|Home | About | Journals | Submit | Contact Us | Français|
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 106.5 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.
Paramyxoviruses have promise as vaccine vectors, particularly for intranasal immunization. For example, intranasal immunization of hamsters with a recombinant human parainfluenza virus type 3 (HPIV3) expressing the measles virus hemagglutinin protein induced a high titer of measles virus-neutralizing serum antibodies (5). Immunization of hamsters with a recombinant HPIV3 expressing the hemagglutinin-neuraminidase (HN) proteins of HPIV1 and HPIV2 induced a high titer of serum antibodies against both HPIV1 and HPIV2 and induced resistance to challenge with these viruses (42). Immunization of the respiratory tract of rhesus monkeys with a recombinant attenuated bovine-human (BH) chimeric PIV3 expressing the attachment (G) and/or fusion (F) glycoprotein of human respiratory syncytial virus (RSV) induced high titers of RSV-neutralizing serum antibodies (35). Indeed, the BHPIV3 construct expressing the RSV F protein is scheduled to be evaluated clinically as a combined RSV/HPIV3 pediatric vaccine (41). Immunization of African green monkeys with the attenuated BHPIV3 vector expressing the spike (S) protein of severe acute respiratory syndrome coronavirus (SARS-CoV) induced a serum neutralizing antibody response against SARS-CoV and protected animals against a subsequent challenge with a high dose of SARS-CoV (1). Finally, intranasal immunization of guinea pigs with HPIV3 expressing the glycoprotein of Ebola virus resulted in a sterilizing immune response against a subsequent challenge with a high dose of Ebola virus (A.Bukreyev et al., unpublished data). HPIV3 can accommodate and express at least three inserts with an aggregate size of at least 7.5 kb and can maintain them stably for multiple passages in vitro (1, 5, 38). However, the use of one or two inserts will likely be optimal because increased insert number and size can result in overattenuation (39).
Although attenuated human paramyxoviruses appear to have excellent properties as intranasal vaccine vectors, they probably would be effective only in individuals who have not been previously exposed to the vector. This is because existing host immunity to the vector would restrict its replication and reduce its immunogenicity. For example, wild-type (wt) RSV infected only 50% of seropositive adult volunteers, and attenuated RSV mutants infected only 10 to 33% (depending on the strain) of volunteers (14). In another study, only 8 to 20% of seropositive adult volunteers were infected with wt HPIV3 or its attenuated mutants (3). Even the antigenically divergent BPIV3 was highly restricted in replication in the respiratory tracts of adults and children (3, 15). Therefore, although the human paramyxoviruses are excellent candidates for use as vectors in the immunologically inexperienced pediatric population, they probably will not be effective in adults because nearly all will have previously been infected with these common pathogens and replication of the vaccine vector would be highly restricted, leading to reduced or absent immunogenicity. This problem likely would hold for any viral vector that is based on a common human pathogen, such as measles virus and common strains of adenovirus. To overcome this limitation, we have examined the infectivity, safety, and immunogenicity properties of Newcastle disease virus (NDV), an avian paramyxovirus that is distinct serotypically (i.e., does not experience significant cross-neutralization) from common human paramyxoviruses.
NDV is a member of the Avulavirus genus of subfamily Paramyxovirinae of family Paramyxoviridae, a genus that does not include any known natural pathogens of humans. NDV isolates can be divided to three groups based on their degree of virulence in birds: (i) lentogenic strains, which cause mild or nonapparent infections that are largely limited to the respiratory tract and which include strains presently in use as live vaccines; (ii) mesogenic stains, which cause systemic infections of intermediate severity; and (iii) velogenic strains, which cause systemic infections with high mortality rates. The genome of NDV is a single negative-sense strand of RNA of15,162 to 15,192 nucleotides (nt) (19) (for examples, see GenBank entries AF077761, AY935499, and AY562989) that encodes seven major structural proteins: nucleoprotein (N), phosphoprotein (P), matrix protein (M), fusion protein (F), hemagglutinin-neuraminidase (HN), and large polymerase protein (L). One of the major determinants of virulence is the structure of the cleavage site of the F protein, where cleavage by a cellular protease must occur to generate the active form of the protein consisting of the F1 and F2 subunits. Cleavage sites that contain the recognition sequence for the intracellular protease furin (R-X-K/R-R) are associated with increased virulence, whereas sites that contain fewer arginine/lysine residues must be cleaved by extracellular secretory proteases and are associated with reduced virulence (6, 34). The HN protein also plays an important role in tropism and pathogenicity (11), and other determinants of pathogenicity remain to be identified. The present study involved two pathogenically distinct strains of NDV: the lentigenic vaccine strain LaSota (NDV-LS) and the mesogenic strain Beaudette C (NDV-BC). The F protein cleavage site of NDV-LS is SGGGRQGR↓LIG and thus lacks the furin motif, whereas that of NDV-BC is SGGRRQKR↓FIG and contains this motif (underlined).
