Influenza A M2 recruits M1 to the plasma membrane: A fluorescence fluctuation microscopy study

Abstract

Influenza A virus (IAV) is a respiratory pathogen that causes seasonal epidemics with significant mortality. One of the most abundant proteins in IAV particles is the matrix protein 1 (M1), which is essential for the virus structural stability. M1 organizes virion assembly and budding at the plasma membrane (PM), where it interacts with other viral components. The recruitment of M1 to the PM as well as its interaction with the other viral envelope proteins (hemagglutinin [HA], neuraminidase, matrix protein 2 [M2]) is controversially discussed in previous studies. Therefore, we used fluorescence fluctuation microscopy techniques (i.e., scanning fluorescence cross-correlation spectroscopy and number and brightness) to quantify the oligomeric state of M1 and its interactions with other viral proteins in co-transfected as well as infected cells. Our results indicate that M1 is recruited to the PM by M2, as a consequence of the strong interaction between the two proteins. In contrast, only a weak interaction between M1 and HA was observed. M1-HA interaction occurred only in the event that M1 was already bound to the PM. We therefore conclude that M2 initiates the assembly of IAV by recruiting M1 to the PM, possibly allowing its further interaction with other viral proteins.

Figure 1

Membrane recruitment of IAV matrix protein 1 (M1) in co-transfected and infected cells. (A and B) Representative confocal fluorescence images of HEK293T cells expressing M1-mEGFP (green) from the influenza A/FPV/Rostock/1934 strain (FPV) alone (A, left). The same construct was also observed in cells co-transfected with the reverse genetic plasmid system of FPV and unlabeled M2, here labeled “all” (A, middle; B, left) and in cells infected with FPV (A, right; B, left). (C) Representative confocal fluorescence images of HEK293T cells co-expressing M1-mEGFP (green) and the FPV M2 (mCherry2-M2, magenta). The right panels show the two channels merged in a single image. (D) Representative confocal fluorescence images of HEK293T cells co-expressing M1-mEGFP (green) and the hemagglutinin (mCherry2-HA_TMD, magenta) in the absence (upper panels) or in the presence (lower panels) of unlabeled M2. (E) Representative confocal fluorescence images of HEK293T cells co-expressing M1-mEGFP (green) and the neuraminidase (NA-mCherry2, magenta) in the absence (upper panels) or in the presence (lower panels) of unlabeled M2. Scale bars represent 10 μm.

Figure 2

M1 oligomerizes in a concentration-dependent manner. N&B analysis of M1-mEGFP in cells expressing only M1-mEGFP, infected with FPV, co-transfected cells expressing unlabeled M2 and the reverse genetic plasmid system for all other FPV proteins (“all”), or co-transfected cells expressing unlabeled M2. Oligomerization and surface concentration values were obtained as described in the section “materials and methods.” (A) Representative average intensity maps of M1-mEGFP in HEK293T cells. The average intensity map is visualized via color scale with units photon counts/dwell time. (B) Representative brightness-intensity maps corresponding to the images represented in (A). The images show pixel brightness as pixel color (counts/dwell time per molecule) and mean photon count rate as pixel intensity. (C) (Left) Box plot of single data points from three independent experiments showing the normalized brightness (i.e., oligomerization) for M1-mEGFP and the corresponding controls; i.e., cytosolic monomer mEGFP(1x), cytosolic dimer mEGFP(2x) in the cytosol of HEK293T cells. (C) (Right) Box plot of single data points from three independent experiments showing the oligomerization of M1-mEGFP at the PM of infected (M1-FPV) or co-transfected (M1-all, M1-M2) cells. Oligomerization values for PM-anchored controls are also shown: monomer mp-mEGFP(1x), dimer mp-mEGFP(2x). Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated at the bottom. Horizontal dotted lines corresponding to oligomerization values 1, 2, 5, and 7 are shown as guide to the eye. (D) M1-mEGFP oligomerization as a function of surface concentration at the PM (in N_monomer/μm²). The number of measured cells were: M1-FPV (n = 46), M1-all (n = 39), and M1-M2 (n = 53).

