Modified Newcastle disease virus vectors expressing the H5 hemagglutinin induce enhanced protection against highly pathogenic H5N1 avian influenza virus in chickens.

Naturally-occurring attenuated strains of Newcastle disease virus (NDV) are being developed as vaccine vectors for use in poultry and humans. However, some NDV strains, such as Beaudette C (BC), may retain too much virulence in poultry for safe use, and more highly attenuated strains may be suboptimally immunogenic. We therefore modified the BC strain by changing the multibasic cleavage site sequence of the F protein to the dibasic sequence of avirulent strain LaSota. Additionally, the BC, F, and HN proteins were modified in several ways to enhance virus replication. These modified BC-derived vectors and the LaSota strain were engineered to express the hemagglutin (HA) protein of H5N1 highly pathogenic influenza virus (HPAIV). In general, the modified BC-based vectors expressing HA replicated better than LaSota/HA, and expressed higher levels of HA protein. Pathogenicity tests indicated that all the modified viruses were highly attenuated in chickens. Based on in vitro characterization, two of the modified BC vectors were chosen for evaluation in chickens as vaccine vectors against H5N1 HPAIV A/Vietnam/1203/04. Immunization of chickens with rNDV vector vaccines followed by challenge with HPAIV demonstrated high levels of protection against clinical disease and mortality. However, only those chickens immunized with modified BC/HA in which residues 271-330 from the F protein had been replaced with the corresponding sequence from the NDV AKO strain conferred complete protection against challenge virus shedding. Our findings suggest that this modified rNDV can be used safely as a vaccine vector with enhanced replication, expression, and protective efficacy in avian species, and potentially in humans.

Introduction

Newcastle disease virus (NDV) is a member of the family Paramyxoviridae and has a nonsegmented, negative-sense RNA genome that contains six genes (3′-N-P-M-F-HN-L-5′) [1].

NDV isolates are categorized into three pathotypes based on their virulence in chickens: lentogenic (avirulent), mesogenic (moderately virulent), and velogenic (virulent) [2]. Lentogenic and mesogenic strains of NDV are widely used as live attenuated vaccines to control Newcastle disease in poultry. In addition, NDV can be modified by reverse genetics to express one or more protective antigens representing heterologous pathogens. One of the applications has been to modify NDV to express a protective antigen from another poultry pathogen to create a bivalent vaccine. For example, NDV has been engineered to express the hemagglutinin HA gene of H5N1 highly pathogenic avian influenza (HPAIV) virus, or the VP2 protein of infectious bursal disease virus (IBDV), to create a bivalent vaccine against NDV and HPAIV or IBDV, respectively [3], [4]. Another application has been used to develop NDV strains as a potential vaccine vector in humans against emerging and re-emerging pathogens, such as H5N1 HPAIV and severe acute respiratory syndrome coronavirus (SARS-CoV) [5], [6].

Although both lentogenic and mesogenic strains have potential as veterinary or human vaccine vectors, in nonhuman primates the mesogenic strain Beaudette C (BC) replicated to a higher titer and induced a substantially higher antibody response compared to the lentogenic strain LaSota, and thus appeared to be more effective [7]. However, NDV strains that have a polybasic cleavage site in the F protein and an intracerebral pathogenicity index (ICPI)>0.7 have been classified as Select Agents. Most mesogenic strains, including strain BC, fall into this category even though their pathogenic potential in chickens is limited. The Select Agent classification renders them impractical for vaccine development and use. Therefore, we have endeavored to develop NDV vaccine vectors that would be more effective than lentogenic strains and yet would have the avirulent phenotype. We further evaluated these new NDV vectors by using them to express the HA protein of H5N1 HPAIV. HPAIV is an economically important disease of poultry worldwide. In particular, HPAIV subtype H5N1 infections have resulted in the culling or death of more than 500 million poultry in more than 62 countries. Furthermore, H5N1 strains have been found to cause disease in humans. Therefore, vaccination of poultry or humans against HPAIV could play an important role in reducing virus shedding and raising the threshold for infection and transmission [9]. However, development of vaccines against HPAIV has been hampered due to poor immunogenicity of the virus [10]. The development and manufacture of HPAIV vaccines based on inactivated virus are costly due to the requirement of enhanced biosafety level 3 containment. Therefore, vaccine platforms that can avoid these shortcomings are needed. Live NDV vectors represent an alternative approach. Remarkably, our modified versions of the BC strain expressing the HA protein of H5N1 HPAIV were found to be avirulent by the ICPI test. We evaluated selected constructs for replication, immunogenicity, and protective efficacy in chickens. Our results showed that two of the modified vectors are superior to the commonly used rLaSota vector.