Recently, recombinant NDV was used to express protective antigens of simian immunodeficiency virus and influenza virus and was shown to induce an immune response in mice (24, 25). However, rodent models are not reliable surrogates alone for humans and are not reliable predictors alone of the attenuation, immunogenicity, and protective efficacy of candidate human vaccines. In principle, any virus that can be manipulated by recombinant DNA technology and engineered to express foreign epitopes or antigens has the potential for use as a vaccine vector. In practice, it is essential to demonstrate satisfactory levels of replication, attenuation, safety, and immunogenicity in an experimental in vivo setting that closely models the intended use with regard to the phylogenetic and anatomical features of the host and the dose, route of administration, and permissiveness to vector replication. In the present study, we evaluated NDV-LS and NDV-BC as vectors to express the HN protein of HPIV3 as a test antigen. These vectors were evaluated for replication, attenuation, and immunogenicity, following intranasal immunization of two species of nonhuman primates, models that have close phylogenetic and anatomical similarity to humans and should be reasonably good surrogates for humans with regard to vector replication.
NDV-LS and NDV-BC are highly related viruses (97% nucleotide sequence identity) and were modified by a common strategy. The genome of each virus was modified to create an XbaI site in the downstream noncoding region of the Pgene (Fig. (Fig.1).1). This involved two C-to-A (positive-sense) nucleotide substitutions at positions 3185 and 3187 in the complete antigenome sequence to create the XbaI site TCTAGA. These substitutions were performed in two PCR steps. First, a cDNA fragment corresponding to positions 2,354 to 3,187 of the NDV antigenomic cDNA was amplified and mutagenized by PCR to create an XbaI site. The forward PCR primer was GATCGAGCTCCGCGGAAACAGTCC AGGAAAGACC, with the SacI site underlined and the SacII site italicized. The reverse primer was ACTCTAGAGAGGATTGCCGCTTGGAGAG, with the XbaI site italicized and the two mutated nucleotides underlined. The PCR product was digested with SacI and XbaI and inserted into the SacI-XbaI window of the plasmid pGem-7Zf(+) (Promega Corporation, Madison, WI). Second, a cDNA fragment corresponding to positions 3182 to 3760 of the NDV antigenome was PCR amplified with the forward primer AGTCTAGATTCCTCA GCCCCACTGAATG, with the XbaI site in boldface type and the two mutated nucleotides underlined, and the reverse primer CATGGACGTCACGTGCTT GACTGCATTCACTGATGAG, with the AatII site underlined and the PmlI site italicized. The PCR product was digested with XbaI and AatII and inserted into the XbaI-AatII window of the plasmid constructed at the previous step. This resulted in a subclone of the SacII-PmlI fragment (NDV genome positions 2354 to 3760) of the NDV antigenome containing the newly created XbaI site. Next, a cDNA of the HPIV3 HN open reading frame (ORF) was amplified by PCR to attach an NDV gene junction (gene end-intergenic-gene start motifs) (Fig. (Fig.1)1) and an XbaI site to its upstream end and an XbaI site to its downstream end (Fig. (Fig.1).1). The forward primer was CTTGAATCTAGATTAGAAAAAACACGGGTAGAACGCCACCATGGAATACTGGAAGCATAC, with the XbaI site italicized, the NDV gene end transcription signal underlined, the C intergenic nucleotide in boldface, the NDV gene start transcription signal underlined, and a 7-nt-long spacer that was designed for efficient translation (17) in boldface. The gene start and gene end signals were identical to those found naturally for the most of the genes of NDV-LS and NDV-BC, and the intergenic sequence is identical to that between the M and F genes. The reverse primer was TACTTGTCTAGATATGATTAAGACGTCGAAAAAGGAA, with the XbaI site italicized and a 5-nt-long spacer to maintain the rule of six (16) in boldface. The PCR product was digested with XbaI and inserted into the SacII-PmlI subclone using the newly created XbaI site. Finally, this SacII-PmlI fragment bearing the HN insert was substituted into the full-length NDV antigenomic cDNA. The DNA sequences in all areas subject to in vitro polymerase reactions were confirmed by nucleotide sequencing.