Figure 3

M2 interacts with M1 in a concentration-dependent manner. sFCCS of M1-mEGFP in HEK293T cells co-expressing mCherry2-M2, mCherry2-HA_TMD/M2-untagged, and NA-mCherry2/M2-untagged. Oligomerization, surface concentration (N_monomer/μm²), cross-correlation, and diffusion coefficient (μm²/s) values were obtained as described in the section,“materials and methods.” (A) Representative confocal fluorescence image of HEK293T cells co-expressing M1-mEGFP (green), mCherry2-HA_TMD (magenta), M2-untagged. Yellow arrow indicates the scanning path used for sFCCS. Scale bar represents 10 μm. (B) Box plot with single data points from three independent experiments shows the oligomerization of the controls monomer mp-mEGFP(1x) and dimer mp-mEGFP(2x), and M1-mEGFP co-expressed with mCherry2-M2, mCherry2-HA_TMD/M2-untagged, and NA-mCherry2/M2-untagged. Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. (C) Box plot with single data points from three independent experiments shows the oligomerization of the controls monomer mp-mCherry2(1x) and dimer mp-mCherry2(2x), and the viral surface proteins mCherry2-M2, mCherry2-HA_TMD, and NA-mCherry2 for the same samples described for (B). Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. (D and E) Scatter plots show the oligomerization of M1-mEGFP as a function of the oligomerization of mCherry2-M2 (D), and the surface concentration of M1-mEGFP as a function of the surface concentration of mCherry2-M2 (E). (F) Box plot with single data points from three independent experiments shows the ratio of the oligomerization, and the surface concentration of M2:M1. Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. (G) Box plot with single data points from three independent experiments shows the rel. cc for the controls negative control mp-mEGFP(1x)/mp-Cherry2 and positive control mp-mCherry2-mEGFP, and between M1-mEGFP and mCherry2-M2, mCherry2-HA_TMD, or NA-mCherry2. Cells expressing mCherry2-HA_TMD and NA-mCherry2 also expressed unlabeled M2. Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. Statistical significance was determined using one-way ANOVA multiple comparison test; ^∗∗∗∗p < 0.0001 compared with the negative control (CTRL−). (H) Box plot with single data points from three independent experiments shows the diffusion coefficient of the controls monomer mp-mEGFP(1x) and dimer: mp-mEGFP(2x), and M1-mEGFP co-expressed with mCherry2-M2, mCherry2-HA_TMD/M2-untagged, and NA-mCherry2/M2-untagged. Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. (I) Box plot with single data points from three independent experiments shows the diffusion coefficient of the controls monomer mp-mCherry2(1x) and dimer mp-mCherry2(2x), and the viral surface proteins mCherry2-M2, mCherry2-HA_TMD, and NA-mCherry2 for the same samples described for (H). Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph.

Figure 4

HA interacts with a membrane-associated M1 construct. (A) Schematic diagram of M1 constructs with N-terminal PM-targeting sequences. One construct has a myristoylation (orange) and palmitoylation (blue) motif (mp-M1-mEGFP), and the other has an additional poly-lysine motif (green letters, mp-KrΦ-M1-mEGFP). (B) Representative M1 subcellular localization images in transfected HEK293T cells expressing mp-M1-mEGFP (left side), or mp-KrΦ-M1-mEGFP (right side). Scale bars represent 10 μm. (C) Box plot with single data points from three independent experiments shows the cross-correlation for the controls negative control mp-mEGFP (1x)/mp-mCherry2 (1x) and positive control mp-mCherry2-mEGFP), and between M1-mEGFP (or mp-M1-mEGFP, or mp-KrΦ-M1-mEGFP) and mCherry2-HA_TMD, or NA-mCherry2. Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. Statistical significance was determined using one-way ANOVA multiple comparison test; ^∗∗p < 0.01, ^∗∗∗∗p < 0.0001 compared with M1-mEGFP/mCherry2-HA_TMD; ^####p < 0.0001 compared with M1-mEGFP/NA-mCherry2; ns, not significant. (D) Box plot with single data points from three independent experiments shows the diffusion coefficient of the monomer control (mp-mEGFP), and M1-mEGFP, mp-M1-mEGFP, and mp-KrΦ-M1-mEGFP co-expressed with mCherry2-HA_TMD, or NA-mCherry2. Median values and whiskers ranging from minimum to maximum values are displayed. Sample size, median, and IQR are indicated in the graph. Statistical significance was determined using one-way ANOVA multiple comparison test; ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001 compared with M1-mEGFP/mCherry2-HA_TMD; ^####p < 0.0001 compared with M1-mEGFP/NA-mCherry2.

Figure 5

M2-binding site on M1 is located in the N-terminal domain. (A) Schematic diagram of different M1 and M2 expression constructs. On top, M1 constructs showing the wild type and the truncated M1 variants with their domains: N-terminal domain (N domain, aa 1–67, blue), linker region (orange, aa 68–86), middle domain (M domain, aa 87–164, green), and C-terminal domain (C domain, aa 165–252, purple). An mEGFP was fused to the C terminus of each protein construct. On the bottom, M2 constructs showing the wild type and the M2 mutant (71-SMR-73 was replaced by three alanine) with their domains: ectodomain (ED, aa 1–25, blue), transmembrane domain (TMD, aa 26–46, orange), cytoplasmic tail (cyto-tail, aa 47–97, green). Each construct had an mCherry2 fused to the N-terminal site of M2. (B) Representative confocal fluorescence images of HEK293T cells expressing truncated M1-mEGFP variants: NM1_1–67, NM1_1–86, NM1_1–164, and CM1_165–252. (C) Representative confocal fluorescence images of HEK293T cells expressing truncated M1-mEGFP variants: NM1_1-67, NM1_1-86, NM1_1-164, and CM1_165-252 (green) in the presence of wild-type mCherry2-M2 (magenta). (D) Representative confocal fluorescence images of HEK293T cells expressing wild-type M1-mEGFP (green) with mCherry2-M2 mutant (M2_mut, magenta). Scale bars represent 10 μm.