A live attenuated cold adapted influenza A H7N3 virus vaccine provides protection against homologous and heterologous H7 viruses in mice and ferrets

Abstract

The appearance of human infections caused by avian influenza A H7 subtype viruses underscore their pandemic potential and the need to develop vaccines to protect humans from viruses of this subtype. A live attenuated H7N3 virus vaccine was generated by reverse genetics using the HA and NA genes of a low pathogenicity A/chicken/BC/CN-6/04 (H7N3) virus and the six internal protein genes of the cold-adapted A/Ann Arbor/6/60 ca (H2N2) virus. The reassortant H7N3 BC 04 ca vaccine virus was temperature sensitive and showed attenuation in mice and ferrets. Intranasal immunization with one dose of the vaccine protected mice and ferrets when challenged with homologous and heterologous H7 viruses. The reassortant H7N3 BC 04 ca vaccine virus showed comparable levels of attenuation, immunogenicity and efficacy in mice and ferret models. The safety, immunogenicity, and efficacy of this vaccine in mice and ferrets support the evaluation of this vaccine in clinical trials.

Introduction

Avian influenza A H7 subtype viruses have caused large outbreaks of disease in domestic poultry in Asia, Europe, North America, and South America in recent years, leading to severe economic losses to the poultry industry (Capua et al., 2002; Capua and Alexander, 2004; Hirst et al., 2004; Jones and Swayne, 2004; Koopmans et al., 2004). Some of these outbreaks have also been associated with the direct transmission of H7 subtype viruses from infected domestic poultry to humans, causing disease and sometimes death. Although sporadic cases of human infections with avian influenza H7 subtype viruses had occurred previously (Campbell et al., 1970; Kurtz et al., 1996; Taylor and Turner 1977; Webster et al., 1981), H7 viruses became a greater concern with the direct transmission of H7N7 viruses to a large number of humans in the Netherlands in 2003. A highly pathogenic (HP) avian influenza A H7N7 subtype virus caused 89 human infections including one fatal case of acute respiratory distress syndrome during a severe disease outbreak in domestic poultry in 2003 in The Netherlands (Fouchier et al., 2004; Koopmans et al., 2004). In November 2003, avian influenza A H7N2 subtype virus was isolated from a patient from New York who was hospitalized for upper and lower respiratory tract illness. The source of this virus remains unknown (http://www.cdc.gov/mmwr/preview/mmwrhtml/mm5325a1.htm). In 2004, two poultry workers were infected when an HP avian influenza H7N3 virus emerged in domestic poultry in British Columbia, Canada; these infections resulted in mild respiratory disease and conjunctivitis (Hirst et al., 2004; Pasick et al., 2005; Tweed et al., 2004). A serological survey detected anti-H7 antibodies in 7 poultry workers who were exposed to a low pathogenicity (LP) avian influenza H7N3 virus during avian influenza outbreaks in Italy in 2002–2003 (Puzelli et al., 2005). Infection of a poultry worker with an avian influenza A H7N3 virus was reported in the United Kingdom during an outbreak of avian influenza in domestic poultry in 2006 (Nguyen-Van-Tam et al., 2006). Elevated antibody titers against H7 subtype virus have been reported in veterinarians in the United States who had a history of exposure to birds (Myers et al., 2007).

This substantial number of human infections following occupational exposure to H7 subtype viruses from avian species in recent years underscores their pandemic potential and the need to develop efficacious vaccines against viruses of this subtype. A subunit vaccine based on the HA and NA proteins of H7 viruses with immune stimulating complexes (ISCOM) as an adjuvant induced protective immunity against challenge infection in mice though non-adjuvanted subunit vaccines proved to be poorly immunogenic and failed to induce protection (de Wit et al., 2005). Two doses of a formalin inactivated H7N2-PR8 reassortant virus vaccine provided protection in mice against H7 viruses from homologous and heterologous lineages (Pappas et al., 2007).

Phylogenetic analysis of the H7 HA gene reveals a separation into two lineages that correspond to the geographic separation of the birds that they infect, namely Eurasian and North American lineages (Campitelli et al., 2005). We have previously analyzed the antigenic relatedness among H7 viruses from the two lineages using post-infection mice and ferret sera (Joseph et al., 2007). Based on the phylogenetic relationships and the ability to induce broadly cross-neutralizing antibodies in mice and ferrets, we selected the wild-type (wt) A/chicken/BC/CN-7/04 (H7N3) (H7N3 BC 04 HP wt) virus from the North American lineage for vaccine development (Joseph et al., 2007). This virus was isolated during an avian influenza disease outbreak in domestic poultry in British Columbia, Canada that resulted in human infections in 2004. The HA protein of this virus had a 7 amino acid insertion near the HA cleavage site (PENPKQAYRKRMTR*GLF) that made the virus highly pathogenic to chickens. The source of this peptide insertion was subsequently identified as the matrix protein (M) of the same virus (Bowes et al., 2004; Hirst et al., 2004; Pasick et al., 2005). The HA gene of H7N3 BC 04 HP wt virus would have to be modified by removing the 7 amino acid peptide insertion in order to generate a LP vaccine virus that could be safely handled under standard BSL-2 laboratory containment. However, the A/chicken/BC/CN-6/04 (H7N3) LP virus (H7N3 BC 04 LP wt), that was isolated from the same outbreak did not have the 7 amino acid peptide insertion at the HA cleavage site (PENPKTR*GLF), was not pathogenic to chickens and was a precursor of the HP virus that caused severe disease in poultry. Because the nucleotide and amino acid sequences of the HA and NA of H7N3 BC 04 LP wt and H7N3 BC 04 HP wt viruses were identical and the HA of H7N3 BC 04 LP wt virus did not require modification of the HA cleavage site, we chose the H7N3 BC 04 LP wt virus as the source of the HA and NA genes for vaccine development.

