted in severe inhibition of bacterial growth, which was overcome by deletion of the DNA sequences encoding the F signal peptide. The region of the F protein which reacted with a virus-neutralizing and fusion-inhibiting monoclonal antibody was probed by expressing cDNA fragments encoding different protein domains in E. coli and testing antibody reactivity by Western blot analysis. Analysis of six fragments yielded an overlapping antibody-reactive region between amino acids 253 and 298. Analysis of reactivity with a cassette of synthetic peptides confirmed that the virus-neutralizing epitope mapped between residues 289 and 298 defined by the amino acid sequence M-S-I-I-K-E-E-V-L-A.

The complete nucleotide sequence of the large (L) polymerase gene of human respiratory syncytial virus (RSV) strain A2 was determined by analysis of cloned-cDNAs representing the entire gene and confirmed in part by dideoxy sequencing of genomic RNA. The RSV L gene is 6578 nucleotides in length and contains a single major open reading frame that encodes a protein of 2165 amino acids. The molecular weight (250,226) and amino acid composition of the deduced RSV L protein are similar to those of other negative-strand RNA viruses. Regions of statistically significant amino acid sequence similarity were identified in pairwise global alignments of the RSV L protein with its counterparts in four paramyxoviruses (parainfluenza virus type 3, Sendai virus, measles virus, Newcastle disease virus) and two rhabdoviruses (rabies virus, vesicular stomatitis virus). In addition, amino acid sequence alignments showed that the RSV L protein has a 70-amino acid amino-terminal extension relative to the others. This is suggested to be due to the acquisition of gene overlap of the RSV L gene with its upstream neighbor, the 22K (M2) gene and the use of a new translational start site. The most highly related region among these seven proteins is located within the amino-terminal half, representing approximately 20% of each protein sequences. This region contains six discrete segments that are colinear and highly conserved in each paramyxovirus and rhabdovirus L protein, and three of these overlapped with sequence motifs found previously in other RNA-dependent RNA and DNA polymerases. A phylogenetic tree was constructed from the paramyxovirus and rhabdovirus L protein sequences to further define their relationships. The branching order indicates that RSV represents a lineage within the paramyxovirus family which is relatively distinct from the others, which in turn are more closely interrelated. Among these other members of the family Paramyxoviridae, the branching order does not entirely conform to their current taxonomic organization, providing support for its reevaluation.

The posttranslational maturation of the hemagglutinin-neuraminidase (HN) glycoprotein of human parainfluenza type 3 virus (PIV3) was investigated in pulse-chase experiments in which folding was monitored by immunoprecipitation with conformation-dependent antibodies and gel electrophoresis under nonreducing conditions and oligomerization was monitored by chemical cross-linking and sedimentation in sucrose gradients. The acquisition of mature immunoreactivity and the formation of correct intramolecular disulfide bonds were concurrent events, with half-times of approximately 10 to 15 min. The finding that newly synthesized HN had little reactivity with postinfection cotton rat serum or with most of the members of a panel of HN-specific monoclonal antibodies indicated that the major epitopes of the PIV3 HN protein are highly conformational in nature. Chemical cross-linking studies indicated that the mature HN protein is present in homoligomers, which are probably tetramers. These findings are consistent with recent observations for the HN protein of Sendai virus (S.D. Thompson, W.G. Laver, K.G. Murti, and A. Portner, J. Virol. 62:4653--4660, 1988; S. Vidal, G. Mottet, D. Kolakofsky, and L. Roux, J. Virol. 63:892--900, 1989). Surprisingly, analysis of pulse-labeled HN protein by sedimentation on sucrose gradients after labeling periods of as little as 2 min indicated that it was present intracellularly only in oligomeric form. The same results were obtained when the labeling period was preceded by a 1.5-h cycloheximide treatment to clear the endoplasmic reticulum of presynthesized HN protein, which indicated that the oligomerization did not involve the incorporation of newly synthesized monomers into partially assembled oligomers. Subsequent chase incubations did not significantly alter the sedimentation profile or stability of the oligomeric forms, suggesting that oligomers detected after short labeling periods were tetramers. Association with cellular proteins did not appear to be responsible for the sedimentation of newly synthesized HN protein as an oligomer. The absence of a detectable monomeric form of intracellular HN protein raised the possibility that oligomerization is cotranslational, and it is possible that the type II membrane orientation of the HN protein might be an important factor in its mode of oligomerization.

The effect of passively transferred RSV immune serum on the antibody response to a single dose of purified RSV fusion (F) and large (G) glycoproteins was studied in cotton rats. Passively transferred antibody that achieved serum antibody levels similar to those seen in newborn human infants resulted in a seven- to eightfold suppression of the neutralizing antibody response of cotton rats to low doses of purified F and G glycoproteins (0.2-1.7 micrograms) and a twofold suppression to higher doses of these antigens (5-15 micrograms). This suppression of the antibody response was accompanied by a reduction in the protective efficacy of the F and G purified glycoprotein vaccine. These results suggest that parenteral immunization with RSV antigens could be less immunogenic in seropositive human infants, but that this suppressive effect might be partially overcome with increased antigen dose.

Effector cells capable of lysing respiratory syncytial virus (RSV)-infected cells were isolated from the lungs of intranasally infected mice. An examination of their specificity showed that cytolysis was major histocompatibility complex restricted. Using recombinant vaccinia viruses containing cloned RSV genes to infect target cells, BALB/c (H-2d) pulmonary effector cells were shown to recognize the fusion protein (F) and to a lesser extent the nucleoprotein (N). Cells specific for the major glycoprotein (G) or the small hydrophobic protein (1A) were not observed.

