Extending the Cytoplasmic Tail of the Influenza A Virus M2 Protein Leads to Reduced Virus Replication In Vivo but Not In Vitro

Abstract

A carboxy-terminal epitope tag introduced into the coding region of the A/WSN/33 M2 protein resulted in a recombinant virus (rWSN M2myc) which replicated to titers similar to those of the parental virus (rWSN) in MDCK cells. The rWSN M2myc virus was attenuated in its ability to induce mortality and weight loss after the intranasal inoculation of BALB/c mice, indicating that the M2 cytoplasmic tail plays a role in virus virulence. Mice infected with rWSN M2myc were completely protected from subsequent challenge with rWSN, suggesting that epitope tagging of the M2 protein may be a useful way of attenuating influenza A virus strains.

Vaccination against influenza A virus is an effective means of controlling morbidity and mortality resulting from annual influenza epidemics (3, 16). An egg-grown, inactivated virus vaccine is the most commonly used influenza vaccine, but a live, attenuated vaccine is also available to most but not all of the general public (1). The immune response induced by the live, attenuated vaccine is believed to be superior to that induced by inactivated virus, resulting in increased protection against antigenic-drift variants and other antigenic subtypes (4, 5).

The influenza A virus M2 protein plays an important role during the entry of influenza A virus into susceptible cells by functioning as a proton channel, which allows for the acidification of the virion interior (17). The M2 cytoplasmic tail can alter the efficiency of the proton channel activity (22), but the addition of an antibody epitope tag at the carboxy terminus of the protein has no demonstrable effect on proton channel activity (18).

Recently, a role for the M2 cytoplasmic tail in infectious-virus production has also been proposed. Truncations of 5 amino acids or less in the M2 cytoplasmic tail had a limited effect on virus replication in vitro; however, larger truncations resulted in a drastic decrease in infectious virus production (2, 8, 11, 12).

Epitope tagging of the M2 protein and in vitro virus replication.

The addition of an antibody epitope tag to the M2 protein cytoplasmic tail does not alter the ion channel activity or expression of the protein (18). In order to further characterize the role of the M2 cytoplasmic tail in the virus life cycle, a recombinant influenza A/WSN/33 virus containing the myc epitope (amino acids EQKLISEEDL; nucleotide sequence 5′-GAGCAGAAGCTGATCTCCGAGGAAGACCTG-3′) (Fig. 1A) at the carboxy terminus of the M2 protein (rWSN M2myc) was generated using standard techniques (as described in references 12 and 13). The replication of this virus after infection of the MDCK cells at a multiplicity of infection (MOI) of 0.001 50% tissue culture infectious dose (TCID₅₀)/ml (methods are described in detail in references 10, 12, and 14) was slightly reduced at early times postinfection but reached titers identical to those of the parental virus (rWSN) at late times postinfection (Fig. 1B). The plaque diameters of the viruses were also nearly identical (Fig. 1C), suggesting that the addition of the amino acids to the M2 protein did not have an adverse effect on virus replication in MDCK cells.

FIG. 1.

In vitro replication of recombinant influenza viruses. (A) Schematic diagram indicating the coding strategy of influenza A virus RNA segment 7 and the sequence and positioning of the myc epitope with respect to the encoded proteins. aa, amino acids; *, stop codon. (B) MDCK cells were infected with the indicated viruses at an MOI of approximately 0.001 TCID₅₀. After a 1-h incubation, the inoculum was removed, and the cells were washed with medium and incubated at 37°C. At the indicated times, the infected-cell supernatant was sampled, and infectious-virus titers were determined by TCID₅₀. The mean and standard error for each time point are graphed. The dashed horizontal line is the limit of detection. The experiment was performed twice at this MOI and twice at an MOI of 0.01 TCID₅₀. (C) The indicated viruses were analyzed by a plaque assay of MDCK cells. After harvesting of the cells at 3 days postinfection, the plaque diameters were measured with a micrometer. The solid horizontal line indicates the average plaque diameter, and n indicates the number of plaques measured.