Recombinant virus was recovered from the NDV-LS/HN and NDV-BC/HN antigenomic cDNAs by transfection into HEp-2 cells in parallel with support plasmids expressing the N, P, and L proteins, as previously described (10, 18). The parental NDV-LS and NDV-BC viruses also were recombinantly derived. The NDV recombinants were propagated in embryonated eggs or monolayers of DF-1 chicken fibroblast cells with Dulbecco's modified Eagle medium (Invitrogen) containing 2.5% of fetal bovine serum (Invitrogen, Carlsbad, CA). For NDV-LS and its derivatives, the medium was supplemented with 5% chicken allantoic fluid as a source of proteases for cleavage of the F protein. Virus titers were determined by plaque assay of DF-1 cells under 0.9% (wt/vol) methylcellulose in the presence of 2.5% fetal bovine serum and, for NDV-LS and NDV-LS/HN, 5% chicken allantoic fluid. The cells were incubated at 37°C for 4 to 5 days, fixed with 80% methanol, and stained with crystal violet. In some experiments, the specificity of the plaques was confirmed by immunostaining with a cocktail of mouse monoclonal antibodies specific to NDV HN protein (30), which were detected using sheep anti-mouse immunoglobulin G (IgG) labeled with horseradish peroxidase.
DF-1 or Vero cells were infected with the recombinant NDVs or their parental viruses at a multiplicity of infection (MOI) of 2 PFU. For comparison, LLC-MK2 or Vero cells were infected in parallel with 2 50% tissue culture infective doses (TCID50) of HPIV3. Twenty-four hours later, the cells were harvested and used for isolation of total intracellular RNA with the RNeasy mini kit (QIAGEN, Inc., Valencia, CA) according to the manufacturer's recommendations. The RNA was analyzed by Northern blot hybridization (7) with a double-stranded DNA (dsDNA) probe prepared from a cDNA of the HN gene using the Megaprime DNA Labeling system (Amersham Biosciences Corp., Piscataway, NJ) and [32P]dCTP. Alternatively, cells were harvested 24 h postinfection and lysed under denaturing and reducing conditions, and the clarified supernatant was subjected to gel electrophoresis on 4 to 12% bis-Tris acrylamide gels (NuPage protein electrophoresis system; Invitrogen, Mountain View, CA) according to the manufacturer's recommendations. The proteins were analyzed by Western blotting using the WesternBreeze immunodetection kit (Invitrogen) with rabbit hyperimmune anti-HPIV3 serum (provided by M. Skiadopoulos). To purify virus particles for analysis, the medium was harvested ~48 h postinfection from cells that had been infected with NDV or HPIV3 as described above. The medium was clarified by low-speed centrifugation, and the virus was concentrated by ultracentrifugation through a 25% sucrose solution at 130,000 × g at 4°C for 2 h, resuspended in a small volume of medium, and further purified by centrifugation on a 30 to 60% (wt/vol) discontinuous sucrose gradient as above, after which the virus band at the interface was isolated. The virion preparations were analyzed directly by silver staining using the SilverQuest kit (Invitrogen) or by Western blotting as described above.
To compare the sensitivity of recombinant NDV or HPIV3 to neutralization by HPIV3-specific antibodies, 50 PFU of NDV or 50 TCID50 of HPIV3 in a volume of 75 μl of Dulbecco's modified Eagle medium (Invitrogen) containing 10% (vol/vol) of a commercial preparation of guinea pig complement (Cambrex Corporation, East Rutherford, N.J.) was combined with 75 μl of various dilutions in the same medium of convalescent-phase serum from an African green monkey that had been infected with HPIV3, or of serum from a rabbit that had been hyperimmunized with sucrose gradient-purified HPIV3 virions, or of serum from an HPIV3-naïve African green monkey or rabbit. The mixtures were incubated for 1 h at 37°C, and the residual infectious virus was quantified by plaque assay of DF-1 cells in the case of NDV or by TCID50 titration of LLC-MK2 cell monolayers in the case of HPIV3. To evaluate neutralization of the NDV recombinants or HPIV3 by NDV-specific antibodies, 105 PFU of recombinant NDV or 105 TCID50 of HPIV3 virus was mixed with an equal volume of various dilutions of a chicken anti-NDV serum (Charles River Laboratories, North Franklin, CT), incubated at 37°C for 1 h, and the residual infectivity was assayed by TCID50 titration of Vero cell monolayers.
