Influenza virus M2 protein ion channel activity helps to maintain pandemic 2009 H1N1 virus hemagglutinin fusion competence during transport to the cell surface

Abstract

The influenza virus hemagglutinin (HA) envelope protein mediates virus entry by first binding to cell surface receptors and then fusing viral and endosomal membranes during endocytosis. Cleavage of the HA precursor (HA0) into a surface receptor-binding subunit (HA1) and a fusion-inducing transmembrane subunit (HA2) by host cell enzymes primes HA for fusion competence by repositioning the fusion peptide to the newly created N terminus of HA2. We previously reported that the influenza virus M2 protein enhances pandemic 2009 influenza A virus [(H1N1)pdm09] HA-pseudovirus infectivity, but the mechanism was unclear. In this study, using cell-cell fusion and HA-pseudovirus infectivity assays, we found that the ion channel function of M2 was required for enhancement of HA fusion and HA-pseudovirus infectivity. The M2 activity was needed only during HA biosynthesis, and proteolysis experiments indicated that M2 proton channel activity helped to protect (H1N1)pdm09 HA from premature conformational changes as it traversed low-pH compartments during transport to the cell surface. While M2 has previously been shown to protect avian influenza virus HA proteins of the H5 and H7 subtypes that have polybasic cleavage motifs, this study demonstrates that M2 can protect HA proteins from human H1N1 strains that lack a polybasic cleavage motif. This finding suggests that M2 proton channel activity may play a wider role in preserving HA fusion competence among a variety of HA subtypes, including HA proteins from emerging strains that may have reduced HA stability.

Importance: Influenza virus infects cells when the hemagglutinin (HA) surface protein undergoes irreversible pH-induced conformational changes after the virus is taken into the cell by endocytosis. HA fusion competence is primed when host cell enzymes cleave the HA precursor. The proton channel function of influenza virus M2 protein has previously been shown to protect avian influenza virus HA proteins that contain a polybasic cleavage site from pH-induced conformational changes during biosynthesis, but this effect is less well understood for human influenza virus HA proteins that lack polybasic cleavage sites. Using assays that focus on HA entry and fusion, we found that the M2 protein also protects (H1N1)pdm09 influenza A virus HA from premature conformational changes as it transits low-pH compartments during biosynthesis. This work suggests that M2 may play a wider role in preserving HA function in a variety of influenza virus subtypes that infect humans and may be especially important for HA proteins that are less stable.

FIG 1

Infectivities of HA-pseudoviruses with and without M2. (A) Infectivities of pseudoviruses bearing A/Mexico/4108/2009 (MX) HA (left) or A/PR/8/34 (PR) HA (right), with and without MX or PR M2. Data are shown as means and standard deviations for three independent experiments. (B) Infectivities of pseudoviruses bearing A-MLV Env, HIV-1 Env, or VSV G, with and without PR M2. (C) Western blots of M2 incorporation into pseudoviruses bearing MX HA, PR HA, A-MLV Env, VSV G, or HIV-1 Env. PR M2 was detected with an M2-specific rabbit antiserum.

FIG 2

Transmission electron microscopy of HA-pseudoviruses with negative staining. (A) MX HA-pseudovirus without M2; (B) MX HA-pseudovirus with MX M2; (C) MX HA-pseudovirus with PR M2.

FIG 3

M2 enhancement of HA-mediated cell-cell fusion. (A) Various levels of MX HA expressed in effector cells (0.5 to 4.5 μg of plasmid transfected) induced cell-cell fusion at pHs 7.4 and 5.0. The bottom left panel shows a Western blot of HA expression in effector cells. The right panels show a flow cytometry analysis of cell surface levels of HA in effector cells (0.5 to 4.5 μg of plasmid transfected). MFI, mean fluorescence intensity. (B) M2 enhanced MX HA-mediated cell-cell fusion at pH 5.0. (Left) A constant level of MX and PR M2 expression (6 μg of plasmid transfected) was tested with various levels of MX HA expression (1 to 4.5 μg of plasmid transfected). The fusion levels were normalized to the individual fusion levels induced by various HA levels without M2 expression. (Right) Various levels of MX and PR M2 expression (0 to 6 μg of plasmid transfected) were tested with a constant level of MX HA expression (1 μg of plasmid transfected). The fusion levels were normalized to the fusion level induced by HA alone without M2 expression. (C) Flow cytometry analysis of cell surface levels of HA in effector cells from the right part of panel B. The MX HA levels on the cell surface with various levels of MX M2 expression are shown in the top panels. The bottom panels show MX HA levels on the cell surface with various levels of PR M2 expression. (D) Western blots of total HA and M2 expression in effector cells from the right part of panel B. The top panels show MX M2 expression with various amounts of transfected DNA. The bottom panels show PR M2 expression with various amounts of transfected DNA. HA0 and HA1 were detected with rabbit HA1 antiserum. MX and PR M2 proteins were detected by use of M2-specific rabbit antiserum. Cell-cell fusion data are shown as means and standard deviations for three independent experiments.