Materials and Methods

Construction and in vitro characterization of modified versions of NDV strain BC

We modified a previously described full length cDNA of the antigenome of strain BC [11] to change its naturally-occurring F protein cleavage site motif (RRQKR↓F) to that of strain LaSota (GRQGR↓L) to create rBC/LascFc. This backbone was further modified by replacing various regions in the F gene with those of velogenic NDV strain AKO [8] or the complete HN protein gene of strain LaSota (Fig. 1A). Particularly, we replaced the region of the F gene encoding amino acids 271–330, or 275–330, or 331–390 of rBC/Las Fc with the corresponding segment of the AKO strain. Strain AKO was chosen because this strain replicates to high titer in vitro and in vivo [15]. We chose these segments, because our preliminary work showed that these modifications enhanced virus replication and syncytium formation in vitro (data not shown). Consequently, this can also enhance the in vivo replication and immunogenicity of vaccine vectors. Infectious viruses were generated using a reverse genetics established in our laboratory [12].

Figure 1

Construction of modified versions of NDV strain rBC and virus replication and induction of serum antibodies in response to infection of 2-week-old chickens. (A) rBC and rLaSota are recombinant versions of the respective biological strains. The other four viruses are versions of rBC that each have the polybasic F protein cleavage site of the BC strain replaced with the cleavage site of LaSota. In addition, in modified rNDV/AKO F 271-330, AKO F 275-330, and AKO F 331-390, the regions of the F gene encoding the indicated F protein amino acids have been swapped with that of virulent strain AKO to improve replication. (B) Chickens (eight birds per group) were inoculated with each virus (264 HA units) by the intranasal route. Three birds in each group were sacrificed on 3 dpi, and virus titers in the collected tissue samples (brain, trachea, lung, and spleen) were determined by limiting dilution in DF1 cells. (C) Serum samples were collected at 1 and 2 wpi and evaluated for virus-specific antibodies by a hemagglutination inhibition assay using chicken erythrocytes.

The in vivo replication and immunogenicity of the modified rNDVs were evaluated in 2-week-old chickens (eight birds per group). Birds were inoculated with 200 μl of each virus (256 HA units/bird) by the intranasal route. Three birds from each group were sacrificed at 3 days post-infection (dpi) and tissues samples (lung, trachea, spleen, and brain) were collected for virus titration. Serum samples collected on days 7 and 14 were evaluated for seroconversion by hemagglutination inhibition (HI) assay [2].

Construction and characterization of modified versions of NDV vectors expressing the HA protein of HPAIV

The HA gene ORF of HPAIV strain A/Vietnam/1203/04 (H5N1) was modified by PCR and inserted between the P and M genes in the antigenomic cDNAs of rLaSota and the modified rNDVs. In addition, the original polybasic cleavage site of the HA gene (PQRERRRKKG) was replaced by that of the low-pathogenicity influenza virus strain A/chicken/Mexico/31381/94 (PQRETG) [13]. Infectious viruses were generated by reverse genetics [12]. The expression of the HA protein by rNDVs and its incorporation into the vector particles were evaluated by Western blotting [4]. Surface expression of the HA protein was evaluated on virus-infected DF1 cells (MOI of 0.1) by immunofluorescence microscopy and flow cytometry. The multicycle growth kinetics of rNDVs was evaluated in DF1 cells in the presence of 10% chicken egg allantoic fluid [12], [14]. Pathogenicity of rNDV/HA vectors was evaluated by mean embryo death time (MDT) in embryonated chicken eggs and ICPI assay in 1-day-old chicks [2].