Several strategies have been used to develop vaccines against avian influenza viruses with pandemic potential. Inactivated virus vaccines, live attenuated virus vaccines, vectored vaccines, and DNA vaccines have been developed against avian influenza A H5N1 subtype viruses and showed promise in preclinical studies (Subbarao and Joseph, 2007). Live attenuated, cold-adapted (ca) influenza virus vaccines against human influenza viruses elicit both systemic and mucosal immunity in the respiratory tract (Maassab and Bryant, 1999; Murphy and Coelingh, 2002). Classical reassortment or plasmid-based reverse genetics (Hoffmann et al., 2000) allow the generation of live attenuated ca vaccine viruses bearing the HA and NA genes of a wt virus of interest and internal protein genes of the vaccine donor virus A/Ann Arbor/6/60 ca (H2N2) (AA ca) that was generated by the serial passage of a wt A/Ann Arbor/6/60 (H2N2) (AA wt) virus at successively lower temperatures (Maassab and Bryant, 1999). Candidate live attenuated virus vaccines against H9N2 and H5N1 avian influenza viruses generated on the AA ca backbone using reassortment and plasmid-based reverse genetics, respectively, were safe and efficacious in mice and ferrets (Chen et al., 2003; Li et al., 1999; Suguitan et al., 2006). Phase 1 clinical evaluation of these vaccines is currently in progress.

In this study we describe the generation of a live attenuated ca H7N3 virus vaccine by plasmid-based reverse genetics using the HA and NA genes of the H7N3 BC 04 LP wt virus and six internal protein genes of the AA ca virus and demonstrate the immunogenicity and efficacy of the vaccine in mice and ferrets.

Results

Generation of the H7N3 BC 04 ca virus

A reassortant virus containing the H7 HA and N3 NA genes derived from the H7N3 BC 04 LP wt virus and remaining gene segments from the AA ca virus was generated by plasmid-based reverse genetics as described previously (Suguitan et al., 2006). The reassortant virus was biologically cloned by limiting dilution in the allantoic cavity of embryonated specific pathogen free (SPF) hen’s eggs. The nucleotide sequence of each gene segment of the reassortant virus was analyzed and the sequence identity with the corresponding gene segment of the parent viruses was confirmed. The five mutations in the internal protein genes, PB1¹¹⁹⁵ (K391E), PB1¹⁷⁶⁶ (E581G), PB1²⁰⁰⁵ (A661T), PB2⁸²¹ (N265S), and NP¹⁴⁶ (D34G) that specify the temperature sensitivity (ts) and attenuation (att) phenotypes of the AA ca vaccine donor virus (Jin et al., 2003, 2004) were present in the reassortant H7N3 BC 04 ca virus.

The H7N3 BC 04 ca virus is ts and trypsin dependent

The phenotypic properties of the H7N3 BC 04 ca and H7N3 BC 04 HP wt viruses were compared in primary chicken kidney (PCK) and chicken embryo fibroblast (CEF) cells. The H7N3 BC 04 ca and the parent AA ca viruses replicated equally well at 25°C and 33°C (ca) but not at 39°C (ts) while the H7N3 BC 04 HP wt virus did not replicate as efficiently at 25°C as at 33°C but replicated equally well at 33°C and 39°C and therefore was neither ca nor ts (Table 1).

Table 1.

The H7N3 BC 04 ca reassortant virus is ca and ts in PCK cells

	Mean virus titer ± SE (log10TCID50/mL) at indicated temperature
Virus				caa phenotype	tsb phenotype
	25°C	33°C	39°C
H7N3 BC 04 wt HP	5.5±0.2	9.7±0	9.7±0	−	−
H7N3 BC 04 ca	8.7±0	8.7±0	6.4±0	+	+
AA ca	10.7±0	10.7±0	6.2±0	+	+

^aca = difference between the mean TCID₅₀ at 33°C and 25°C ≤ 100-fold

^bts = difference between the mean TCID₅₀ at 33°C and 39°C ≥ 100-fold

The H7N3 BC 04 HP wt virus that possesses a 7 amino acid long insertion at the cleavage site of the HA protein formed plaques efficiently in CEF cells in the presence and absence of trypsin. The H7N3 BC 04 ca failed to form plaques in the absence of trypsin, consistent with the absence of a multi-basic cleavage site motif in the HA protein (Table 2).

Table 2.

The H7N3 BC 04 ca virus requires trypsin for efficient plaque formation in CEF cells

			Virus titer (log10 pfu/mL)
Virus	Amino acid sequence near HA cleavage sitea	Temperature
			With trypsin	Without trypsin
H7N3 BC 04 wt HP	PKQAVRKRMTR/GLF	37°C 33°C	8.0 8.2	8.0 7.8
H7N3 BC 04 ca	PK - - - - - - -TR/GLF	37°C 33°C	6.2 7.0	≤2.7b 4.2

^a/ indicates site of cleavage of HA into HA1 and HA2; - indicates absence of a corresponding residue in the aligned sequences.

^bLower limit of detection

Attenuation of the H7N3 BC 04 ca virus in chickens, mice, and ferrets

Level of attenuation in chickens following intravenous (i.v.) and intranasal (i.n.) inoculation

Clinical signs of illness were not observed in chickens upon i.n. and i.v. inoculation with the H7N3 BC 04 ca virus. The H7N3 BC 04 ca virus was not lethal for chickens and virus was not isolated from oropharyngeal or cloacal swabs that were collected on day 3 following i.n. inoculation or from the kidney, heart, brain, and lungs of birds from the i.n. inoculated group that were euthanized 3 dpi (data not shown). The H7N3 BC 04 ca virus also failed to elicit an antibody response following i.n. inoculation (data not shown). The failure of the H7N3 BC 04 ca virus to replicate in chickens could be because it is a temperature sensitive virus or because the AA virus is a human influenza virus that may be restricted in replication in chickens.

Lethality for mice

The H7N3 BC 04 wt and ca viruses were tested for lethality in mice. Consistent with our previous findings (Joseph et al., 2007), a dose of 10⁶ fifty percent tissue culture infectious doses (TCID₅₀) of the H7N3 BC 04 HP wt virus was lethal for mice by 7 day post-infection (dpi). Although the H7N3 BC 04 LP wt virus caused a slight reduction in weight by 3 dpi, all of the mice rapidly regained weight. The H7N3 BC 04 ca virus did not cause weight loss in mice (Fig. 1).

Figure. 1.

