Plasma membrane protrusions mediate host cell-cell fusion induced by Burkholderia thailandensis

Abstract

Cell-cell fusion is important for biological processes including fertilization, development, immunity, and microbial pathogenesis. Bacteria in the pseudomallei group of the Burkholderia species, including B. thailandensis, spread between host cells by inducing cell-cell fusion. Previous work showed that B. thailandensis-induced cell-cell fusion requires intracellular bacterial motility and a bacterial protein secretion apparatus called the type VI secretion system-5 (T6SS-5), including the T6SS-5 protein VgrG5. However, the cellular-level mechanism of and T6SS-5 proteins important for bacteria-induced cell-cell fusion remained incompletely described. Using live-cell imaging, we found bacteria used actin-based motility to push on the host cell plasma membrane to form plasma membrane protrusions that extended into neighboring cells. Then, membrane fusion occurred within membrane protrusions either proximal to the bacterium at the tip or elsewhere within protrusions. Expression of VgrG5 by bacteria within membrane protrusions was required to promote cell-cell fusion. Furthermore, a second predicted T6SS-5 protein, TagD5, was also required for cell-cell fusion. In the absence of VgrG5 or TagD5, bacteria in plasma membrane protrusions were engulfed into neighboring cells. Our results suggest that the T6SS-5 effectors VgrG5 and TagD5 are secreted within membrane protrusions and act locally to promote membrane fusion.

FIGURE 1:

In one observed pathway, B. thailandensis spreads by inducing cell–cell fusion at the protrusion tip. (A and B) Live-cell imaging stills of two examples of BtGFP WT while inducing cell–cell fusion. A 1:1 mixture of A549 cells that expressed the plasma membrane marker TagRFP-T-farnesyl or cytoplasmic GFP was used. Times represent min:s post protrusion formation. Images taken at ∼16 hpi. Scale bars are 5 µm. White arrows highlight the bacterium forming the protrusion. Black arrows highlight the region of protrusion entry. Where GFP signal is difficult to see, insets with increased brightness are shown. (C) Model of cell–cell fusion occurring at the protrusion tip.

FIGURE 2:

In another observed pathway, B. thailandensis spreads by inducing cell–cell fusion elsewhere within the protrusion. (A and B) Live-cell imaging stills of two examples of BtGFP WT while inducing cell–cell fusion. A 1:1 mixture of A549 cells that expressed the plasma membrane marker TagRFP-T-farnesyl or cytoplasmic GFP was used. Times represent min:s post protrusion formation. Images taken at ∼16 hpi. Scale bars are 5 µm. White arrows highlight the bacterium forming the protrusion. Black arrows highlight the region of protrusion entry. Where GFP signal is difficult to see, insets with increased brightness are shown. (C) Model of cell–cell fusion occurring elsewhere within the protrusion. (D) Still showing visible detachment of the bacterium-containing protrusion from the donor cell.

FIGURE 3:

Quantification of B. thailandensis inducing cell–cell fusion live-cell imaging dataset. (A) Graph of maximum protrusion length from videos where entire protrusion was visible (n = 12). (B) Graph of maximum protrusion length for membrane fusion that occurred at the protrusion tip (n = 9) vs. elsewhere in the protrusion (n = 4). (C) Graph of time to cytoplasmic mixing (n = 20). (D) Graph of time to cytoplasmic mixing vs. maximum protrusion length (n = 12). R² = 0.09197, p = 0.3138. (E) Graph of time to cytoplasmic mixing for membrane fusion that occurred at the protrusion tip (n = 8) vs. elsewhere in the protrusion (n = 12). For A–E, P values were calculated by unpaired Mann–Whitney test, data are mean ± SD.

FIGURE 4:

VgrG5 acts at the membrane fusion step (A) Live-cell imaging stills of BtGFP ΔvgrG5 during cell-to-cell spread. A549 cells that expressed TagRFP-T-farnesyl were used. Times represent min:s post protrusion formation. All images taken at ∼24 hpi. Scale bars are 5 µm. (B) Model of spread. (C) Graph of maximum protrusion length for BtGFP WT (n = 12) and BtGFP ΔvgrG5 (n = 14). (D) Graph of time to cytoplasmic mixing (n = 20) or protrusion engulfment (n = 20). For C and D, P values were calculated by unpaired Mann–Whitney tests; data are mean ± SD.

FIGURE 5:

B. thailandensis must express VgrG5 within a protrusion to induce cell–cell fusion. (A) Experimental design and possible outcomes. (B) Live-cell imaging stills of BtGFP Δvgrg5 spreading from an MNGC initially formed by cell–cell fusion induced by BtBFP WT bacteria. Times represent min:s after the video began. Scale bars are 5 µm.

FIGURE 6:

TagD5 is required for inducing cell–cell fusion and acts at the membrane fusion step. (A) Plaque areas of Vero cells infected with the indicated strains. N = 3 experiments, 9–11 plaques per experiment. (B) Live-cell imaging stills of BtGFP ΔtagD5 during cell-to-cell spread. A549 cells that expressed TagRFP-T-farnesyl were used. Times represent min:s post protrusion formation. All images taken at ∼24 hpi. Scale bars are 5 µm. (C) Model of spread. (D) Graph of maximum protrusion length for BtGFP WT (n = 12) and BtGFP ΔtagD5 (n = 15). (E) Graph of time to cytoplasmic mixing (n = 20) or protrusion engulfment (n = 23). For A, D, and E, P values were calculated by unpaired Mann–Whitney tests; data are mean ± SD.