Structure-based design of a quadrivalent fusion glycoprotein vaccine for human parainfluenza virus types 1-4.

Parainfluenza virus types 1-4 (PIV1-4) are highly infectious human pathogens, of which PIV3 is most commonly responsible for severe respiratory illness in newborns, elderly, and immunocompromised individuals. To obtain a vaccine effective against all four PIV types, we engineered mutations in each of the four PIV fusion (F) glycoproteins to stabilize their metastable prefusion states, as such stabilization had previously enabled the elicitation of high-titer neutralizing antibodies against the related respiratory syncytial virus. A cryoelectron microscopy structure of an engineered PIV3 F prefusion-stabilized trimer, bound to the prefusion-specific antibody PIA174, revealed atomic-level details for how introduced mutations improved stability as well as how a single PIA174 antibody recognized the trimeric apex of prefusion PIV3 F. Nine combinations of six newly identified disulfides and two cavity-filling mutations stabilized the prefusion PIV3 F immunogens and induced 200- to 500-fold higher neutralizing titers in mice than were elicited by PIV3 F in the postfusion conformation. For PIV1, PIV2, and PIV4, we also obtained stabilized prefusion Fs, for which prefusion versus postfusion titers were 2- to 20-fold higher. Elicited murine responses were PIV type-specific, with little cross-neutralization of other PIVs. In nonhuman primates (NHPs), quadrivalent immunization with prefusion-stabilized Fs from PIV1-4 consistently induced potent neutralizing responses against all four PIVs. For PIV3, the average elicited NHP titer from the quadrivalent immunization was more than fivefold higher than any titer observed in a cohort of over 100 human adults, highlighting the ability of a prefusion-stabilized immunogen to elicit especially potent neutralization.

Genome sequencing and phylogenetic analysis of 39 human parainfluenza virus type 1 strains isolated from 1997-2010.

Thirty-nine human parainfluenza type 1 (HPIV-1) genomes were sequenced from samples collected in Milwaukee, Wisconsin from 1997-2010. Following sequencing, phylogenetic analyses of these sequences plus any publicly available HPIV-1 sequences (from GenBank) were performed. Phylogenetic analysis of the whole genomes, as well as individual genes, revealed that the current HPIV-1 viruses group into three different clades. Previous evolutionary studies of HPIV-1 in Milwaukee revealed that there were two genotypes of HPIV-1 co-circulating in 1991 (previously described as HPIV-1 genotypes C and D). The current study reveals that there are still two different HPIV-1 viruses co-circulating in Milwaukee; however, both groups of HPIV-1 viruses are derived from genotype C indicating that genotype D may no longer be in circulation in Milwaukee. Analyses of genetic diversity indicate that while most of the genome is under purifying selection some regions of the genome are more tolerant of mutation. In the 40 HPIV-1 genomes sequenced in this study, the nucleotide sequence of the L gene is the most conserved while the sequence of the P gene is the most variable. Over the entire protein coding region of the genome, 81 variable amino acid residues were observed and as with nucleotide diversity, the P protein seemed to be the most tolerant of mutation (and contains the greatest proportion of non-synonymous to synonymous substitutions) while the M protein appears to be the least tolerant of amino acid substitution.

Origin of the inhibitory activity of 4-O-substituted sialic derivatives of human parainfluenza virus.

Human parainfluenza virus (hPIV) is a serious human pathogen causing upper and lower respiratory tract disease, yet there are no effective vaccines or therapies to control parainfluenza virus infections. Recently, we found that 4-O-substituted sialic derivatives have potent inhibitory activity against hPIV-1, whereas the anti-influenza inhibitor Zanamivir was less inhibitory. To elucidate the origin of the high potency inhibitory activities of these 4-O-substituted derivatives, we performed correlated fragment molecular orbital (FMO)-interfragment interaction energy (IFIE) analysis for hemagglutinin-neuraminidase (HN) glycoprotein complexes of hPIV with the derivatives and compared them with those for Zanamivir. We found key interactions between the inhibitors and the hPIV HN glycoprotein and identified important factors for the inhibitory activity. These theoretical results will be useful for the development of novel anti-hPIV drugs.