Virulence of the M2 epitope-tagged viruses.

To assess the virulence of rWSN M2myc, 6- to 8-week-old BALB/c mice were anesthetized and inoculated intranasally with various doses (10³, 10⁴, or 10⁵ PFU) of rWSN M2myc (methods are described in reference 15). Surprisingly, there was no mortality of mice infected with any dose of the rWSN M2myc virus, but the corresponding doses of rWSN showed the expected degree of mortality (Fig. 2A). Mice inoculated with rWSN M2myc had significantly reduced weight loss compared to that of mice inoculated with rWSN, again indicating that the rWSN M2myc virus had reduced virulence in the mouse model of infection (Fig. 2C). When the viral loads in the tracheas (Fig. 2E) and lungs (Fig. 2F) of virus-infected mice were determined, it was clear that the rWSN M2myc virus replicated to titers lower than those of the parental rWSN virus. Taken together, the data indicate that the addition of a myc epitope tag to the carboxy terminus of M2 significantly attenuates virus replication and virulence in mice.

FIG. 2.

In vivo replication and pathogenesis of recombinant influenza A viruses. (A and B) Mice were inoculated intranasally with the indicated viruses, and mortality was assessed over 2 weeks. (C and D) Mice inoculated intranasally with the indicated viruses were weighed daily as a surrogate for virus-induced morbidity. Mouse weights were normalized to their weights at the time of infection, and the averages and standard errors are graphed. (E and F) At the indicated days (d.) postinfection with 10⁵ PFU of virus, tracheas (E) and lungs (F) were harvested and virus loads determined by TCID₅₀. The virus load in each tissue is graphed as the number of TCID₅₀ per gram of tissue, and the solid horizontal line indicates the average. The dashed horizontal line is the limit of detection. Virus doses used to inoculate the mice are given in numbers of PFU, and n indicates the number of mice.

In order to determine whether the addition of amino acids to the M2 cytoplasmic tail or simply the presence of additional nucleotide sequences in the engineered virus mediated the loss of virus virulence, a virus containing a stop codon inserted in place of the codon for the first amino acid of the myc sequence (rWSN M2stopmyc) was generated. This resulted in the production of an M2 protein that did not contain the myc epitope tag but that did contain the nucleotide sequence for the epitope in the viral RNA species (data not shown). In MDCK cell cultures, this virus replicated in a manner indistinguishable from that of rWSN after administration at a low MOI (Fig. 1B) and with respect to plaque size (Fig. 1C). When 10⁵ PFU of rWSN M2stopmyc was administered intranasally to mice, the virus regained the ability to cause mortality (Fig. 2B) and induce weight loss (Fig. 2D), similarly to rWSN. These data indicate that the presence of the amino acids at the carboxy terminus of the M2 protein but not the additional nucleotides present in the viral RNA was responsible for attenuating the virulence of the virus in vivo.

M2 epitope-tagged viruses as vaccines.

Since the rWSN M2myc virus replicated but did not cause significant morbidity or mortality in infected mice, its potential as a live, attenuated influenza vaccine was assessed. Mice that had been infected with rWSN M2myc were challenged 28 days postinfection with a lethal dose of rWSN. Irrespective of virus dose, animals that were previously infected with rWSN M2myc survived a lethal rWSN challenge (Fig. 3A) and had very little weight loss (Fig. 3B), unlike the age-matched, naive mice. The protection afforded by rWSN M2myc infection was equivalent to that seen in animals that were given a sublethal (10³ PFU) dose of rWSN (Fig. 3A and B), although it must be noted that the morbidity induced by a sublethal rWSN infection was significantly greater than that observed with any dose of rWSN M2myc (Fig. 2C). Serum samples were collected from the mice either before or 28 days after rWSN challenge, and the neutralizing antibody titers against rWSN were determined. The neutralizing antibody titers in rWSN M2myc-immunized mice were comparable to those in mice immunized with a sublethal dose of rWSN (Fig. 3C). Neutralizing antibody titers rose slightly after the challenge (Fig. 3C). Taken together, the data indicate that the addition of a carboxy-terminal epitope tag to the M2 protein results in significant attenuation of virus virulence but that the immune response induced by virus infection can protect against a lethal influenza A virus challenge.