Percent change in body weight following i.n. inoculation of H7N3 viruses. Groups of 5 mice received 10⁶ TCID₅₀/mouse of BC 04 ca (□), H7N3 BC 04 wt LP (■), H7N3 BC 04 wt HP (▲) viruses. Mock infected mice (●) received L-15.

Replication kinetics of the H7N3 BC 04 ca virus in mice

The level of replication of the H7N3 BC 04 wt and ca viruses was assessed in mice on 2, 3, and 4 dpi. The AA wt and AA ca viruses were included for comparison. The H7N3 BC 04 LP and HP wt viruses replicated to similar levels in the respiratory tract of mice (Fig. 2). The level of replication of the H7N3 BC 04 ca virus in the upper respiratory tract was significantly lower than the wt viruses on 2 and 3 dpi. An increase in the level of replication on 4 dpi suggests that peak replication of H7N3 BC 04 ca virus may be delayed in the upper respiratory tract as compared to wt parent virus. However, replication of the H7N3 BC 04 ca virus in the lower respiratory tract was significantly lower than the wt viruses on 2 and 3 dpi and it was below the detection limit on 4 dpi. (Fig. 2). With the exception of a single time point (day 4 in the nasal turbinates), the level of replication of the H7N3 BC 04 ca virus was also lower than that of the attenuated AA ca virus. The H7N3 BC 04 ca virus was undetectable in the spleen (Fig. 2). Although low levels of wt viruses were detected in the brain, the ca viruses were undetectable. Thus, the replication of the H7N3 BC 04 ca virus was highly restricted in mice as compared to the wt H7N3 HP and LP viruses.

Figure. 2.

Replication of the H7N3 BC 04 ca virus in mice. Level of replication of ca and wt viruses in mice following i.n. inoculation with 10⁶ TCID₅₀/mouse of H7N3 BC 04 wt HP (hatched bars), H7N3 BC 04 wt LP (solid bars), H7N3 BC 04 ca (gray bars), AA wt (open bars) and AA ca (horizontal hatch bars) viruses. Virus titers in the (A) nasal turbinates, (B) lungs, (C) spleen, and (D) brain of 4 mice per group sacrificed on 2, 3, and 4 dpi, respectively are expressed as mean ± SE log₁₀TCID₅₀/gm of tissue. The dashed horizontal line indicates the lower limit of detection. * indicates statistically significant (p<0.05) reduction in virus titers compared to the corresponding wt virus.

Level of replication of the H7N3 BC 04 ca virus in ferrets

The level of replication of the H7N3 BC 04 wt and ca viruses was evaluated in ferrets 3 days following i.n. inoculation. The AA wt and AA ca viruses were included for comparison. While the H7N3 and AA wt viruses and the AA ca virus replicated to high titers in the upper respiratory tract, the H7N3 BC 04 ca virus replicated only to a moderate level, between 10⁴ and 10⁵ fifty percent egg infectious dose (EID₅₀)/gm (Fig. 3) and replication of the H7N3 BC 04 ca and AA ca virus were not detected in the lower respiratory tract of ferrets. In contrast, the H7N3 BC 04 wt viruses replicated to high titers in the lower respiratory tract (Fig. 3). Moderate levels of H7N3 and AA wt viruses were detected in the olfactory bulb. While a low level of the AA ca virus was detectable, the H7N3 BC 04 ca virus was not detected in the olfactory bulbs (Fig. 3). The H7N3 BC 04 HP and LP wt viruses were detected at moderate and low levels, respectively, but the H7N3 BC 04 ca and the AA ca viruses were not detected in the rest of the brain (Fig. 3).

Figure. 3.

Replication of the H7N3 BC 04 ca virus in ferrets. Level of replication of ca and wt viruses in ferrets following i.n. inoculation with 10⁷ TCID₅₀/ferret of H7N3 BC 04 wt HP (hatched bars), H7N3 BC 04 wt LP (solid bars), H7N3 BC 04 ca (gray bars), AA wt (open bars), and AA ca (horizontal hatch bars) viruses on 3 dpi. Virus titers are expressed as mean ± SE log₁₀EID₅₀/gm of tissue. The dashed horizontal line indicates the lower limit of detection. * indicates statistically significant (p<0.05) reduction in virus titer compared to the H7N3 BC 04 wt viruses.

Immunogenicity of H7N3 BC 04 ca virus

Immunogenicity in mice

The immunogenicity of the H7N3 BC 04 LP wt and H7N3 BC 04 ca viruses was evaluated in mice following immunization with 1 or 2 doses of the immunizing virus. A single dose of H7N3 BC 04 ca induced a detectable neutralizing antibody response against the homologous virus, with a geometric mean neutralizing antibody titer (GMT) of 87 by 4 weeks post-immunization (Table 3) that increased over the next 4 weeks. The final titer achieved following two doses of vaccine was similar to that seen 8 weeks after a single dose (Table 3). The GMT achieved following infection with the H7N3 BC 04 ca virus was 2 to 4-fold lower than the titer following infection with the H7N3 BC 04 LP wt virus (Table 3).

Table 3.

The H7N3 BC 04 ca virus is immunogenic in mice and ferrets

		Geometric mean neutralizing antibody titers achieved at indicated time post-immunization in indicated species
Virus used for immunization	Number of Dosesa
		Mice b			Ferrets c
		D 0	D 28	D 56	D 0	D 28	D 56
H7N3 BC 04 ca	1 2	<10 <10	87 45	403 470	<10 <10	160 84	ND d 264
H7N3 BC 04 wt LP	1 2	<10 <10	320 154	941 1701	<10 ND	571 ND	ND ND

^aMice and ferrets in the 1-dose group were immunized i.n. on day 0 and in the 2-dose group were immunized on day 0 and 28 with 10⁶ TCID₅₀ (mice) and 10⁷ TCID₅₀ (ferrets) of the indicated virus

^bGeometric mean of serum neutralizing antibody titers of 3 mice against the homologous H7N3 BC 04 wt LP virus

^cGeometric mean of serum neutralizing antibody titers of 9 ferrets that were immunized with the H7N3 BC 04 ca virus and 3 ferrets immunized with the H7N3 BC 04 wt LP virus against the homologous H7N3 BC 04 wt LP virus.