Rhesus monkeys (2 to 4 years old; body weight, 3.0 to 3.9 kg each) were obtained from LABS of Virginia (Morgan Island, SC); African green monkeys (juveniles, 4.4 to 7.7 kg each) were obtained from a free-ranging colony on the island of St. Kitts. The animals were confirmed to be seronegative for HPIV3 and NDV by hemagglutination inhibition (HAI) assay. In the HAI assay specific for HPIV3 antibodies, we used guinea pig erythrocytes and HPIV3 strain JS as the indicator virus; for NDV antibodies, we used turkey erythrocytes and parental NDV-BC as the indicator virus. The monkeys were infected as described previously (1) by combined intranasal and intratracheal routes with 106.5 PFU at 1 ml per site on day 1 and day 28. Nose-throat swabs were taken daily on days 0 to 10 and on day 12 following the first dose and on days 0, 2, 4, 6, and 8 following the second dose. Tracheal lavages were performed on days 2, 4, 6, 8, 10, and 12 following the first dose and on days 2, 4, 6, and 8 following the second dose. These samples were snap-frozen, and virus titers were determined later by plaque assay of DF-1 cells as described above. Clinical observations of the animals were performed daily after the first dose on days 0 to 10 and on day 12 and after the second dose on days 0, 2, 4, 6, and 8. In a separate experiment, monkeys were euthanized on day 4 after infection. At necropsy, tissue blocks of 1 cm3 from each lobe of the lungs were frozen, homogenized later in the presence of Leibovitz's L-15 medium (Invitrogen), and analyzed for the virus by plaque assay. The primate experiments were performed at a site approved by the Association for Assessment andAccreditation of Laboratory Animal Care International with a protocol approved by the Animal Care and Use Committee of the National Institute of Allergy and Infectious Diseases.
Full-length antigenomic cDNAs of vaccine strain NDV-LS and mesogenic strain NDV-BC were each modified by the introduction of an XbaI site into the downstream nontranslated region of the P gene, immediately upstream of the P gene-end transcription signal (Fig. (Fig.1).1). This site was then used for the insertion of a transcription cassette containing the ORF encoding the HN glycoprotein of HPIV3. The HPIV3 HN protein was chosen as the foreign insert because we have extensive historic data involving the immune response against this protein expressed by wild-type and attenuated HPIV3 in nonhuman primates and humans. This foreign ORF was engineered to be preceded by an NDV gene junction consisting of (in upstream-to-downstream order) a gene end signal, an intergenic nucleotide, and a gene start signal (Fig. (Fig.1).1). Insertion of this cassette into the XbaI site placed the HN ORF under the control of its own set of NDV transcription signals as a separate added gene between the P and M genes. The construct was designed to conform to the rule of six, which is the requirement that the nucleotide length of the genome be an even multiple of six for efficient RNA replication to occur (16, 31). This insert increased the length of the NDV genome by 1,758 nt (12%), from 15,186 to 16,944 nt.
The recombinant viruses were recovered by cotransfection of plasmids encoding each full-length antigenomic RNA together with the NDV N, P, and L support proteins in HEp-2 cells as described previously (10, 18). The correct structure of the HPIV3 HN insert in each recovered virus was confirmed by reverse transcription-PCR amplification and sequence analysis of the entire insert. The viruses were propagated in chicken embryos and titrated; the yields of the viruses were 8.6 × 107 (NDV-BC/HN) and 4.0 × 107 (NDV-LS/HN) PFU/ml of allantoic fluid, which are somewhat lower than typical concentrations of parental viruses without inserts in chicken embryos and are consistent with previously published data demonstrating moderate attenuation of replication due to insertion of a foreign gene (10, 18).