Immunogenicity and protective efficacy of the NDV/HA vectors in 2-week-old chickens

Groups (n  = 16 per group) of 2-week-old SPF chickens were infected by the oculonasal route with 106 EID50 per bird of rLaSota/HA, rNDV-AKO F 271–330/HA, or rNDV-Las HN/HA, and an additional group of birds (n  = 6) was left uninfected. Following immunization, pre-challenge serum samples were collected weekly from all of the birds. Eight birds from each group were challenged with 104 ELD50 of HPAIV strain A/Vietnam/1203/2004 at 1 week post-immunization (wpi), and the remaining eight birds were challenged in the same way at 3 wpi. For the immunized control group, 3 birds were challenged 1 and 3 wpi. To monitor shedding of the challenge virus, oral and cloacal swabs were collected on days 4 and 7 post-challenge, inoculated into 9-day-old SPF embryonated chicken eggs, and confirmed by HA assay using chicken erythrocytes. Three chickens from each group were sacrificed at 4 days post-challenge to evaluate challenge virus replication in different organs (brain, trachea, lungs, and spleen). All experiments involving virulent NDV and HPAIV were performed in our USDA approved enhanced Biosafety Level-3 facility.

Results

Generation of new NDV vectors

A series of modified versions of rBC (Fig. 1A) were recovered and passaged seven times in 1-day-old chicks. The sequence of the F gene of these viruses was confirmed by RT-PCR and sequence analysis to be correct. The in vivo replication and immunogenicity of these rNDVs were evaluated in 2-week-old chickens (Fig. 1B and C). When birds were sacrificed at 3 dpi, rBC replication was detected in the trachea, lung, and spleen, but not in the brain, whereas replication of rLaSota was detected only in the trachea (Fig. 1B). Two of the modified viruses, rBC/Las Fc and rNDV-AKO F 275–330, remained restricted to the trachea while the other two modified viruses, rNDV-AKO F 271–390 and rNDV-AKO F 331–390, replicated in both the trachea and the lung (Fig. 1A). Each of the modified vectors replicated more efficiently than rLaSota (p  < 0.05). rNDV-AKO F 271–390 and rNDV-AKO F 331–390, the two viruses that replicated in both the trachea and the lungs, induced higher levels of NDV-specific serum antibodies than rLaSota and the other modified viruses at 1 wpi. All of the modified viruses induced higher levels of serum antibodies than rLaSota at 2 wpi (p  < 0.05) (Fig. 1C).

Generation of and characterization of NDV vectors expressing the HA protein of H5N1 HPAIV

We then engineered rLaSota and three of the modified rBC vectors described above as well as rBC in which the HN gene was replaced by that of LaSota (Fig. 2A) to express the HA protein of H5N1 HPAIV strain A/Vietnam/1203/2004 (Fig. 2A). The recovered viruses were passaged seven times in 9-day-old embryonated chicken eggs. All of the viruses replicated efficiently in the eggs (>28  HAU/ml), and the sequence of F and HA genes were found to be correct. The multi-cycle growth kinetics of the rNDVs bearing the H5N1 HPAIV HA gene showed that each of the modified rBC viruses replicated more efficiently than rLaSota or rLaSota/HA for the first 24 hpi (Fig. 2B). Specifically, the rBC-derived viruses reached 8 log10 TCID50/ml at 24 hpi, whereas rLaSota and rLaSota/HA had titers of 6.5 and 5.4 log10 TCID50/ml, respectively, at 24 hpi. The expression of HA by the rNDVs was evaluated in DF1 cells in the presence of 10% allantoic fluid. Despite the presence of the added protease, under these conditions HA was detected only as the uncleaved precursor (HA0, molecular weight of 66 kDa). The expression of the HA0 protein could be detected at 12 h post-infection (hpi) in the case of three modified rNDVs (AKO F 271–330, AKO F 275–330, and Las HN) (Fig. 3A). Subsequently, the expression of HA0 was detected for all the rNDVs at 24 hpi. At 24 hpi, modified rNDV containing AKO F 331–390 or Las HN expressed higher levels of HA0 than the other viruses.

Figure 2

Construction of modified NDVs expressing the H5N1 HPAIV HA gene and multi-cycle replication in vitro. (A) The HA ORF of HPAIV H5N1 strain A/Vietnam/1203/04 was engineered to be flanked by the NDV gene-start and gene-end signals and was inserted into the intergenic region between the P and M genes in the rNDV vectors shown. Genes derived from rBC or rLaSota are shown as black or gray rectangles, respectively. F gene segments derived from the NDV AKO strain are shown as white bars with designated location. All viruses contained the F protein cleavage sequence of LaSota, GRQGRL, as shown. (B) Monolayer cultures of the chicken fibroblast DF-1 cell line were infected with the indicated viruses at an MOI of 0.01 PFU/cell. The medium contained 10% uninfected egg allantoic fluid as a source of exogenous protease. Aliquots of the media were taken at 8-h intervals and viral titers were determined by limiting dilution on DF1 cells.