Loss of the N-linked glycan at residue 173 of human parainfluenza virus type 1 hemagglutinin-neuraminidase exposes a second receptor-binding site.

BCX 2798 (4-azido-5-isobutyrylamino-2,3-didehydro-2,3,4,5-tetradeoxy-d-glycero-d-galacto-2-nonulopyranosic acid) effectively inhibited the activities of the hemagglutinin-neuraminidase (HN) of human parainfluenza viruses (hPIV) in vitro and protected mice from lethal infection with a recombinant Sendai virus whose HN was replaced with that of hPIV-1 (rSeV[hPIV-1HN]) (I. V. Alymova, G. Taylor, T. Takimoto, T. H. Lin., P. Chand, Y. S. Babu, C. Li, X. Xiong, and A. Portner, Antimicrob. Agents Chemother. 48:1495-1502, 2004). The ability of BCX 2798 to select drug-resistant variants in vivo was examined. A variant with an Asn-to-Ser mutation at residue 173 (N173S) in HN was recovered from mice after a second passage of rSeV(hPIV-1HN) in the presence of BCX 2798 (10 mg/kg of body weight daily). The N173S mutant remained sensitive to BCX 2798 in neuraminidase inhibition assays but was more than 10,000-fold less sensitive to the compound in hemagglutination inhibition tests than rSeV(hPIV-1HN). Its susceptibility to BCX 2798 in plaque reduction assays was reduced fivefold and did not differ from that of rSeV(hPIV-1HN) in mice. The N173S mutant failed to be efficiently eluted from erythrocytes and released from cells. It demonstrated reduced growth in cell culture and superior growth in mice. The results for gel electrophoresis analysis were consistent with the loss of the N-linked glycan at residue 173 in the mutant. Sequence and structural comparisons revealed that residue 173 on hPIV-1 HN is located close to the region of the second receptor-binding site identified in Newcastle disease virus HN. Our study suggests that the N-linked glycan at residue 173 masks a second receptor-binding site on hPIV-1 HN.

Human parainfluenza virus type 1 phosphoprotein is constitutively phosphorylated at Ser-120 and Ser-184.

RNA-dependent RNA polymerases of single-stranded, negative-sense RNA viruses comprise a phosphoprotein (P) and a large protein. The constitutive phosphorylation of the P protein in these viruses is highly conserved, yet the functional significance of phosphorylation is enigmatic. To approach this problem, phosphorylation sites were determined in two closely related paramyxovirus P proteins. Sendai virus (SV) is a prototypic paramyxovirus. Previously, using a phosphopeptide mapping technique, the primary constitutive phosphorylation site of SV P protein was mapped to Ser-249. Phosphorylation at Ser-249 is dependent on the presence of Pro-250. Human parainfluenza virus type 1 (HPIV-1) P protein has 66% similarity to SV P protein and its predicted secondary structure is highly similar to that of SV P protein. However, there is no obvious conserved phosphorylation site in HPIV-1 P protein. Using the phosphopeptide mapping strategy, the constitutive phosphorylation sites of HPIV-1 P protein were mapped. The HPIV-1 P protein is primarily phosphorylated at Ser-120. Phosphorylation at Ser-120 is dependent on the presence of Pro-121. It also has a minor phosphorylation site at Ser-184. The sequence at Ser-184 does not match any consensus phosphorylation target site for the known kinases. Significantly, the P proteins from both viruses are constitutively and primarily phosphorylated at one serine and the phosphorylation of that serine is dependent on the presence of a proline on its carboxyl side.

The influence of imperfectly paired helices I and III on the catalytic activity of hammerhead ribozymes.

Several catalytic antisense RNAs directed against different regions of the genomic or antigenomic RNA of Sendai virus were constructed. All RNAs contained the same catalytic domain based on hammerhead ribozymes but some had deletions or mutations resulting in imperfect helices I and III. Pre-annealed substrate/ribozyme complexes were used to determine the rates of the cleavage process for the different ribozymes under single-turnover conditions. It was found that the sequence context surrounding the cleavable motif influenced the cleavage efficiencies. Deletions or mutations of nucleotides 2.1 or 15.1 and 15.2 according to the numbering system for hammerhead ribozymes of Hertel et al. destroyed catalytic activity. Deletions of nucleotide 2.2 or additional nucleotides in the helix I-forming region of the ribozyme did not destruct, but only reduced the cleavage efficiencies. Similar results were observed for a deletion of nucleotide 15.3. Simultaneous deletions within helices I and III resulted in alternative cleavage sites. The potential consequences for the specificity of the ribozyme reaction are discussed.