^dNot done

The level of cross-neutralizing antibodies was determined in the sera of mice that were challenged with the heterologous NL/03 (H7N7) and EG/63 (H7N3) viruses. The GMTs against the NL/03 virus were < 20 (the lowest dilution tested in this experiment) and 34 following 1 dose of the H7N3 BC 04 ca virus at 4 and 8 weeks, respectively and 60 in sera following 2 doses of vaccine. Similarly, the GMTs against the EG/63 virus were <20 and 26 following 1 dose at 4 and 8 weeks, respectively and 39 in sera following 2 doses of vaccine.

Immunogenicity in ferrets

As was seen in mice, neutralizing antibody was detectable in ferrets 4 weeks after 1 dose of vaccine but, unlike what was seen in mice, a higher GMT was noted following two doses of vaccine in ferrets. We did not determine whether the antibody titer would continue to rise in ferrets without a second dose of vaccine (Table 3). The GMT achieved following infection with the H7N3 BC 04 ca virus was 3 to 6-fold lower than following infection with the H7N3 BC 04 LP wt virus (Table 3). The GMT against the H7N3 BC 04 HP wt virus was similar to that against the H7N3 BC 04 LP wt virus (data not shown). One and 2 doses of the H7N3 BC 04 ca virus induced GMTs of 11 and 10, respectively against the heterologous NL/03 (H7N7) virus in ferrets.

Efficacy of the H7N3 BC 04 ca virus in mice

Protection from lethal challenge with homologous and heterologous H7 viruses

Following immunization with 1 or 2 doses of the H7N3 BC 04 ca vaccine virus, groups of 8 mice were challenged with 50 LD₅₀ of the H7N3 BC 04 HP wt from the North American or NL/03 (H7N7) and EG/63 (H7N3) viruses from the Eurasian lineages of H7 viruses. A single dose of the H7N3 BC 04 ca virus provided complete protection from lethal challenge with all three viruses while mock-immunized mice succumbed to death between days 6 and 8 post-challenge (data not shown). Morbidity (weight loss) following challenge administered 4 weeks after a single dose of vaccine is shown in Fig 4.

Figure. 4.

Percent change in body weight of mice i.n. immunized with a single dose of the H7N3 BC 04 ca vaccine virus following lethal challenge with HP H7 viruses. Groups of 8 mice that were immunized with 10⁶ TCID₅₀ of H7N3 BC 04 ca vaccine virus (□) or were mock-immunized (L-15) (▲) were challenged 4 weeks later with 50 LD₅₀ of (A) H7N3 BC 04 wt HP, (B) NL/03 (H7N7), and (C) EG/63 (H7N3) viruses. All immunized mice survived and mock-immunized mice succumbed to death between 6 and 8 dpi.

Protection from replication of homologous and heterologous H7 challenge viruses

The efficacy of 1 or 2 doses of the H7N3 BC 04 ca virus vaccine in preventing the replication of challenge viruses was evaluated on day 4 following challenge, when all of the challenge viruses should be present at high titer. A single dose of the H7N3 BC 04 ca virus vaccine (Fig 5A and 5B) provided protection from pulmonary replication of North American lineage viruses but not of Eurasian lineage viruses on day 4 post-challenge. Two doses (Fig. 5B) of the H7N3 BC 04 ca virus vaccine provided protection from pulmonary replication of Eurasian lineage viruses on day 4 post-challenge.

Figure. 5.

Replication of H7 wt challenge viruses in the respiratory tract of mice immunized with the H7N3 BC 04 ca vaccine virus. Groups of mice were immunized with 1 (open bars) or 2 (gray bars) doses of 10⁶ TCID₅₀ of the H7N3 BC 04 ca vaccine virus or were mock-immunized (L-15) (solid bars) and 4 mice per group were challenged with 10⁵ TCID₅₀ of each indicated virus (A) 4 weeks or (B) 8 weeks post-immunization. H7N3 BC 04 wt LP and H7N3 BC 04 wt HP are homologous and UT/95 and VA/02 are heterologous LP viruses from the North American lineage. EG/63 (H7N3) and NL/03 (H7N7) are antigenically heterologous HP viruses from the Eurasian lineage. Virus titers are expressed as mean ± SE log₁₀TCID₅₀/gm of tissue from groups of 4 mice. The dashed horizontal line indicates the lower limit of detection. The reduction in virus titer in the respiratory tract of the immunized mice compared to mock-immunized group was statistically significant (p<0.05).

Efficacy of the H7N3 BC 04 ca vaccine virus in ferrets

Homologous H7N3 viruses from the North American lineage and a heterologous NL/03 (H7N7) virus from the Eurasian lineage were also used for challenge studies in ferrets following 1 or 2 doses of the vaccine. Significant reductions (p<0.05) in the level of challenge virus replication was observed in the upper respiratory tract (nasal turbinates) when ferrets were challenged 4 weeks after 1 dose of the vaccine (Fig. 6). A single dose of the vaccine provided complete protection from the replication of the homologous H7N3 BC 04 LP wt virus and significantly reduced (p<0.05) the levels of replication of the homologous H7N3 BC 04 HP wt and the heterologous NL/03 (H7N7) viruses in the lower respiratory tract (lungs) (Fig. 6). One dose of the vaccine also prevented the spread of NL/03 (H7N7) to the brain beyond the olfactory bulb and significantly reduced (p<0.05) the virus titers of the H7N3 BC 04 HP wt and NL/03 (H7N7) viruses in the olfactory bulb (Fig. 6). Two doses of the vaccine further reduced the virus titers in the upper respiratory tract and completely prevented the spread of the H7N3 BC 04 HP wt and NL/03 (H7N7) viruses to the olfactory bulb (Fig. 6). However, low levels of the homologous H7N3 BC 04 HP wt and the heterologous NL/03 (H7N7) viruses were detected in the lower respiratory tract after 2 doses of the vaccine.

Figure. 6.