Figure 3

Expression of H5N1 HPAIV HA protein in vitro by rNDV/HA vectors and incorporation of HA into rNDV virions. (A) Expression of the HA protein in DF1 cells. DF1 cells were infected with each virus at MOI 1, and cell lysates were collected at 12 or 24 h post-infection for Western blot analysis using convalescent serum from chickens that had been infected with H5N1 HPAIV to visualize the HA protein and monoclonal antibody to visualize the NDV HN protein. (B) Incorporation of the HA protein into the NDV vector particle. The viruses were harvested from allantoic fluids of infected eggs at 72 hpi, purified through a 30% sucrose cushion, and analyzed by Western blot as in part A. (A, B) Lanes: rLaSota (lane 1), rLaSota/HA (lane 2), modified rNDV-AKO 271-330/HA (lane 3), modified rNDV-AKO 275-330/HA (lane 4), modified rNDV-AKO 331-390/HA (lane 5), and modified rNDV-Las HN/HA (lane 6). (C, D) Expression of the HA protein on the surface of DF-1 cells infected with the indicated NDV/HA vector. DF1 cells were infected with each virus (MOI of 0.1) and, at 24 h post-infection, were stained with anti-peptide antiserum against the HA protein followed by anti-Alexa Fluor 488 antibody, fixed with 4% paraformaldehyde and were analyzed by immunofluorescence microscopy (C) and flow cytometry (D).

The incorporation of HA into the vector particle was further analyzed by Western blotting (Fig. 3B). The HA protein was detected in the form of HA0 as well as the larger subunit of the cleaved form, HA1 (molecular weight 40 kDa). Incorporation of HA into the vector particle was observed to various extents for the rLaSota parent vector and the modified rNDV vectors, in particular those containing F 271–330 and Las HN (Fig. 3B). There was a two-fold increase in the incorporation of HA1 protein into the vector particle for rNDV-Las HN/HA compared to rLaSota/HA.

We next evaluated the surface expression of the HA protein in DF1 cells by immunofluorescence microscopy (Fig. 3C). Similar to the results with the Western blots, modified rNDV containing AKO F 271–330 or Las HN expressed higher levels of the HA protein on the surface of DF1 cells. Furthermore, this was quantitatively confirmed by flow cytometry (Fig. 3D). These results identified rNDV vector containing AKO F 271–330 or Las HN to have enhanced HA expression and incorporation of the HA protein into virus particles.

The pathogenicity of rNDVs expressing the HA protein was evaluated by the MDT and ICPI assays (Table 1 ). The avirulent nature of rLaSota was not affected in either assay by expression of the HPAIV HA protein. The MDT values of the modified rNDVs expressing the HA protein were more than 90 h, and the ICPI values were 0.00, indicating that these modified rNDV vectors are avirulent in chickens. Chicks infected with rNDVs had no apparent clinical signs during the 8-day period of the ICPI test, confirming the attenuation of all modified rNDVs.

Table 1

Pathogenicity of parental and modified rNDVs in embryonated eggs and 1-day-old chicks.

Virus	MDTa (h)	ICPIb
rBC	58	1.57
rLaSota	117	0.00
rLaSota/HA	121	0.00
Modified rNDV-AKO F 271-330/HA	112	0.00
Modified rNDV-AKO F 275-330/HA	114	0.00
Modified rNDV-AKO F 331-390/HA	98	0.00
Modified rNDV-Las HN/HA	102	0.00

Mean embryo death time (MDT): the mean time (h) for the minimum lethal dose of virus to kill all of the inoculated embryos [2]. Pathotype definition: virulent strains, <60 h; intermediate virulent strains, 60 to 90 h; avirulent strains, >90 h.

Intracerebral pathogenicity index (ICPI): evaluation of disease and death following intracerebral inoculation in 1-day-old SPF chicks [2]. Pathotype definition: virulent strains, 1.5–2.0; intermediate virulent strains, 0.7–1.5; and avirulent strains, 0.0–0.7.