The carboxy-terminal domain of Sendai virus nucleocapsid protein is involved in complex formation between phosphoprotein and nucleocapsid-like particles.

The replicative unit of Sendai virus, the nucleocapsid, is composed of the viral genomic RNA and the viral proteins NP, P, and L. P and L proteins form a polymerase complex and are associated with the nucleocapsid core (NP/RNA complex). The NP protein has recently been shown to be composed of two domains. While the amino-terminal domain I encapsidates RNA and gives rise to the characteristic helical nucleocapsid structure, the precise function of the carboxy-terminal domain II remained unclear. It was however supposed to be involved in binding the polymerase complex. To test this hypothesis, we established a cosedimentation assay which detects complex formation between P protein and nucleocapsid-like (NC-like) particles. Four mutant NC-like particles all carrying amino acid deletions in domain II of the NP protein were tested. Complex formation was abolished after deletion of amino acids 426-497 or 456-524, while deletions of amino acids 400-415 or 414-439 had no influence on the interaction. The results indicate that domain II of NP protein is involved in binding P protein to nucleocapsids. The function and position of a putative P protein binding site in domain II are discussed.

Assignment of disulfide bridges in the fusion glycoprotein of Sendai virus.

The mature fusion (F) glycoprotein of the paramyxovirus family consists of two disulfide-linked subunits, the N-terminal F2 and the C-terminal F1 subunits, and contains 10 cysteine residues which are highly conserved at specific positions. The high level of conservation strongly suggests that they are indeed disulfide linked and play important roles in the folding and functioning of the molecule. However, it has not even been clarified which cysteine residues link the F2 and F1 subunits. This report describes our assignment of the disulfide bridges in purified Sendai virus F glycoprotein by fragmentation of the polypeptide and isolation of cystine-containing peptides and determination of their N-terminal sequences. The data demonstrate that all of the 10 cysteine residues participate in disulfide bridges and that Cys-70, the only cysteine in F2, and Cys-199, the most upstream cysteine in F1, form the interchain bond. Of the remaining eight cysteine residues clustered near the transmembrane domain of F1, the specific bridges identified are Cys-338 to Cys-347 and Cys-362 to Cys-370. Although no exact pairings between the subsequent four residues were defined, it seems likely that the most downstream, Cys-424, is linked to Cys-394, Cys-399, or Cys-401. Thus, we conclude that the cysteine-rich domain indeed contributes to the formation of a bunched structure containing at least two tandem cystine loops.

The human parainfluenza virus type-1 prototypic strain contains a heat-labile hemagglutinin-neuraminidase protein.

The virus yield of human parainfluenza virus type-1 (hPIV-1) in cultured cells at 38 degrees C is reduced more than 100-fold compared to 34 degrees C, while the virus yield of Sendai virus (SV, Enders strain), a murine parainfluenza virus type-1 with high homology to hPIV-1 was almost equal at both temperatures. To understand the basis for the differences in the temperature growth characteristics of the two viruses, we examined the heat-stability of hPIV-1 and SV glycoproteins expressed from cDNAs by pulse-chase experiments. The hemagglutinin-neuraminidase (HN) protein of hPIV-1 was stable after a 6-h chase at 34 degrees C, while at 38 degrees C prominent protein degradation was observed starting at 3 h chase and by 6 h HN was reduced by 65%. In contrast, SV HN protein was stable at both 34 and 38 degrees C. The other hPIV-1 glycoprotein, the fusion (F) protein was stable at both temperatures. To identify the amino acids which are responsible for the heat-lability of hPIV-1 HN, mutant HN proteins were constructed by site-directed mutagenesis. Mutant hPIV-1 HN which had substitutions at positions 461 and 462 became heat-stable at 38 degrees C. These data indicate amino acids around 461 are responsible for the heat-lability of the wild type hPIV-1 HN protein and the reduced yield of the virus at 38 degrees C.