FIG 4

M2 enhancement of MX HA-mediated cell-cell fusion across a range of pH values. (A) Relative cell-cell fusion levels mediated by MX HA at various pHs in the presence or absence of MX M2 (top) or PR M2 (bottom). For determinations of relative fusion, the fusion levels were normalized to the maximum fusion obtained with MX M2 (top) or PR M2 (bottom). (B) Fusion index comparisons. Fusion levels in each curve from panel A were normalized to the maximum fusion level for that curve. For example, the individual fusion levels at different pHs in the curve for MX HA were normalized to the maximum fusion level in the curve for MX HA. (C) Flow cytometry analysis of cell surface expression of HA in the effector cells used for panels A and B. Fusion data are shown as means and standard deviations for three independent experiments. Plasmids used for transfection included 1 μg of MX HA plasmid and 3 μg of M2 plasmid.

FIG 5

M2 proton channel function requirement for enhancement of MX HA-mediated cell-cell fusion. (A) Alignment of amino acids 21 to 35 of A/Udorn/72 (Udorn), PR, and MX M2 proteins. (B) MX HA-mediated cell-cell fusion at various pHs, with and without MX M2 proton channel function. (C) MX HA-mediated cell-cell fusion at various pHs, with and without PR M2 proton channel function. (Left) M2 proteins and M2 A30P mutants with proton channel dysfunction. (Middle) Controls (HA or HA plus M2), with and without amantadine. (Right) M2 mutants sensitive to amantadine, with and without amantadine. AM, 10 μM amantadine. Fusion data are shown as means and standard deviations for three independent experiments. Plasmids used for transfection included 1 μg of MX HA plasmid and 3 μg of M2 plasmid.

FIG 6

M2 proton channel function during HA biosynthesis accounts for the enhancement of HA-mediated cell-cell fusion. (A) Amantadine (AM) treatment windows. Treat 1, amantadine was maintained from 0.5 h post-HA and -M2 transfection of effector cells and thereafter, until the fusion assay ended, at the point of lysing cells; treat 2, amantadine was maintained from the time of low-pH treatment of effector-target cocultures and thereafter, until the fusion assay ended, at the point of lysing cells; treat 3, amantadine was maintained from the point right after the pH treatment of effector-target cocultures and thereafter, until the fusion assay ended, at the point of lysing cells. (B) Effects of amantadine treatment on MX M2-N31S enhancement of MX HA-mediated cell-cell fusion at pH 5.0. Amantadine (10 μM final concentration) was maintained according to treatments 1, 2, and 3. Fusion data are shown as means and standard deviations for three independent experiments. Plasmids used for transfection included 1 μg of MX HA plasmid and 3 μg of MX M2-N31S plasmid. *, P < 0.001 (compared to no AM treatment; t test).

FIG 7

M2 enhancement of TPCK-trypsin-activated HA-mediated fusion and M2 protection of HA from premature conformational changes. (A) TPCK-trypsin-digested HA induced cell-cell fusion at pH 5.0 in the presence or absence of MX M2. The fusion levels were normalized to the MX HA-mediated fusion level with TPCK-trypsin digestion. The bottom panels show Western blots of expression of HA, M2, and actin. Plasmids used for transfection included 1 μg of MX HA plasmid and 3 μg of MX M2 plasmid. (B) Nonreducing Western blots of TPCK-trypsin-digested pseudovirus HA in the presence or absence of MX or PR M2 or M2 with the A30P mutation. HA0 and HA fragments were detected with antisera against the HA2 C helix. Cell-cell fusion data are shown as means and standard deviations for three independent experiments. HA fragments were disulfide-bonded HA1 and HA2 proteolytic fragments.

FIG 8

Comparison of pH dependences of MX and PR HA-mediated cell-cell fusion. Fusion responses to pH were compared for MX and PR HA proteins. The fusion response to pH was normalized to the maximum fusion for that curve. Fusion data are shown as means and standard deviations for three independent experiments.