To confirm expression of the HPIV3 HN insert, DF1 chicken cells were infected with NDV-LS/HN, NDV-BC/HN, and their parental viruses lacking the insert at an MOI of 2 PFU per cell. In parallel, LLC-MK2 rhesus monkey cells were infected with wild-type HPIV3 at multiplicity of 2 TCID50 per cell. Twenty-four hours postinfection, cells were harvested, and total intracellular RNA was isolated. The RNA was subjected to electrophoresis in formaldehyde agarose gel and analyzed by Northern blot hybridization with a double-stranded DNA (dsDNA) probe prepared from a cDNA of the HPIV3 HN gene (Fig. (Fig.2A).2A). This showed that NDV-BC/HN and NDV-LS/HN (Fig. (Fig.2A,2A, lanes 2 and 4, respectively) each expressed a major RNA of the appropriate size to be a monocistronic mRNA of the HN gene cassette, as well as a larger mRNA that was of the appropriate size to be a readthrough of the inserted HPIV3 HN gene with either the upstream NDV P gene or the downstream NDV M gene; this RNA was not analyzed further. HPIV3 expressed a monocistronic HN mRNA of similar size (Fig. (Fig.2A,2A, lane 5); its apparent size was slightly larger than that expressed by either NDV construct, which is consistent with its calculated length being 74 nt larger due to a longer noncoding sequence. Interestingly, the NDV-BC/HN vector expressed a substantially higher level of HN mRNA at 24 h postinfection than did NDV-LC/HN or HPIV3. We also compared these viruses side by side in Vero African green monkey cells with the same inocula mentioned above and the same 24-h period of infection, followed by Northern blot analysis with the HN-specific probe (Fig. (Fig.2B).2B). The same spectrum of mRNAs described above was observed, and the higher level of expression for the NDV-BC/HN virus that was noted above was also seen under these conditions.
To investigate the intracellular expression of the HN protein, DF-1 cells were infected with 2 PFU per cell of NDV-BC, NDV-LS, and their HN-bearing derivatives. In parallel, LLC-MK2 cells were infected with 2 TCID50 per cell of HPIV3 or were mock infected. The cells were harvested 24 h postinfection and subjected to Western blot analysis with rabbit antiserum raised against sucrose gradient-purified HPIV3: the results for NDV-BC and NDV-BC/HN are shown in Fig. Fig.3A.3A. Lysates of HPIV3-infected cells contained strong bands corresponding to the P, N, and M proteins (Fig. (Fig.3A,3A, lane 4). In addition, the HN protein was detected as a minor band, which likely reflects poor reactivity of the antiserum due to the highly conformational and denaturation-sensitive nature of the epitopes of HN (4, 23). Although the HPIV3 HN protein was only faintly discernible in lysates from HPIV3-infected cells, a band of the appropriate size was evident in lysates from DF-1 cells infected with NDV-BC/HN (Fig. (Fig.3A,3A, lane 3) and NDV-LS/HN (not shown) and was absent (as expected) in lysates from cells infected with the parental NDV-BC (lane 2) and NDV-LS (not shown) viruses.
The insertion of genes encoding transmembrane proteins of heterologous viruses into genomes of negative-strand viruses sometimes results in the incorporation of these proteins into the virus particle (33). We investigated this possibility by purifying NDV-LS/HN, NDV-BC/HN, and their parental viruses on discontinuous sucrose gradients and analyzing the partially purified preparations by silver staining (not shown) and Western blot analysis (the results for HPIV3 and the NDV-BC-based viruses are shown in Fig. Fig.3B).3B). In this case, the HPIV3 HN protein was detected by Western blot analysis as a minor band in the virus preparations of HPIV3 (Fig. (Fig.3B,3B, lane 4) as well as for NDV-BC/HN (lane 3) and NDV-LS/HN (not shown) but not NDV-BC (lane 2) or NDV-LS (not shown). It was of interest to estimate the relative abundance of the HPIV3 HN band in the preparations of NDV-BC/HN compared to that in HPIV3. However, the HN band in the silver-stained gel of the NDV preparation was obscured by background bands, precluding direct comparison of silver-stained material (not shown). Therefore, we estimated this abundance by using an indirect method. Specifically, we measured the intensity of the HN bands in the NDV-BC/HN and HPIV3 lanes in the Western blot shown in Fig. Fig.3B.3B. Each value was then normalized relative to the intensity of N protein in the corresponding lane of the silver-stained gel (not shown). This value was further normalized to account for the difference in size between the N proteins of HPIV3 and NDV, thus converting the value into a relative molar amount. These calculations indicated that the relative molar amount of HN in the NDV-BC/HN preparation compared to HPIV3 was 15%.
The observed incorporation of the HPIV3 HN protein into virions of the NDV vectors raised the question as to whether this could confer sensitivity to neutralization by antibodies specific to HPIV3. To test this possibility, we incubated aliquots that each contained 50 PFU of the HN-bearing NDV vectors, their parental viruses lacking the insert, or 50 TCID50 of HPIV3 with 1:20 to 1:1,280 dilutions of HPIV3-positive African green monkey serum, a serum of a rabbit hyperimmunized with gradient-purified HPIV3 virions, or a nonimmune control serum of monkeys or rabbits in the presence of guinea pig complement for 1 h at 37°C. The residual infectious virus in each aliquot was quantified by plaque assay of DF-1 cell monolyers (NDV) or TCID50 titration on LLC-MK2 cells (HPIV3). Whereas either HPIV3-specific antibody preparation completely neutralized HPIV3 at a dilution of 1:1,280, no neutralization was observed for the recombinant NDVs bearing the HPIV3 HN protein, even at a dilution of 1:20 (not shown). Thus, the level of incorporation of the HPIV3 HN protein into the virus particle of the NDV vectors apparently was insufficient to confer a significant sensitivity to HPIV3-neutralizing antibodies, even in the presence of complement.