Fusion of Sendai virus with model membranes as affected by dehydrating agents and pH: correlation between viral envelope hydrophobicity and fusion activity.

We have studied the importance of hydrophobic interactions in the fusion activity of Sendai virus. Viral fusion activity towards model lipid membranes was monitored using a fluorescence dequenching assay. Fusion was dependent on the composition of the target membrane and was more rapid and extensive at low pH. Dehydrating agents such as dimethylsulfoxide and dimethylsulfone greatly enhanced the initial rate and extent of the fusion process, the effect of dimethylsulfone doubling that of dimethylsulfoxide. However, dehydrating agents had a less pronounced stimulatory effect at low pH. Using the fluorescent probe ANS (1-aminonaphthalene-8-sulfonate) we have shown that low pH by itself involves a marked increase in the viral envelope hydrophobicity, probably resulting in the observed lower stimulation of viral fusion activity by dehydrating agents. Therefore, there seems to be a strong correlation between viral envelope hydrophobicity and viral fusion activity.

Fusogenic properties of Sendai virosome envelopes in rat brain preparations.

Sendai virosomes were characterized with respect to their ability to bind to, fuse with, and introduce substances into several rat brain preparations. Encapsulation efficiency for Sendai virosomes was enhanced but binding to cerebral cortical P2 preparations was attenuated by addition of bovine brain phosphatidylcholine during reconstitution. A higher percentage of Sendai virosomes than phosphatidylcholine liposomes appeared to bind to, fuse with and subsequently deliver [14C]sucrose into osmotically labile pools of the P2 preparation. Fusogenic activity was estimated by measuring dequenching of fluorescently labelled N-NBD-phosphatidylethanolamine. More virosomally encapsulated [14C]sucrose was bound to the P2 fraction than introduced into osmotically labile organelles, and the fraction of vesicles undergoing fusion was intermediate between these two values. Non-encapsulated [14C]sucrose did not bind to and was not taken up by the P2 fraction in a quantifiable manner. Virosomal envelopes also bound to primary cultures of rat brain neurons and glia in an apparently saturable manner. Addition of increasing amounts of the adenoassociated virus-derived vector pJDT95 increased encapsulation efficiency, and virosomes reconstituted in the presence of 60 micrograms DNA retained most of their binding activity (5.4% of total label) compared to those containing [14C]sucrose alone (8.4%). These data indicate that Sendai virosomes may be useful in the delivery of substances into brain-derived tissues, potentially for the modulation of gene expression and neurotransmission.

Effect of glycosaminoglycans and PEG on fusion of Sendai virus with phosphatidylserine vesicles.

The fusion of Sendai virus with phosphatidylserine vesicles was monitored by a pyrene-phosphatidylcholine fluorescence assay. A strong influence of pH and ionic strength on the extent of fusion was observed. The negatively-charged polymers (dextran sulfate, heparin and chondroitin sulfate) inhibited the ability of the viruses to fuse with the liposomes. The extent of inhibition, for a given amount (w/v) of the polymers, was the greatest for dextran sulfate followed by heparin and chondroitin sulfate. The extent of inhibition depended on the pH and ionic strength of the solution; the lower the pH of the solution, the more effective the fusion inhibition by the polymers. The molecular weight of dextran sulfate (DS) influenced the inhibition effect, i.e., DS with higher molecular weight exhibited a stronger inhibition effect. The presence of sodium sulfate, even in excess concentration, had no inhibitory effect on fusion. On the other hand, PEG had an opposite effect on fusion compared to the negatively-charged polymers, and it decreased their inhibition effect when both were present in the same media. It is concluded that the inhibition of the fusion activity of Sendai virus results from the adherence of negatively-charged polymers to the virus surface preventing close contacts between the virus and liposome surface.

Effect of anionic polymers on fusion of Sendai virus with human erythrocyte ghosts.