Immunogenicity and protective efficacy of rNDV/HA vectors in chickens against HPAIV challenge

Two-week-old chickens in groups of 16 were infected intranasally with rLaSota/HA, rNDV-AKO F 271–330/HA, or rNDV-Las HN/HA, or were left uninfected (n  = 6). To determine the ability of the modified viruses to induce rNDV-specific or HPAIV-specific immune responses, serum samples were collected at 1, 2, and 3 wpi and tested against HPAIV A/Vietnam/1203/2004 (Fig. 4A) and highly virulent NDV strain GB Texas (Fig. 4B) using HPAIV-specific and NDV-specific HI assays. In this study, immune response against GB Texas was evaluated for potential use of NDV vectors as a dual vaccine. In the HPAIV-specific assay, HI titers increased with time, reaching the highest titer at 3 wpi. The two modified rNDV vectors induced higher HA-specific antibody titers than the rLaSota vector (p  < 0.05). In the NDV-specific assay, the increase in antibody titer with time was somewhat less and the differences between the viruses were somewhat less. The highest NDV-specific antibody titers were observed with the modified rNDV vectors, although this was less evident at 3 wpi.

Figure 4

Induction of serum antibodies in 2-week-old chickens in response to infection with rNDV vector vaccines. Chickens were inoculated with each virus (106 EID50) by the intranasal route. Serum samples were collected at 1, 2, and 3 wpi. Virus-specific antibodies were determined by a hemagglutination inhibition assay using HPAIV A/Vietnam/1203/2004 (A) or NDV strain GB Texas (B).

To evaluate the protective efficacy of rNDVs, the immunized chickens were divided into two groups and challenged at two different time points, 1 wpi and 3 wpi. Each group (eight birds each; three for the unimmunized control) was challenged with 104 ELD50 of homologous HPAIV A/Vietnam/1203/2004 via the oculo-nasal route. For the unimmunized chickens, challenge with the HPAIV virus resulted in clinical signs at 1 day post-challenge (dpc), and 100% mortality at 2 dpc. In contrast, all of the immunized chickens were completely protected from clinical disease and mortality following HPAIV challenge.

We determined shedding and replication of HPAIV challenge virus by collecting oral and cloacal swabs (4 and 7 dpc) and tissue samples (brain, lung, trachea, spleen; 4 dpc) from the immunized chickens. For the groups challenged at 1 wpi, challenge HPAIV shedding was detected 4 dpc in oral and cloacal swabs in some chickens immunized with rLaSota/HA or modified rNDV-Las HN/HA, whereas no shedding was detected 4 dpc from chickens immunized with modified rNDV-AKO F 271–330/HA virus (Table 2 ). No shedding was observed in any of the swabs taken at 7 dpc, and no virus was recovered from the tissue specimens collected 4 dpc (data not shown). For the groups that were challenged at 3 wpi, only a single bird (in the rLaSota/HA group) had detectable shedding (in an oral swab) at 4 dpc (Table 2). No shedding was detected from any bird on 7 dpi, and no virus was detected in any of the tissue samples collected 4 dpc (data not shown). These data indicated that all of the immunized groups of the chickens were efficiently protected against a dose of HPAIV that was rapidly fatal in unimmunized birds. The modified rNDV vectors appeared to be more protective than rLaSota.

Table 2

Oral and cloacal shedding of A/Vietnam/1203/2004 HPAIV challenge virus in 2-week-old chickens immunized 1 or 3 weeks earlier with rNDV expressing the HPAIV HA protein.

Chicken groupa	1 week post-immunization		3 weeks post-immunization
	Oral	Cloacal	Oral	Cloacal
rLaSota/HA	2/5	1/5	1/5	0/5
Modified rNDV-AKO F 271-330/HA	0/5	0/5	0/5	0/5
Modified rNDV-Las HN/HA	1/5	1/5	0/5	0/5

Two-week-old chickens (16 birds per group) were immunized intranasally with the indicated NDV strain. One week post-immunization, eight birds from each group were challenged with HPAIV A/Vietnam/1203/2004 by the intranasal route, and at 3 weeks post-immunization the remaining eight birds were challenged in the same way. Oral (A) and cloacal (B) swabs were collected from five birds in each group on day 4 and 7 post challenge. To identify the shedding of challenge virus, aliquots (100 μl each) of the collected samples were inoculated into three eggs per sample, and allantoic fluids were collected 3 dpi. Virus replication was determined by hemagglutination assay. Detection of virus in any of the three eggs was considered positive. Only the day 4 swabs are shown because no shedding was detected in any of the day 7 swabs. Note that this study included a mock-immunized group for each challenge (n = 3), but these birds died on day 2 post-challenge and were not sampled.