FIG. 3.

Infection with rWSN M2myc protects mice from a lethal challenge with rWSN. (A) Mice previously infected with rWSN M2myc were challenged with an rWSN intranasal dose of 10⁵ PFU (approximately 20 times the 50% mouse lethal dose) and monitored for mortality (A) and weight loss (B). All rWSN M2myc-infected animals are graphed together under mortality, as no deaths were documented for any group. (B) Mouse weights were normalized to their weights at the time of infection, and the averages and standard errors are graphed. The virus doses used to inoculate the mice are given in numbers of PFU. (C) Serum samples were collected from the mice that survived infection with either rWSN or rWSN M2myc. Sera were also collected from the immunized mice that survived a subsequent challenge with rWSN and pooled. Dilutions of sera were incubated with 10³ PFU of rWSN and then analyzed in sextuplet on the basis of the TCID₅₀. At 4 days postinfection, the dishes were stained with naphthol blue-black and the serum concentration which inhibited the cytopathic effect in 50% of the wells was determined. The experiment was repeated three times with similar results.

Engineering a recombinant influenza A virus that expresses an M2 protein with a carboxy-terminal myc epitope tag resulted in no change in virus yield after infection of MDCK cells but significant reductions in virulence using the mouse model of infection. This method for attenuating influenza virus virulence could have important implications for the design of live, attenuated vaccines, and experiments testing whether this strategy can attenuate a variety of influenza A virus strains are under way. Since virus yield in vitro is not altered, the vaccine yields will not be adversely affected by this method of attenuation. The engineered virus replicates in vivo but does not cause significant morbidity or mortality, which may allow for the generation of cell-mediated as well as humoral immune responses which can broaden the protective response. The reasons for this in vivo attenuation are not clear, but they may reflect the fact that influenza A virus particle assembly may have additional requirements in vivo. Influenza A virus infects and replicates in the ciliated cells of mouse tracheas and bronchioles (7), cells which are quite different morphologically and functionally from MDCK cells.

The ability to modify the M2 protein of a circulating vaccine strain in order to attenuate it allows for the inclusion of all the coding regions of the virus, which allows for maximal protection against all virally encoded epitopes. The inclusion of an antibody epitope as a unique marker in the vaccine strain allows for more-efficient differentiation of vaccinated animals from infected animals, which may be of particular use in veterinary applications (20).

We have isolated and sequenced the M2 open reading frames from 66 plaques isolated from the lungs of three mice at day 5 postinfection. In each case, the M2 epitope tag was still present. We also sequenced infected-cell supernatant from MDCK cells infected with rWSN M2myc (MOI = 0.001 TCID₅₀) at 48 and 60 h postinfection and confirmed that the M2 epitope was still present (data not shown). The genetic stability of the epitope-tagged M2 protein after repeated passage in mice and MDCK cells clearly needs to be tested, but these initial results suggest that the selective pressure against viruses encoding epitope-tagged M2 does not lead to the immediate selection of revertant viruses. The identification of second-site mutations that restore virus virulence without eliminating the M2 epitope tag could also provide insights into the mechanism responsible for attenuating the virus.

A single, attenuating mutation may not be sufficient to ensure a lack of virus reversion, but epitope tagging of the M2 protein in combination with other attenuating mutations, such as deletions in the NS1 coding region (21), alterations in hemagglutinin cleavage sites (19), and mutations that confer cold adaption (6, 9), could allow for the rapid engineering of vaccine strains which are nearly identical to the circulating viruses but which possess various levels of attenuation in vivo. While the molecular basis for rWSN M2myc attenuation, as well as the nature and composition of the sequences which can be added to the M2 cytoplasmic tail, needs to be better defined, this approach to attenuating virus virulence has the potential to dramatically improve vaccine efficacy as well as shorten the time needed to generate master vaccine strains.