The incorporation of small amounts of HPIV3 HN into the virion particles of recombinant NDVs had the potential to contribute to initiating infection. To test this possibility, we incubated aliquots that each contained 105 PFU per well of the HN-bearing NDV vectors or their parental viruses lacking the insert, or 105 TCID50 of HPIV3, with 1:10 to 1:160 dilutions of chicken anti-NDV serum in the presence of guinea pig complement for 1 h at 37°C. Following incubation, the residual infectious virus was quantified by TCID50 titration of Vero cells that effectively support replication of both NDV and HPIV3. We observed a complete neutralization of both NDV-BC and NDV-BC/HN, whereas (as expected) the NDV-specific serum did not neutralize HPIV3. This suggested that the small amount of HPIV3 incorporated into the NDV virion did not confer NDV-independent ability to infect susceptible cells.
We evaluated the replication and immunogenicity of NDV-BC/HN in African green monkeys and rhesus monkeys and of NDV-LS/HN in African green monkeys (Table (Table1).1). Animals in groups of four were immunized by the combined intranasal and intratracheal routes with 106.5 PFU of virus per site on day 0. A second identical immunization was administered on day 28 to investigate whether the immune response to the first dose would interfere with the replication and immunogenicity of the second dose and whether the second dose would result in a boosting effect for the insert-specific immune response. Virus replication was monitored by collecting respiratory secretions by combined nose-throat (nasopharyngeal) swabs on days 1 to 10 daily and day 12 following the first dose and on days 0, 2, 4, 6, and 8 following the second dose and by tracheal lavage on days 2, 4, 6, 8, 10, and 12 following the first dose and on days 2, 4, 6 and 8 following the second dose. Analysis of the nasopharyngeal swabs and tracheal lavages for the presence of shed virus was performed by plaque assay; no detectable virus was found in either location for any virus except for a very low (5 PFU/ml) titer in a nasal swab of one African green monkey on day 2 after the first dose of NDV-BC/HPIV3-HN (Table (Table11).
In a second experiment, we infected four African green monkeys by the combined intranasal and intratracheal routes with 106.5 PFU of the parental NDV-BC virus per site (Table (Table2).2). On day 4, the animals were euthanized, and a total of 10 1-cm3 (each) tissue samples were taken from multiple lobes of the lungs. These were homogenized, and virus titers were determined by plaque assay. For each set of 10 samples representing a single animal, 2 or 3 samples were positive for infectious NDV, with titers of up to 103.4 PFU per g (Table (Table2).2). This suggested that the NDV recombinants were replicating at low levels at scattered locations within the lungs. Rhinorrhea, cough, sneeze, or other disease signs were not observed after immunization with any recombinant NDV.
Sera were collected from African green monkeys infected with NDV-BC/HN or NDV-LS/HN and rhesus monkeys infected with NDV-BC/HN on days 0 (prior to the first infection), 28 (prior to the second infection), and 56 (28 days following the second infection) (Table (Table3).3). The serum antibody response to the NDV vector was evaluated by an HAI assay against NDV. NDV-LS/HN induced a very low titer of serum NDV-HAI antibodies (1.8 log2, or 1:3.5) after the first dose, whereas NDV-BC/HN induced a moderate level of serum NDV-HAI antibodies (4.5 log2, or 1:23). After the second immunization, by day 56, the level of NDV-specific serum antibodies markedly increased in both groups (5.8 to 6.8 log2, or 1:56 to 1:111), with that for NDV-LS/HN remaining slightly lower.
Next, the HPIV3-specific antibody response was determined by HAI (Table (Table3).3). HPIV3-specific antibodies were not detected on day 0, whereas on day 28 a high titer of antibody was detected in both species of monkeys and for both NDV vectors, in the range of 7.0 to 11 log2, or 1:388 to 1:446. After the second immunization, the level of antibody titer on day 56 further increased to a mean log2 titer of 10.5 (1:1,448) in NDV-LS/HN-immunized African green monkeys, 11.1 log2 (1:2,195) in NDV-BC/HN-immunized African green monkeys, and 11.7 log2 (1:3,327) in NDV-BC/HN-immunized rhesus monkeys.