Replication of H7 wt challenge viruses in ferrets immunized with H7N3 BC 04 ca vaccine virus. Groups of ferrets immunized with 1 (open bars) or 2 (gray bars) doses of 10⁷ TCID₅₀ of H7N3 BC 04 ca vaccine virus and mock-immunized (L-15) (solid bars) ferrets were challenged 4 weeks post-dose 1 and 2 with 10⁶ TCID₅₀ of the indicated viruses. The H7N3 BC 04 wt LP and H7N3 BC 04 wt HP viruses are homologous viruses from the North American lineage and NL/03 (H7N7) is a heterologous HP virus from the Eurasian lineage. Virus titers in the nasal turbinates, olfactory bulbs, and brain are expressed as mean ± SE log₁₀TCID₅₀/gm of tissue. Virus titers in the lungs are expressed as mean ± SE log₁₀EID₅₀/gm of tissue. * indicates statistically significant (p<0.05) reduction in the virus titers. The dashed horizontal line indicates the lower limit of detection.

Discussion

Direct transmission of avian influenza A H7 subtype viruses from infected domestic poultry to humans resulting in disease and some times death underscores the threat posed to human health by viruses of this subtype (Fouchier et al., 2004; Koopmans et al., 2004; Subbarao and Joseph, 2007; Tweed et al., 2004). It is prudent to develop efficacious vaccines against H7 subtype viruses because of their pandemic potential. We have generated a live attenuated H7N3 BC 04 ca virus vaccine by plasmid-based reverse genetics using the HA and NA genes of the H7N3 BC 04 LP wt virus and six internal protein genes of the ca influenza A vaccine donor strain, AA ca virus. The vaccine virus was generated in Vero cells that are qualified for human vaccine production and subsequent amplification and biological cloning were carried out in SPF embryonated hen’s eggs. The absence of multi-basic amino acid residues at the cleavage site in the HA of the H7N3 BC 04 LP wt virus that was the precursor of the H7N3 BC 04 HP wt virus obviated the need to genetically modify the HA gene and allowed the safe handling of the reassortant vaccine virus under BSL-2 laboratory containment.

The H7N3 BC 04 ca vaccine virus displayed ca, ts, and att phenotypes that are specified by the internal protein genes of the AA ca virus (Jin et al., 2003, 2004). Consistent with the absence of multi-basic amino acid residues at the cleavage site in the HA, the H7N3 BC 04 ca vaccine virus was dependent on exogenous trypsin for plaque formation in CEF cells. The H7N3 BC 04 LP wt virus and the H7N3 BC 04 ca vaccine virus were not pathogenic to chickens (Bowes et al., 2004; Hirst et al., 2004; Pasick et al., 2005). Failure of the vaccine virus to replicate in chickens is likely a consequence of the temperature sensitivity of the virus and suggests that the use of the H7N3 BC 04 ca vaccine will not pose a threat to the poultry industry.

The reassortant H7N3 BC 04 ca virus was not lethal for mice. Unlike the wt parent virus, the replication of the H7N3 BC 04 ca virus was highly restricted in the respiratory tract of mice and systemic (extrapulmonary) spread was not observed. The H7N3 BC 04 ca virus was also restricted in replication in ferrets. In ferrets, the H7N3 BC 04 ca virus replicated moderately well in the upper respiratory tract but replication was not detected in the lower respiratory tract and there was no spread to the brain. These characteristics of the H7N3 BC 04 ca vaccine virus in ferrets fit the attenuation criteria for the live attenuated influenza virus vaccines (FluMist) that are licensed for human use. The attenuation of the H7N3 BC 04 ca vaccine virus in both mice and ferrets can be attributed to the acquisition of the 6 internal gene segments from the AA ca virus. Live attenuated ca H5 and H9 avian influenza A subtype virus vaccines bearing the 6 internal protein genes of the AA ca virus have demonstrated similar levels of attenuation in mice and ferrets (Chen et al., 2003; Suguitan et al., 2006). Restricted replication in the respiratory tract and lack of systemic spread in mice and ferrets suggest that H7N3 BC 04 ca vaccine virus is sufficiently attenuated for use in humans.

The H7N3 BC 04 ca vaccine virus induced neutralizing antibodies with comparable cross-reactivity in mice and ferrets. A single dose of the H7N3 BC 04 ca vaccine virus induced neutralizing antibodies in mice and ferrets 4 weeks post-immunization. A further rise in the antibody titer in the mouse sera collected 4 weeks later is consistent with our previous findings (Joseph et al., 2007). A similar phenomenon was observed in the case of H5N1 ca vaccine viruses where antibody titers in mice continued to rise over 8 weeks following immunization with a single dose of H5N1 ca vaccine viruses but, the final titer achieved was lower than that induced by 2 doses of the vaccine viruses (Suguitan et al., 2006). Despite eliciting an increase in homologous neutralizing antibody titers, two doses of the vaccine did not increase the breadth of cross-reactivity of neutralizing antibodies against heterologous H7 viruses from the Eurasian lineage. The pattern of antibody response induced by the H7N3 BC 04 ca virus was similar to that of H9N2 ca vaccine virus. Intranasal administration of a single dose of an H9N2 ca vaccine virus induced detectable levels of hemagglutination inhibition (HI) antibodies in ferrets and mice (Chen et al., 2003), but the titers induced by the vaccine virus were lower than those induced by the wt H9N2 parent virus (Chen et al., 2003). The lower titers of HI and neutralizing antibody induced in mice and ferrets by infection with the ca viruses compared to the wt parent viruses may be a consequence of the reduced antigen load presented by the attenuated virus due to restricted replication. In contrast to H7N3 BC 04 ca and H9N2 ca vaccine viruses, a single dose of H5N1 ca vaccine virus was poorly immunogenic in mice. Although anti-H5 HA ELISA antibodies were detected in the serum, HI and virus neutralizing antibody titers were low. A second dose of H5N1 ca vaccine elicited a significant boost in HI and Nt Ab response (Suguitan et al., 2006).