Discussion

NDV is being developed as a bivalent vaccine vector for protection of chickens against NDV and other poultry pathogens [3], [4], [9], [16], [17]. Additionally, NDV has shown promising results as a potential vaccine vector for human use [18], [19], [20]. NDV is attractive as a vaccine vector in humans because it is highly attenuated in non-human primates, it is not a common human pathogen and is antigenically distinct from common human pathogens and thus should be unaffected by pre-existing immunity, and it can be given topically to the respiratory tract to induce both mucosal and systemic immunity. Although both lentogenic and mesogenic NDV strains can be used as vaccine vectors, the mesogenic strains are more easily grown in vitro and are more immunogenic in vivo. However, mesogenic strains might cause disease in poultry, and most strains are classified as Select Agents, which precludes their development as a vaccine vector. Therefore, we changed the polybasic F cleavage site of mesogenic strain BC to the dibasic F cleavage site of strain LaSota. It would eliminate one of the criteria in classifying BC as a Select Agent. We also confirmed that the recovered virus stably maintained modified F protein cleavage sites during seven passages in chickens. Since reversion of the F protein cleavage site sequence will require mutation of 11 nucleotides, there exists a distance possibility of reversion to virulence. Further, the genetic stability of the cleavage site sequence and virus attenuation need to be carefully evaluated with insertion of different foreign genes into this vaccine vector prior to its use in the field. In addition, we swapped regions of the F gene with strain AKO, or swapped the HN gene with rLaSota. This was done because the fusogenic activity of NDV can affect virus replication in vitro and in vivo, and its modulation can enhance the potential of rNDVs for use as vaccine vectors [21]. The resulting modified rBC viruses were found to be more restricted for replication than their rBC parent and were more immunogenic than rLaSota strain in 2-week-old chickens.

We then used these vectors to express the HA glycoprotein of H5N1 HPAIV. H5N1 HPAIV viruses have been a major concern both in poultry industry and public health. There is a need to develop effective vaccines against HPAIV. Our results showed that all the modified rNDV/HA vectors were avirulent in 1-day-old chicks, with an ICPI value of 0.00. Thus, these vectors are not Select Agents and would not pose a threat to the poultry industry.

In this study, two modified rNDV vectors, one containing Las HN and the other containing AKO F 271–330, showed enhanced expression of the HPAIV HA protein and incorporation of the HA protein into the NDV particles. Thus, these two modified rNDV/HA vectors were chosen for evaluation of immunogenicity and protective efficacy against H5N1 HPAIV in parallel with the LaSota/HA vector. Our immunization study showed that most serum samples from immunized chickens with the three rNDVs at 1 wpi had high levels of HI titers against NDV and HPAIV, suggesting that these vectors can be efficient as early as 1 wpi. At 3 wpi, titers of serum samples from all immunized chickens remained high. Subsequent challenge with HPAIV H5N1 showed that each of the vectors induced substantial protection against HPAIV challenge. Specifically, whereas all of the unimmunized chickens died by 2 dpc, all of the immunized chickens remained free of disease without replication of detectable infectious challenge virus in tissue samples. However, evaluation of tracheal and cloacal swabs taken at 4 and 7 dpc did reveal differences in the level of protection. Specifically, only the rNDV-AKO F 271–330/HA vector completely prevented detectable HPAIV shedding in all of the birds. The rNDV-Las HN/HA vector was the next most protective, with shedding detected in 2 swabs. The rLaSota/HA was the least protective, with shedding detected in 4 swabs. These results suggest that the two modified rNDV vectors can be more efficient in restricting replication of HPAIV challenge virus than the current rLaSota vector. Since virus shedding can transmit HPAIV to unvaccinated flocks and to humans, an ideal HPAIV vaccine should completely prevent any challenge virus replication and shedding. This suggests that the rNDV-AKO F 271–330/HA vector represents an improved H5N1 HPAIV vaccine for poultry.

In summary, our finding showed that the mesogenic NDV strain BC can be modified for possible use as an avirulent, safe, and effective vaccine vectors in chickens. These rNDV vectors can replicate to higher levels in vitro and in chickens than the rLaSota virus, without causing any clinical signs. Our study has identified two improved rNDV vectors that induce higher levels of immunogenicity and protective efficacy against HPAIV. Whether these two rNDV vectors would also provide increased immunogenicity and protection against HPAIV in mammalian species remains to be identified.