Visualizing infection of individual influenza viruses

Abstract

Influenza is a paradigm for understanding viral infections. As an opportunistic pathogen exploiting the cellular endocytic machinery for infection, influenza is also a valuable model system for exploring the cell's constitutive endocytic pathway. We have studied the transport, acidification, and fusion of single influenza viruses in living cells by using real-time fluorescence microscopy and have dissected individual stages of the viral entry pathway. The movement of individual viruses revealed a striking three-stage active transport process that preceded viral fusion with endosomes starting with an actin-dependent movement in the cell periphery, followed by a rapid, dynein-directed translocation to the perinuclear region, and finally an intermittent movement involving both plus- and minus-end-directed microtubule-based motilities in the perinuclear region. Surprisingly, the majority of viruses experience their initial acidification in the perinuclear region immediately following the dynein-directed rapid translocation step. This finding suggests a previously undescribed scenario of the endocytic pathway toward late endosomes: endosome maturation, including initial acidification, largely occurs in the perinuclear region.

Fig. 1.

Fluorescence images and fusion sites of individual influenza viruses in CHO cells. (a) DIC image of a CHO cell (Left) and fluorescence images of DiD-labeled viruses 2 min (Center) and 15 min (Right) after infection. The thin white lines trace the nuclear boundary. (b) The fusion sites of individual viruses are marked by red stars. The boundaries of a cell and its nucleus are highlighted by thick gray and thin white lines, respectively. The boundaries between cells are discernible in the DIC image as distinct cleft-like contours. The nuclei are visible as ovoid regions near the cell centers surrounded by high-contrast vesicular structures. The stars seemingly inside the nuclei most likely indicate fusion events above or below the nuclei.

Fig. 2.

Tracking the transport and fusion of individual influenza viruses. (a) The trajectory of a DiD-labeled virus inside a cell. The color of the trajectory codes time with the colored bar indicating a uniform time axis from 0 s (black) to 500 s (yellow). The red star indicates the fusion site. (Scale bar: 10 μm.) (b) Time trajectories of the velocity (black) and the DiD fluorescence intensity (blue) of the virus. t₁, t₂, and t₃ are the durations of stages I, II, and III, respectively. Stage II movements can be consistently identified for each viral trajectories as the rapid unidirectional translocation from the cell periphery to the perinuclear region. Stage I is then defined as the period before this transient motion, and stage III is defined as the period after stage II but before fusion. (c) Histogram of the viral velocity in stage I. (Inset) Shown is the measured average mean square displacement (〈Δr²〉) vs. time (Δt) for a virus (green symbols). The green line is a fit to 〈Δr²〉 = constant + DΔt + (vΔt)² with D = 0.001 μm²/s and v = 0.02 μm/s. The small constant term is due to noise. About 60% of the viral trajectories in stage I show such superlinear dependence of 〈Δr²〉 on Δt. Because of the diffusion-like component of the movement (DΔt), the instantaneous speed in the histogram (=Δr/Δt) depends on Δt, which is chosen to be 0.5 s in c–e. (d) Histogram of the viral velocity in stage II. (e) Histogram of the viral velocity in stage III. (Inset) Shown is a typical example of 〈Δr²〉 vs. Δt. To represent the bursts of relatively fast movements in stage III, the 〈Δr²〉 vs. Δt plot was calculated by using only those points where the virus is traveling with speed >0.3 μm/s. The green line is a fit of the first eight data points to 〈Δr²〉 = constant + D′Δt + (v′Δt)² with D′ = 0.5 μm²/s and v′ = 0.4 μm/s. The small constant term is due to noise.

Fig. 3.

Effects of drugs and antibodies on viral transport properties. (a) Viral trajectories of the first 100 s of viral movement in untreated, cyto-D-treated, and nocodazole-treated cells. The arrow in Upper Right indicates the minimum radius (r_bound) of a circle centered at the origin of the trajectory that can cover the entire trajectory at a given time. (b) The average r_bound over many trajectories as a function of time. The r_bound in untreated and nocodazole-treated cells quickly exceeds the r_bound attained in cyto-D-treated cells. The rapid, initial increase of r_bound in the cyto-D-treated curve likely arises from restricted diffusion of the viruses or undulations of the cells, and does not occur for viruses immobolized on glass. (c) Viral trajectories (10–15 min) in untreated cells, nocodazole-treated cells, and cells injected with anti-dynein (1 mg/ml). The time-color scale is similar to that described in Fig. 2. (Scale bar: 10 μm.)

Fig. 4.

Tracking the transport and acidification of individual influenza viruses. (a) The trajectory of a Cy3/CypHer5-labeled virus inside a cell. The colored bar indicates a uniform time axis from 0 to 450 s. The green star indicates the initial acidification site. (Scale bar: 10 μm.) (b) Time trajectories of the velocity (black) and the ratio of CypHer5 and Cy3 fluorescent emission (red) of the virus. t₄ defines the time elapsed between the end of stage II and the initial acidification event. The pH values are labeled according to the calibrated pH dependence of the intensity ratio between CypHer 5 and Cy3 (data not shown). (c) Histograms of t₃ and t₄ as defined in Figs. 2b and 4b, respectively.

Fig. 5.

A model of the endocytic pathway toward late endosomes. (1) The virus is internalized and transported to the early endosome (EE) in an actin (AT)-dependent way (stage I movement). (2) The virus-containing endocytic compartment leaves the EE, still at the extracellular pH. This may occur either through a virus-bearing endocytic carrier vesicle (ECV) budding from the EE (13, 27) or the membrane-rich tubular region of the EE recycling to leave a more vesicular EE that contains the virus (33). (3) The ECV or vesicular EE is transported to the perinuclear region via a dynein-directed movement on a microtubule (MT) (stage II movement). (4) The ECV or vesicular EE matures into a maturing endosome (ME) by changing the membrane-bound motor protein activity (transition from stage II to stage III movement). (5) The endosome further matures by changing its pH from the extracellular value to pH ≈6 (initial acidification as indicated by the change in fluorescence ratio between CypHer 5 and Cy3 conjugated to viruses). (6) Further acidification brings the pH of the endosome to the late endosomal (LE) value, pH ≈5 (second acidification as indicated by viral fusion).