The efficacy of the H7N3 BC 04 ca vaccine virus was demonstrated in mice and ferrets. Viruses for the challenge studies were selected based on our evaluation of mice infected with H7 viruses (Joseph et al., 2007). A single dose of the vaccine completely protected mice from lethal challenge with homologous and heterologous HP viruses from the North American and Eurasian lineages and prevented dissemination of the challenge viruses to the brain. However, protection from viral replication in the respiratory tract of mice following a single dose of vaccine was only observed in mice that were challenged with viruses from the North American lineage. Despite an increase in homologous neutralizing antibody titers over 8 weeks following a single dose of vaccine, the breadth of protection conferred by the vaccine was not significantly enhanced 4 weeks later and mice were still not completely protected from pulmonary replication of the challenge viruses from the Eurasian lineage. Although sera collected following 2 doses of vaccine did not cross-neutralize heterologous viruses from the Eurasian lineage, mice that received 2 doses of vaccine were protected from replication of the heterologous viruses. Thus, a statistical correlation could not be established between the heterologous neutralizing antibody titers and level of pulmonary replication of heterologous challenge viruses from the Eurasian lineage in mice, suggesting that other arms of the immune system such as mucosal antibodies or cellular immunity may contribute to protection.

A single dose of the vaccine conferred protection from replication of the homologous LP virus in the lower respiratory tract of ferrets. A single dose of the vaccine also significantly reduced the levels of replication of homologous and heterologous HP viruses in the upper and lower respiratory tract of ferrets and effectively prevented dissemination of the challenge viruses to the brain and olfactory bulb. Two doses of vaccine further reduced titers of the challenge viruses in the upper respiratory tract but complete protection from the replication of all challenge viruses was not observed in the lungs despite the increase in the homologous antibody titers. In ferrets, the number of doses of vaccine (1 or 2) and rise in the homologous antibody titers did not affect the titers of the HP challenge viruses in the lower respiratory tract.

We have previously evaluated the efficacy of H5N1 ca and H9N2 ca vaccine candidates in mice and ferrets. As was seen in the case of H7N3 BC 04 ca vaccine virus, a single dose of H5N1 ca vaccines provided complete protection from lethality following challenge with homologous and antigenically distinct heterologous H5N1 wt viruses (Suguitan et al., 2006). Although, a single dose of the H9N2 ca vaccine protected mice from pulmonary replication of the homologous and antigenically heterologous H9N2 wt challenge viruses, 2 doses of H5N1 ca vaccines were needed to provide complete protection from pulmonary replication of homologous and antigenically distinct heterologous H5N1 wt viruses in mice and in ferrets (Chen et al., 2003; Suguitan et al., 2006).

In summary, the H7N3 BC 04 ca vaccine virus was attenuated in mice, ferrets, and chickens and was immunogenic in mice and ferrets. The vaccine was also efficacious in protecting mice and ferrets from challenge with homologous and antigenically distinct heterologous H7 viruses from the Eurasian and North American lineages. Comparable attenuation, immunogenicity, and efficacy data in mice and ferrets support the use of the mouse model for preclinical evaluation of vaccines against avian influenza A H7 subtype viruses. Based on the promising preclinical data in this report, careful clinical evaluation of the H7N3 BC 04 ca vaccine is warranted.

Material and methods

Viruses

The LP avian influenza virus H7N3 BC 04 LP wt was used to derive the HA and NA genes for the development of the live attenuated ca reassortant vaccine virus H7N3 BC 04 ca. The 6 internal protein gene segments of H7N3 BC 04 ca were derived from the ca influenza A vaccine donor strain, AA ca (MedImmune, Mountain View, CA). The AA wt virus that was used for this study was provided by MedImmune. Additional viruses used for the efficacy study include HP avian influenza virus H7N3 BC 04 HP wt, LP viruses A/turkey/VA/55/02 (VA/02) (H7N2), and A/turkey/UT/24721-10/95 (UT/95) (H7N3) from the North American lineage and HP viruses A/Netherlands/219/03 (NL/03) (H7N7) and A/turkey/England/63 (EG/63) (H7N3) from the Eurasian lineage. These viruses were kindly provided by Dr. Robert G. Webster, St. Jude Children’s Research Hospital, Memphis, TN; Dr. David Swayne, Southeast Poultry Research Laboratory, USDA, Athens, GA; Dr. Nancy Cox, Influenza Division, Centers for Disease Control and Prevention, Atlanta, GA; and Dr. John Pasick, Canadian Food Inspection Agency, National Centre for Foreign Animal Disease, Winnipeg, Canada.

The wt and reassortant ca viruses were propagated in the allantoic cavity of 10-day-old SPF embryonated chicken eggs at 37°C and 33°C, respectively, and stored at −80°C until use. TCID₅₀ were determined by titration of serial 10-fold dilutions of the viruses in Madin-Darby canine kidney (MDCK) cells and titers were determined by the method described by Reed and Muench (Reed and Muench, 1938).

All experiments, including animal studies with LP and HP H7 avian influenza viruses, were conducted in BSL-3+ containment laboratories approved for use by the US Department of Agriculture and Centers for Disease Control and Prevention. Animal experiments were approved by the National Institutes of Health Animal Care and Use Committee.

Cloning of HA and NA expression plasmids

The HA and NA genes of H7N3 BC 04 LP wt were amplified by RT-PCR using Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA) and pfu DNA polymerase (Stratagene, La Jolla, CA). The HA vRNA was reverse transcribed with primer H7RT (5’-AGCAAAAGCAGGGGATACAAAATG) and PCR amplified with primers BCH7F-Aar (5’cacttatattcacctgcctcagggAGCAAAAGCAGGGG) and BCH7R-Aar (5’cctaacatatcacctgcctcgtattAGTAGAAACAAGGGTGTT). The NA vRNA was reverse transcribed with BCN3RT (5’-AGCAAAAGCAGGTGCG) and PCR amplified with primers BCN3F-Aar (5’-cacttatattcacctgcctcagggAGCAAAAGCAGGTGCG) and BCN3R-Aar (5’-ctaacatatcacctgcctcgtattAGTAGAAACAAGGTGCTT). The HA and NA cDNAs were cloned into the pAD3000 expression vector, a derivative of plasmid pHW2000 (Hoffmann et al., 2000), that contains pol I and pol II promoters but in which the bovine growth hormone polyadenylation signal was replaced by the SV40 late mRNA polyadenylation sequence. Both the HA and NA primer pairs contained Aar I restriction sites that were designed to be compatible with the BsmB I sites present in the pAD3000 plasmid. The HA and NA plasmids were analyzed by restriction enzyme digestion and sequence analysis. Plasmids containing viral cDNAs that were identical to the consensus sequences of the wt vRNA were selected for virus rescue experiments.

