Bistable forespore engulfment in Bacillus subtilis by a zipper mechanism in absence of the cell wall - PubMed (original) (raw)
Bistable forespore engulfment in Bacillus subtilis by a zipper mechanism in absence of the cell wall
Nikola Ojkic et al. PLoS Comput Biol. 2014.
Abstract
To survive starvation, the bacterium Bacillus subtilis forms durable spores. The initial step of sporulation is asymmetric cell division, leading to a large mother-cell and a small forespore compartment. After division is completed and the dividing septum is thinned, the mother cell engulfs the forespore in a slow process based on cell-wall degradation and synthesis. However, recently a new cell-wall independent mechanism was shown to significantly contribute, which can even lead to fast engulfment in [Formula: see text] 60 [Formula: see text] of the cases when the cell wall is completely removed. In this backup mechanism, strong ligand-receptor binding between mother-cell protein SpoIIIAH and forespore-protein SpoIIQ leads to zipper-like engulfment, but quantitative understanding is missing. In our work, we combined fluorescence image analysis and stochastic Langevin simulations of the fluctuating membrane to investigate the origin of fast bistable engulfment in absence of the cell wall. Our cell morphologies compare favorably with experimental time-lapse microscopy, with engulfment sensitive to the number of SpoIIQ-SpoIIIAH bonds in a threshold-like manner. By systematic exploration of model parameters, we predict regions of osmotic pressure and membrane-surface tension that produce successful engulfment. Indeed, decreasing the medium osmolarity in experiments prevents engulfment in line with our predictions. Forespore engulfment may thus not only be an ideal model system to study decision-making in single cells, but its biophysical principles are likely applicable to engulfment in other cell types, e.g. during phagocytosis in eukaryotes.
Conflict of interest statement
The authors have declared that no competing interests exist.
Figures
Figure 1. Bistable forespore engulfment of B. subtilis after cell-wall removal.
(A and B) Images adopted from . Medial focal plane images of sporulating bacteria treated with cell-wall removal lysozyme in osmotically protected medium with 0.5 M of sucrose . Fluorescent membrane stain FM 4–64 was used to track the progressing mother-cell membrane engulfing the forespore. Arrows point to the moving edges of the mother membrane. Double-headed arrows show mother-forespore cell separation. (A) In wild-type (WT) cells upon cell-wall removal, mother cell either engulfs the forespore (top) or retracts (bottom), see Movie S1. Time 0 minutes (0') is assigned to the onset of volume loss (see Fig. 2A). (B) Absence of the zipper proteins SpoIIQ (top) or SpoIIIAH (bottom) prevents membrane from forward progression causing protoplast separation. Time 0 minutes is used as the time of physical separation of mother cell and forespore. (C) Cartoon of fast bistable forespore engulfment in WT cells. Mother-cell compartment and forespore are shown in blue and red, respectively. After cell-wall removal 
60 
of the sporulating cells engulf the forespore, while 
40 
fail to engulf . (D) Cartoon showing the protoplast separation as observed in mutants of panel B. Scale bars: 2 
m.
Figure 2. Image analysis reveals mother-cell volume loss during engulfment.
(A–D) Using active contours we measured volume, surface area, and engulfment over time for mother cells and forespores (for details see Materials and Methods). The onset of volume loss for each cell was set to 0 minutes and all cell measurements were aligned in time based on 0' points. 3D volume and surface area were calculated assuming rotational symmetry around the axis that connects center of masses of forespore and mother cells. All analysis was performed on previously published movies from . (A) Mother-cell volume loss amounts to 
35 
during engulfment, while forespore volume remains the same. (Inset) Typical cell shrinks longitudinally causing volume loss. (B) Surface area for mother cell was calculated using two models, termed “tight cup” or “broad cup” shown in Fig. 3A. The broad-cup model is assumed for times 
0' as no morphological membrane changes occur for these times (see Fig. 3B). The mother-surface area reduction is 
5–15 
, while forespore-surface area remains the same. (C) Engulfment shown by FM 4–64 kymograph (pixel intensities along the forespore contour versus time) of a single representative cell treated with lysozyme coincides with onset of volume loss. (D) Average engulfment over time. (Inset) Cross-correlation coefficient versus time showing anticorrelation between mother-cell volume and engulfment, and mother-surface area and engulfment. No time delays were observed. Average +/- SEM of 
= 6 cells.
Figure 3. Cup-shape analysis.