Six amino-terminal deletion mutants of the NH2-terminally anchored (type II orientation) hemagglutinin-neuraminidase (HN) protein of parainfluenza virus type 3 were expressed in tissue culture by recombinant SV-40 viruses. The mutations consisted of progressive deletions of the cytoplasmic domain and, in some cases, of the hydrophobic signal/anchor. Three activities were dissociated for the signal/anchor: membrane insertion, translocation, and anchoring/transport. HN protein lacking the entire cytoplasmic tail was inserted efficiently into the membrane of the endoplasmic reticulum but was translocated inefficiently into the lumen. However, the small amounts that were successfully translocated appeared to be processed subsequently in a manner indistinguishable from that of parental HN. Thus, the cytoplasmic domain was not required for maturation of this type II glycoprotein. Progressive deletions into the membrane anchor restored efficient translocation, indicating that the NH2-terminal 44 amino acids were fully dispensable for membrane insertion and translocation and that a 10-amino acid hydrophobic signal sequence was sufficient for both activities. These latter HN molecules appeared to be folded authentically as assayed by hemagglutination activity, reactivity with a conformation-specific antiserum, correct formation of intramolecular disulfide bonds, and homooligomerization. However, most (85-90%) of these molecules accumulated in the ER. This showed that folding and oligomerization into a biologically active form, which presumably represents a virion spike, occurs essentially to completion within that compartment but is not sufficient for efficient transport through the exocytotic pathway. Protein transport also appeared to depend on the structure of the membrane anchor. These latter mutants were not stably integrated in the membrane, and the small proportion (10-15%) that was processed through the exocytotic pathway was secreted. The maturation steps and some of the effects of mutations described here for a type II glycoprotein resemble previous observations for prototypic type I glycoproteins and are indicative of close similarities in these processes for proteins of both membrane orientations.

The replication of RSV in unimmunized cotton rats was evaluated by quantitating the amount of infectious virus in the lung and the number of RSV infected cells in a histopathological section of lung by in situ hybridization. RSV infected cells were detected only in alveoli and bronchioles and constituted only a small minority of the cell population. The temporal patterns of rise to the peak number of infected cells (day 4) and the peak titer of infectious virus (day 3) were similar. The clearance of both infected cells and infectious virus was nearly complete by day 7. In animals previously immunized with purified RSV glycoproteins or formalin-inactivated RSV there also was a good correlation between the number of infected cells detected by in situ hybridization and the amount of infectious virus recovered. It was previously demonstrated that cotton rats immunized with formalin-inactivated vaccine developed enhanced pulmonary histopathology following challenge with RSV. In such animals, there was approximately a 90% reduction in the number of infected cells compared to control unimmunized, RSV-challenged animals. Formalin-inactivated RSV vaccine-enhanced lung histopathology developed despite the effective elimination of virus and virus-infected cells suggesting that the enhanced pathology is the result of an exaggeration of normal immune mechanisms involved in clearance of virus infection, an aberrant immune response during infection, or both.

The sequences of the P mRNA and protein of strain 18537 of antigenic subgroup B of human respiratory syncytial virus were determined by sequencing cloned cDNAs of intracellular mRNA. Comparison with the corresponding sequences of the A2 strain of subgroup A showed that there was extensive sequence identity at both the nucleotide (80% identity) and amino acid (90% identity) levels. The P proteins contained a single divergent region (52% amino acid identity) flanked by highly conserved domains (96% identity). The previously observed differences in electrophoretic mobilities between the P proteins of subgroup A strains and certain subgroup B strains could not be attributed to differences in the polypeptide Mr of the primary translation product.

Posttranslational processing and cell surface expression were examined for three C-terminally truncated mutants of the G protein of respiratory syncytial virus expressed from engineered cDNAs. The truncated mutants, encoded by cDNAs designated G71, G180, and G230, contained the N-terminal 71, 180, and 230 amino acids, respectively, of the 298-amino-acid G protein. To facilitate detection of G71, which reacted inefficiently with G-specific antisera, we constructed a parallel set of cDNAs, designated G71/13, G180/13, and G230/13, to encode the same truncated species with the addition of a C-terminal 13-amino-acid reporter peptide which could be detected efficiently with an antipeptide serum. G71, G180, and G230 were expressed as species of Mr 7,500, 48,000, and 51,000, respectively, compared with 84,000 for parental G protein. The proteins encoded by G180 and G230, like parental G protein, contained both N-linked and O-linked carbohydrate. Also, the protein encoded by G71/13 appeared to be O glycosylated, showing that even this highly truncated form contained the structural information required to target the protein for O glycosylation. As for parental G protein, the estimated Mrs of the proteins encoded by G180 and G230 were approximately twice the calculated molecular weight of the polypeptide chain. Experiments with monensin showed that most of this difference between the calculated and observed Mr was due to posttranslational processing in or beyond the trans-Golgi compartment, presumably owing to the addition of carbohydrate or aggregation into dimers or both. Like parental G protein, all three truncated forms accumulated abundantly at the cell surface, and in each case the C terminus was extracellular. Thus, the N-terminal 71 amino acids of the G protein contained all the structural information required for efficient membrane insertion and cell surface expression, whereas the extracellular domain was dispensable for these activities. Cotton rats were immunized with recombinant vaccinia viruses expressing the G71, G180, G230, or parental G protein to compare their abilities to induce serum antibodies and resistance to challenge virus replication. The G71 and G180 recombinants failed to induce significant levels of G-specific antibodies or resistance to challenge, whereas the immunogenicity of G230 equaled or exceeded that of parental G protein. This suggested that the C-terminal 68 amino acids of the 236-amino-acid ectodomain do not contribute to the major epitope(s) of the G protein that is involved in inducing protective immunity.