Generation of reassortant ca virus

The reassortant ca virus was generated by plasmid-based reverse genetics. A recombinant 6:2 reassortant that contained the HA and NA gene segments of H7N3 BC 04 LP wt and internal protein genes of AA _ca was generated in Vero cells by electroporation using a Bio-Rad Gene Pulser Xcell System (Bio-Rad, Hercules, CA) as previously described (Hoffmann et al., 2000; Suguitan et al., 2006). Seed virus stocks were generated by three rounds of cloning by limiting dilution in SPF eggs, followed by expansion in SPF eggs. The eight gene segments of the reassortant virus were sequenced and verified.

Evaluation of phenotypic properties of the reassortant virus

The reassortant H7N3 BC 04 ca virus was assessed for temperature ts and ca properties. The ts phenotype of the reassortant virus was assessed by evaluating viral replication by cytopathic effect (CPE) in PCK cells (Charles River Laboratories) at 33°C (permissive) and 39°C (non-permissive temperature). Virus was grown in PCK cells in MEM (Invitrogen) supplemented with non-essential amino acids (Invitrogen), L-glutamine (Invitrogen), and 2x concentration of antibiotic-antimycotic (penicillin, streptomycin, and amphotericin B) (Invitrogen). Cells were examined for CPE 6 dpi. Viruses that displayed ≥100-fold reduction in titer at 39°C compared with that observed at 33°C are considered ts.

The ca phenotype of the parent and reassortant viruses was determined by comparing their infectivity in PCK cells at 25°C and 33°C. Cells incubated at 33°C and 25°C were examined for CPE at 6 and 10 dpi respectively. The ca phenotype is defined as less than 100-fold reduction of virus titer at 25°C as compared to virus titer at 33°C.

Plaque assay in the presence and absence of trypsin

Chicken embryo fibroblast cells in 6-well tissue culture plates were inoculated with 0.1 ml of virus serially diluted in Leibowitz (L-15) medium. The H7N3 BC 04 HP wt and H7N3 BC 04 ca viruses were adsorbed for 1 h, with shaking every 15 minutes. Wells were overlaid with 1.8% w/v Bacto agar (Difco) mixed 1:1 with 2X Medium 199 containing antibiotics and fungizone, with or without 0.1 □g/ml porcine trypsin (Sigma). Plates were incubated at 33°C and 37°C for 4 – 5 days. Agarose overlays were carefully dislodged and the cell monolayers were fixed with 100% methanol. Cell monolayers in each well were rinsed with 2 ml of Blotto (5% non-fat dry milk in 1x PBS) and incubated for 1 h at room temperature with 0.5 ml/well of 1:1000 of the primary antibody in Blotto (Chicken anti-AA ca antibody, MedImmune). Plates were washed 3 times with Blotto and incubated with 0.5 ml/well of 1:2000 of rabbit anti-chicken Y H+L horseradish peroxidase (HRP) (Pierce, Rockford, IL) secondary antibody in Blotto for 1 hr at room temperature. Plates were washed 3 times with PBS and plaques were stained using 0.5 ml/well of AEC substrate solution (DAKO, Carpinteria, CA) at 37°C for 30–45 min. Plates were washed once with water and dried at room temperature. The number of plaques in each well was counted and titers were recorded in plaque forming units (pfu) per ml.

Pathogenicity and infectivity studies in chickens

An intravenous pathogenicity test (USAHA 1994) was used to determine the pathogenicity of the H7N3 BC 04 ca virus for chickens. Eight 4-week old White Leghorns chickens were inoculated i.v. with the H7N3 BC 04 ca virus at a dose of 0.2 ml of a 1:10 dilution of stock virus and were monitored for mortality up to 10 dpi. To determine infectivity via a simulated natural route of exposure, a separate group of 10 chickens were inoculated i.n. with 10⁶ TCID₅₀ of the H7N3 BC 04 ca virus. Oropharyngeal and cloacal swabs were collected for virus isolation 3 dpi. Two birds from the i.n. inoculated group were euthanized 3 dpi for virus isolation from kidney, heart, brain, and lungs. All the remaining chickens were euthanized and bled 14 dpi and sera were tested for evidence of seroconversion by agar gel precipitin assay.

Pathogenicity and replication studies in mice

To assess the lethality in mice, groups of 5 four to six week-old female BALB/c mice (Taconic Farms, Hudson, NY) were lightly anesthetized and infected i.n. with 10⁶ TCID₅₀/mouse of the H7N3 BC 04 LP wt and H7N3 BC 04 HP wt viruses and serial 10-fold dilutions ranging from 10⁰ TCID₅₀ to 10⁶ TCID₅₀/mouse of the H7N3 BC 04 ca virus in a volume of 50 µl. Mice were monitored and weighed daily for 14 days.

To study the ability of the H7N3 BC 04 ca virus to replicate in different tissues, groups of 12 female BALB/c mice were inoculated i.n. with 10⁵ TCID₅₀ of H7N3 BC 04 HP wt, H7N3 BC 04 LP wt, H7N3 BC 04 ca, AA wt, and AA ca viruses. Four mice from each group were euthanized on 2, 3, and 4 dpi. Nasal turbinates, lungs, brains, and spleens were harvested, weighed, and homogenized in L-15 medium (Invitrogen) containing 2x concentration of antibiotic-antimycotic (penicillin, streptomycin, and amphotericin B) (Invitrogen) to make a 5% (nasal turbinates and spleen) or 10% (lungs and brain) w/v tissue homogenate. Tissue homogenates were clarified by centrifugation and titrated in 24- and 96-well tissue culture plates containing MDCK cells. Virus titer in each organ was determined by the method described by Reed and Muench (1938) and expressed as log₁₀TCID₅₀/gm of tissue. Log-transformed viral titers were compared using the Mann-Whitney U test.