The effect of anionic polymers (dextran sulfate, heparin and chondroitin sulfate) on fusion of Sendai virus with erythrocyte ghosts was studied. The effect of pH on the activity of these anionic polymers was also investigated. In order to examine the interaction of such polymers with the Sendai virion and erythrocyte ghost surfaces, the binding of virions to erythrocyte ghosts and the aggregation of virions and/or erythrocyte ghosts were also measured with respect to the same parameters. It was found that the anionic polymers suppressed the fusion of Sendai virus with erythrocyte ghosts. The order of effectiveness of the polymers in suppression was dextran sulfate greater than heparin greater than chondroitin sulfate, for the application of a same quantity (weight/ml) of the polymers. The lower the pH of the suspending medium, the more effective were the polymers in suppressing virion-erythrocyte ghost aggregation and fusion. The suppression of fusion was dependent on the concentration of the polymers applied: the higher the concentration of the polymer applied, the more the suppression was observed. Evidence from binding studies, turbidity measurements and electrophoretic mobility measurements indicates that the anionic polymers interact preferentially with the virion surface.

Sendai virus M protein is found in two distinct isoforms defined by monoclonal antibodies.

The use of a monoclonal antibody defines a subset of Sendai virus M protein representing about 30% of total. This M protein acquires, during the hour following synthesis, an epitope not present on the bulk of M. This epitope maturation is observed in acutely as well as in persistently infected cells. It takes place in vivo in absence of other viral proteins, but it is not observed when the protein is synthesized in a reticulocyte lysate. Epitope maturation does not appear to result from phosphorylation, acylation or disulfide bond formation. If immunofluorescent staining seems to indicate a preferential association of this subset of M protein with nucleocapsids, this is not confirmed by immunogold staining or by nucleocapsid isolation. Incubation of cytoplasmic extracts or of purified M protein in conditions which do not favor M to M protein association results in a relative increase of M protein carrying the maturing epitope. It is concluded that M protein exists in two distinct isoforms.

Absorption spectra of viral components of Sendai virus in the wavelength region from 130 to 320 nm.

Using synchrotron radiation as a light source, the absorption spectra of purified viral components of the Sendai virus, i.e. messenger RNA, lipids, spike (envelope) proteins, reconstructed envelopes, core proteins and whole virions, were obtained in the wavelength region 130-320 nm by measuring the transmission of thin films. Viral (messenger) RNA two peaks at 260 and 190 nm, and a large increase below 160 nm. The absorption spectrum of lipids exhibited a broad peak at 190 nm and a very sharp increase below 160 nm. With spike proteins, a slight peak at 280 nm and a shoulder at 230 nm were observed in addition to a sharper peak at 190 nm and a rather slow increasing absorption below 160 nm. Reconstructed envelopes showed the features of a combination of lipids and proteins. The absorption spectra of core proteins and whole virions exhibited similar characteristics to spike proteins. Conventional UV data were also obtained in the wavelength range 210-320 nm with RNA and lipids. The UV and synchrotron radiation data were in good agreement in terms of the mass absorption coefficients. The molecular splitting of spike proteins was also examined. Proteins gave more diffuse reflection than their subunits, causing a reduction in absorption. This was explained by a loss of transparency with increasing molecular weight.

Localization of P, NP, and M proteins on Sendai virus nucleocapsid using immunogold labeling.

The distribution of NP, P, and M proteins on Sendai virus nucleocapsids purified from cells and virions were studied by immunogold staining using monoclonal antibodies. NP molecules were found uniformly along the entire length of both cytosol and virion derived nucleocapsids. This observation is in accord with the earlier proposals that NP molecules maintained the structural integrity of the nucleocapsid. The distribution of P in nucleocapsids derived from the cytosol differed from the distribution in those originating from virions. In nucleocapsids derived from the cytosol, P molecules occurred in 4 to 10 discreet clusters at varying locations along the length of the nucleocapsid. In contrast, on nucleocapsids derived from virions, P molecules were uniformly distributed over the entire length of the nucleocapsid. These observations suggest that the distribution of P depends on the functional state of the nucleocapsid. The occurrence of P clusters at different locations on intracellular nucleocapsids indicates that P is a mobile molecule; this suggestion is consistent with P's role in viral RNA synthesis. The distribution of the matrix (M) protein also depended on where the nucleocapsids were derived from. Large quantities of M protein were found along the entire length of nucleocapsids derived from the cytosol, while in virion nucleocapsids, many fewer molecules of M were observed. The large amounts of M on the nucleocapsids originating from the cytosol supports the hypothesis that M protein mediates the recognition between the nucleocapsid and the envelope glycoproteins.