For comparison, we assayed parallel sera from African green monkeys and rhesus monkeys that had been infected in a previous experiment with 106 or 105 TCID50 of wt HPIV3, respectively, by the combined intranasal and intratracheal routes. In this previous experiment, sera were obtained 56 and 28 days postinfection, respectively. The mean antibody titers from African green monkeys or rhesus monkeys were 1:1,261 and 1:1,448, respectively. Thus, a single dose of either NDV strain expressing the HPIV3 HN protein induced a level of HPIV3-specific HAI antibodies that was approximately threefold lower than that induced by HPIV3, whereas the second dose resulted in titers that exceeded that of the HPIV3-infected animals by nearly twofold.
This study provided evidence that NDV is a promising intranasal vaccine vector for human use. A number of previous studies showed that human parainfluenza viruses are attenuated and highly immunogenic in experimental animals that are seronegative for the vector (see the introduction). These human viruses have considerable promise as vectors for immunization of the immunologically naive pediatric population. However, the adult population characteristically is immunologically experienced for these common pathogens and thus likely will be refractory to successful immunization (see the introduction). This obstacle could be surmounted by the identification of a vector that is serotypically distinct from common pathogens of humans; that is infectious, attenuated, and immunogenic in primates; and that replicates efficiently in a cell substrate such as Vero cells that are suitable for vaccine manufacture. NDV appears to possess these desirable qualities.
Intranasal-intratracheal immunization with 106.5 PFU of NDV was well tolerated in two species of nonhuman primates, with no overt disease signs and, at most, very low levels of virus shedding detected in respiratory secretions. However, shedding of viruses from the respiratory tract does not always reveal the extent of pulmonary replication. For example, in nonhuman primates infected with SARS-CoV, consistently higher titers of virus were seen in lung, tracheal, or nasal turbinate tissue homogenates than in nose-throat swabs or tracheal lavage samples (22). Therefore, a second experiment was performed, in which direct examination of lung tissue from African green monkeys provided evidence of low, scattered virus replication. This high degree of attenuation suggests that NDV would be a very safe intranasal vector.
There was previous evidence suggesting that NDV can cause conjunctivitis in bird handlers exposed to the virus, resulting in seroconversion (20, 27). Although conjunctivitis could be produced in monkeys when lentogenic, mesogenic, or velogenic strains of NDV were instillated onto traumatized conjunctiva, multiple instillations of large doses of viruses over the intact conjunctiva failed to do so (2). This suggests that humans can potentially be infected with NDV but probably are not susceptible to serious disease. In addition, various strains of NDV have been administered parenterally to humans as an oncolytic agent in studies over several decades that sometimes employed high doses and sometimes involved immunosuppressed patients (26). These studies involved only small numbers of patients, did not monitor virus replication or tropism, and therefore do not provide complete guidance on the replication of NDV in humans. Nonetheless, this experience gives some historic evidence that NDV is well tolerated in humans even under the somewhat drastic condition of parenteral administration of a high dose to immunosuppressed individuals.
A single intranasal immunization with 106.5 PFU of NDV expressing the HPIV3 HN protein induced a titer of serum HAI antibodies to the foreign HN protein that was only slightly less than that induced by infection with wild-type HPIV3 and equaled or exceeded that induced in previous studies against the HN protein of the attenuated BHPIV3 vector bearing one or two foreign inserts (36). This high level of immunogenicity was remarkable, since HPIV3 and BHPIV3 appear to replicate much more efficiently than NDV. Specifically, the titers of shed virus with HPIV3 and BHPIV3 were 100- to 10,000-fold higher than that for NDV. Our experience with immunization and challenge studies with wild-type HPIV3 and attenuated derivatives in these species of nonhuman primates indicate that this level of HPIV3-specific immunity would be highly protective against HPIV3 challenge. In the animals immunized with the NDV vectors, administration of a second dose 28 days later provided a modest boost and resulted in a titer of HPIV3-specific serum antibodies that exceeded that observed following HPIV3 infection. We did not challenge these doubly immunized animals with HPIV3 because the level of immunity was so high that, based on past experience, challenge virus replication was expected to be highly restricted (9, 37).