Replication in ferrets

Six to seven week-old ferrets (Triple F, Sayre, PA, USA) that were sero-negative for antibodies to currently circulating H3N2 and H1N1 human influenza viruses were used to compare the ability of wt and ca viruses to replicate in ferret tissues. Groups of 3 ferrets were inoculated with H7N3 BC 04 LP wt, H7N3 BC 04 ca, and AA ca viruses and 2 ferrets were inoculated with H7N3 BC 04 HP wt and AA wt viruses. Each ferret was inoculated i.n. with 10⁷ TCID₅₀ of the virus in a volume of 0.5 ml (0.25 ml per nostril). Ferrets were monitored daily for clinical signs of influenza and body temperatures. Ferrets were euthanized at 3 dpi and nasal turbinates, lungs (left lower lobe), and portions of the brain were harvested. Tissues were homogenized in L-15 medium (Invitrogen) containing a 2x concentration of antibiotic-antimycotic (penicillin, streptomycin, and amphotericin B) (Invitrogen) to make a 5% (nasal turbinates) or 10% (lung and brain) w/v tissue homogenate. Serial 10-fold dilutions of the tissue homogenates were prepared and 0.1 ml was inoculated into four 9–11 day-old embryonated SPF hen’s eggs. Eggs were incubated at 35°C for 24 hours for the H7N3 BC 04 HP wt virus or 72 hours for the remaining viruses. Allantoic fluid from each egg was subjected to a hemagglutination assay using 0.5% turkey erythrocytes. Virus titers were determined as EID₅₀/gm of tissue using the Reed-Muench method (1938). Log-transformed viral titers were compared using the Mann-Whitney U test.

Evaluation of immunogenicity of ca virus in mice and ferrets

Groups of 3 four to six-week old female BALB/c mice were immunized i.n. with 1 or 2 doses at a 4 week interval of 50 µl containing 10⁶ TCID₅₀ of the H7N3 BC 04 LP wt or H7N3 BC 04 ca viruses. Sera were collected on 0, 28, and 56 dpi. Similarly, 7-week old ferrets that were sero-negative for antibodies to currently circulating H3N2 and H1N1 human influenza viruses in groups of 3 were immunized i.n. with 1 or 2 doses at a 4 week interval of 200 µl (100 µl/nostril) containing 10⁷ TCID₅₀/ferret of the H7N3 BC 04 ca virus and 1 dose of the H7N3 BC 04 LP wt virus. Sera were collected on 0, 28, and 56 dpi and 0 and 28 dpi from ferrets that were immunized with H7N3 BC 04 ca virus and H7N3 BC 04 LP wt virus respectively. Neutralizing antibody titers in the post-immunization sera against the homologous and heterologous viruses were determined in a microneutralization (MN) assay. Serial two-fold dilutions of heat-inactivated serum were prepared starting from 1:10 or 1:20 dilution. Equal volumes of serum and virus (100 TCID₅₀) were mixed and incubated for 60 minutes at room temperature. The residual infectivity of the virus-serum mixture was determined in MDCK cells in four replicates for each dilution. Neutralizing antibody titer was defined as the reciprocal of the highest dilution of serum that completely neutralized the infectivity of 100 TCID₅₀ of the virus as determined by the absence of CPE at day 4.

Evaluation of the efficacy of H7N3 BC 04 ca virus vaccine in mice

Protection from lethality

Groups of four to six-week old female BALB/c mice were immunized i.n. with 1 or 2 doses at a 4 week interval with 10⁶ TCID₅₀ of H7N3 BC 04 ca or L-15 medium (mock-immunized). Groups of 8 mice were challenged 4 and 8 weeks post-dose 1 and 4 weeks postdose 2 with 50 µl containing 50 times the 50% lethal dose (50 LD₅₀) of the homologous H7N3 BC 04 HP wt, and heterologous NL/03 (H7N7) and EG/63 (H7N3) HP viruses (Joseph et al., 2007). Mice were monitored and weighed daily for 14 days post-challenge.

Protection from replication of the challenge virus in mice

The levels of pulmonary replication and extra-pulmonary spread of the challenge viruses were also evaluated. Groups of four to six-week old female BALB/c mice were immunized i.n. with 50 µl containing 10⁶ TCID₅₀ of H7N3 BC 04 ca virus or L-15 medium (mock-immunized) in 1 or 2 doses 4 weeks apart. Mice were challenged 4 or 8 weeks post-dose 1 and 4 weeks postdose 2 with 10⁵ TCID₅₀ of the homologous H7N3 BC 04 LP wt and H7N3 BC 04 HP wt viruses and the heterologous VA/02 (H7N2) and UT/95 (H7N3) viruses from the North American lineage and heterologous NL/03 (H7N7) and EG/63 (H7N3) viruses from the Eurasian lineage. Nasal turbinates, lungs, and brain were harvested at 4 days post-challenge. Tissue homogenates 26 were prepared as described before and titrated in MDCK cells. Virus titer in each organ was expressed as log₁₀TCID₅₀/gm of tissue from groups of 4 mice. Log-transformed viral titers were compared using the Mann-Whitney U test.

Evaluation of the efficacy of H7N3 BC 04 ca virus vaccine in ferrets

Groups of three 7-wk old ferrets that were sero-negative for antibodies to currently circulating H3N2 and H1N1 human influenza viruses were immunized i.n. with 1 or 2 doses at a 4 week interval with 10⁷ TCID₅₀ of H7N3 BC 04 ca or L-15 medium (mock-immunized), in a volume of 200 µl/ferret (100 µl /nostril). Ferrets were challenged 4 weeks post-dose 1 and 2 with 10⁶ TCID₅₀ of the homologous H7N3 BC 04 LP wt and H7N3 BC 04 HP wt viruses from the North American lineage and heterologous NL/03 (H7N7) virus from the Eurasian lineage. Ferrets were euthanized 3 days post-challenge and nasal turbinates, lungs (right and left lower lobes), and brain were harvested and homogenized. Tissue homogenates of nasal turbinates and brain tissues were titrated in MDCK cells and virus titers were determined and expressed as log₁₀TCID₅₀/gm of tissue. Because embryonated eggs are more sensitive for detecting low concentrations of virus, lung homogenates from the ferrets were titrated in embryonated eggs and virus titers were expressed as EID₅₀/gm of tissue. Log-transformed viral titers were compared using the Mann-Whitney U test.