(A) Theoretical predictions of two exclusive cup-shape models in two different fluorescence channels. Fluorescent FM 4–64 labels all membrane exposed to medium, while SpoIIIJ-GFP localizes at the mother-cell membrane only . Therefore, the “tight-cup” model predicts four membrane folds at the mother-forespore boundary in the FM 4–64 channel and three membrane folds observed in the GFP channel. Likewise, “broad-cup” model predicts two membrane folds in the FM 4–64 channel and a single membrane in the GFP channel. (B–C) For a single cell at certain time points two average intensities were measured: average pixel intensity of mother-forespore boundary (
), and average pixel intensity of mother membrane far from the boundary (
). Ratio 
versus time is plotted for FM 4–64 channel (B) and for SpoIIIJ-GFP (C). As before, time 0' is the onset of volume loss (see Fig. 2). Average +/− SEM for different cells is plotted. See Movie S2.
Figure 4. Engulfment model and Langevin simulations.
(A) 3D mother-cell membrane is represented by a string of beads assuming rotational symmetry around 
-axis. Each bead at position (
, 
) represents a ribbon of width 
and length 
. Forespore is modeled as a solid sphere. See Materials and Methods, and Text S1 for further model explanations. (B) Snapshot of 3D simulation showing example of early-stage engulfment.
Figure 5. Simulation results for engulfment and volume/surface area changes.
(A–D) Simulation snapshots for different parameter combinations for surface tension (
) and 
(pressure difference) at 5 s for fixed SpoIIQ-SpoIIIAH surface density 
. Simulations that reached full engulfment earlier than 5 s were terminated and last snapshots are displayed. (B) Percentage of forespore-surface area enclosed by mother membrane. White dashed line separates regions of full and partial engulfment. White cross shows the parameters used for Fig. 6A and B. (C and D) Volume and surface area of mother cell.
Figure 6. Bistable engulfment depends on zipper-molecule density.
(A) Engulfment as a function of SpoIIQ protein surface density for 
= 50 pN/_µ_m and 
= 500 Pa. The total binding energy was converted to SpoIIQ protein-surface density using the binding energy of a single SpoIIQ-SpoIIIAH bond (see Text S1) . Consistent with experimental results shown in (D) (extracted from [8]), engulfment is threshold-dependent on number of SpoIIQ proteins expressed in forespores. Gray vertical arrows point to surface densities for which snapshots are shown in (C) at 5 s. (B) Simulations lead to bistable outcome at later times (
= 10 s) with two subpopulations of stalled and fully completed cups. (E) For each set of constraint parameters (
and 
) we performed a surface-density scan as in (A). The lower bound on the critical number of SpoIIQ molecules necessary for engulfment ranges from 
120 to 
7200 molecules depending on constraint parameters.
Figure 7. Decreasing medium osmolarity causes retraction as theoretically predicted.
(A) Medial focal plane images of sporulating cells treated with lysozyme in osmotically protected medium but without sucrose (see Materials and Methods, Movie S4). Fully engulfed cells retract (cells 1 and 3) while partially engulfed cells (cells 2 and 4) fail to engulf and undergo retraction as well. (B) Percentage of forespore engulfment after lysozyme treatment in medium with 0.5 M sucrose and without sucrose. Average +/− STD of four different microscopy fields, each containing 136–160 cells in sucrose condition and 93–173 cells without sucrose. (C) Images of typical cell undergoing swelling and retraction. Time 0 minutes corresponds to fully rounded mother cell. (D) Mother-cell and forespore volume aligned in time based on 0 time point defined as in (C). After mother cells round, volume steadily increases. (Inset) Volume comparison for experiments with and without sucrose (
). 
is mother-cell volume 
2 minutes after engulfment or retraction corresponding to 0.5 M surose or no sucrose, respectively. 
is initial mother-cell volume 
8 minutes before transition. (E) Mother-cell and forespore surface area. (Inset) Surface area comparison for cases with and without sucrose (
). Average +/− SEM of 
= 12 cells in panel (D, E). Scale bars 2 _µ_m.
References
- Stragier P, Losick R (1996) Molecular genetics of sporulation in Bacillus subtilis . Annu Rev Genet 30: 297–341. -PubMed
- Underhill DM, Ozinsky A (2002) Phagocytosis of microbes: complexity in action. Annu Rev Immunol 20: 825–852. -PubMed
Publication types
MeSH terms
LinkOut - more resources
Full Text Sources
Other Literature Sources
Molecular Biology Databases