The two NDV strains, NDV-BC/HN and NDV-LC/HN, were comparable in most respects in vivo in this study. The major difference was that NDV-BC/HN induced a substantially higher (almost eightfold-higher) level of NDV-specific antibodies following the first infection, although the difference was much less following the second infection. The NDV-BC/HN vector also induced a somewhat higher level of serum HAI antibody to the HPIV3 HN protein, although this difference was only twofold or less. One possibility is that the NDV-LS/HN virus was slightly more restricted and hence less immunogenic in primates than the NDV-BC/HN virus. In general, this suggests that the two NDV strains do not differ greatly in their infectivity in primates, in contrast to the situation in birds. It may be that the determinants of virulence in birds do not have the same impact in primates. Another difference between the two NDV strains was evident in vitro, namely that NDV-BC/HN expressed substantially higher levels of intracellular HN mRNA in a 24-h period than did NDV-LS/HN or, unexpectedly, wild-type HPIV3. It has been noted previously that lentigenic strains of NDV, including NDV-LS, have reduced levels of transcription in vitro (21). Whether this influences the immunogenicity of the vector is not known, although the comparison of NDV-LS/HN and NDV-BC/HN in the present study did not provide evidence of a major increase in the immunogenicity of the latter virus concomitant with its increased gene expression in vitro.
NDV-based vaccines can potentially be developed against known highly pathogenic agents such as SARS-CoV, avian influenza, and Ebola virus, as well as against any new pathogen that might emerge in the future. Such a vaccine could be used locally by single intranasal administration to control regional outbreaks, since in this setting the ability of a single immunization to induce a high level of immunity would be advantageous. Such as vaccine could also be used prophylactically to protect medical personnel working in areas endemic for such viruses. To optimize the immune response in such medical personnel, the vaccine could be used alone or in prime-boost strategies with other vaccines such as DNA vaccines. Since immunization with NDV-vectored vaccines results in an induction of immune response specific for the vector and since NDV strains have a high degree of antigenic relatedness, the repeated use of an NDV-based vaccine beyond the two doses used in the present study probably would not be feasible. However, this problem can potentially be solved by the development of vaccine vectors based on avian paramyxoviruses belonging to serotypes other than NDV. Nine serotypes of avian paramyxoviruses are known; NDV belongs to serotype 1. Representatives of serotypes 2 to 9 cause disease with various levels of severity in birds (32), and we plan to evaluate them for infectivity, safety, and immunogenicity in birds, rodents, and nonhuman primates.
An intranasal vaccine based on NDV or another comparable paramyxovirus would have a number of advantages for controlling highly pathogenic agents. First, intranasal immunization induces both local and systemic immunity, which should prevent acquisition of infection or decrease its severity. A number of highly pathogenic agents use the respiratory tract as a major portal of entry; in some cases, they replicate extensively and cause disease at that site. Restriction of replication of a viral pathogen in the respiratory tract should greatly decrease its transmission to susceptible contacts. Because of this, direct stimulation of immunity in the respiratory tract should improve vaccine efficacy and usefulness for local outbreak control. Second, there is extensive experience with human clinical trials of administration of human paramyxoviruses to the respiratory tract (12-14). This is particularly true for human parainfluenza viruses, which are reasonably close relatives of NDV and other avian paramyxoviruses. This experience will help guide vaccine development. Third, if pretested vaccine vectors based on NDV or comparable viruses are available, the development of a vaccine against any future pathogen that might unexpectedly emerge could be rapid, since it would only require cloning of the gene of a protective antigen and insertion into the vector backbone. Clinical evaluation, manufacture, and delivery of such a vaccine would be expedited by the experience from prior testing of the vectors. Fourth, manufacture and use of such vaccines would be safe, since propagation of highly pathogenic viruses would not be involved. The value of this safety feature is illustrated by the recent laboratory-based outbreaks that occurred during work with SARS-CoV (28, 29). Finally, genetic reassortment does not occur with a nonsegmented virus like NDV, and RNA recombination by nonsegmented negative-strand RNA viruses is negligible (40). This contrasts with the situation of certain existing live vaccines, such as poliovirus, for which recombination with circulating strains has been described (8), or for potential live vaccines, such as for avian influenza virus or SARS-CoV, which would have the strong possibility of reassortment or recombination, respectively, with circulating strains. The observation that NDV-vectored vaccines will be essentially free of genetic exchange with circulating viruses represents a critical safety feature for the individual vaccinee and for the overall population.
We thank Brad Fineyfrock for performing NDV infections and taking biological samples from the animals, Ernest Williams and Fatemeh Davoodi for performing HAI assays, and Mario Skiadopoulos for providing rabbit anti-HPIV3 serum.
This project was funded as a part of the NIAID Intramural Program.
We have no commercial interests in NDV vectors.