Limits...
FtsZ does not initiate membrane constriction at the onset of division

View Article: PubMed Central - PubMed

ABSTRACT

The source of constriction required for division of a bacterial cell remains enigmatic. FtsZ is widely believed to be a key player, because in vitro experiments indicate that it can deform liposomes when membrane tethered. However in vivo evidence for such a role has remained elusive as it has been challenging to distinguish the contribution of FtsZ from that of peptidoglycan-ingrowth. To differentiate between these two possibilities we studied the early stages of division in Escherichia coli, when FtsZ is present at the division site but peptidoglycan synthesizing enzymes such as FtsI and FtsN are not. Our approach was to use correlative cryo-fluorescence and cryo-electron microscopy (cryo-CLEM) to monitor the localization of fluorescently labeled FtsZ, FtsI or FtsN correlated with the septal ultra-structural geometry in the same cell. We noted that the presence of FtsZ at the division septum is not sufficient to deform membranes. This observation suggests that, although FtsZ can provide a constrictive force, the force is not substantial at the onset of division. Conversely, the presence of FtsN always correlated with membrane invagination, indicating that allosteric activation of peptidoglycan ingrowth is the trigger for constriction of the cell envelope during cell division in E. coli.

No MeSH data available.


Related in: MedlinePlus

The presence of FtsZ at the division septum is not enough to initiate inner membrane constriction.E. coli cells expressing chromosomally encoded FtsZ-GFP were analysed by cryo-CLEM. (a–d) Upper row, cryo-fluorescence image. Lower row, cryo-electron microscopy image of the same cells as above. (a) In cells where FtsZ-GFP had not yet condensed to a single ring, membrane invagination was not initiated. (b) Cells with FtsZ-GFP accumulated at the midcell, but without visible constrictions. (c,d) Cells with FtsZ-GFP accumulated at the midcell that also showed visible constrictions, indicative of a later stage during division. All cells in this stage had uniform inner and outer membrane invaginations. FtsZ-GFP was also observed in deeply constricted cells (Supplementary Fig. 3), but was not observed in cells that had completed division. Cells expressing FtsZ-GFP exhibited no apparent growth phenotype (Supplementary Figs 4 and 6) and the amount of FtsZ-GFP was less than 20% of the total cellular FtsZ (Supplementary Fig. 5). The total number cells examined by cryo-CLEM during early FtsZ-GFP accumulation at midcell was 127 (the total number of cells for all stages was >200). (e,f) The membrane tethers FtsA and ZipA localize to midcell together with FtsZ, shown by dual color fluorescence microscopy imaging on live cells simultaneously expressing (e) FtsZ-GFP and ZipA-mCherry or (f) FtsZ-mCherry and FtsA-GFP. n > 100. Scale bars = 2 μm. Images are best viewed on a digital screen.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5016829&req=5

f1: The presence of FtsZ at the division septum is not enough to initiate inner membrane constriction.E. coli cells expressing chromosomally encoded FtsZ-GFP were analysed by cryo-CLEM. (a–d) Upper row, cryo-fluorescence image. Lower row, cryo-electron microscopy image of the same cells as above. (a) In cells where FtsZ-GFP had not yet condensed to a single ring, membrane invagination was not initiated. (b) Cells with FtsZ-GFP accumulated at the midcell, but without visible constrictions. (c,d) Cells with FtsZ-GFP accumulated at the midcell that also showed visible constrictions, indicative of a later stage during division. All cells in this stage had uniform inner and outer membrane invaginations. FtsZ-GFP was also observed in deeply constricted cells (Supplementary Fig. 3), but was not observed in cells that had completed division. Cells expressing FtsZ-GFP exhibited no apparent growth phenotype (Supplementary Figs 4 and 6) and the amount of FtsZ-GFP was less than 20% of the total cellular FtsZ (Supplementary Fig. 5). The total number cells examined by cryo-CLEM during early FtsZ-GFP accumulation at midcell was 127 (the total number of cells for all stages was >200). (e,f) The membrane tethers FtsA and ZipA localize to midcell together with FtsZ, shown by dual color fluorescence microscopy imaging on live cells simultaneously expressing (e) FtsZ-GFP and ZipA-mCherry or (f) FtsZ-mCherry and FtsA-GFP. n > 100. Scale bars = 2 μm. Images are best viewed on a digital screen.

Mentions: Our initial goal was to determine if FtsZ was sufficient to generate a contractile force in vivo. Our working hypothesis was that, if FtsZ was sufficient we should see deformations of the inner membrane when only FtsZ and its membrane tethers FtsA23 and ZipA24 (aka the proto-ring) are present at midcell (but PG synthesizing enzymes are not). We cryogenically preserved Escherichia coli cells expressing a chromosomal copy of FtsZ-GFP by vitrification and imaged them by cryo-CLEM (Supplementary Fig. 2). When examining cells in a pre-divisional stage, indicated by the typical helical arrangement of FtsZ-GFP2526, we never observed membrane invagination (Fig. 1a). In cells where FtsZ-GFP had condensed to a single band (as judged from the cryo-fluorescence images) we noted that 27% lacked a visible invagination (as judged from the cryo-EM images) (Fig. 1b). The cells that had visible membrane invaginations, could be further classified as having either minor (~17%) (Fig. 1c) or major invaginations (~56%) (Fig. 1d). In all cells where membrane invagination was observed we noted that both the inner and outer membranes were equally deformed (see also Supplementary Fig. 3). These observations suggest that the presence of FtsZ-GFP at the division septum is not sufficient to deform the inner membrane in vivo.


FtsZ does not initiate membrane constriction at the onset of division
The presence of FtsZ at the division septum is not enough to initiate inner membrane constriction.E. coli cells expressing chromosomally encoded FtsZ-GFP were analysed by cryo-CLEM. (a–d) Upper row, cryo-fluorescence image. Lower row, cryo-electron microscopy image of the same cells as above. (a) In cells where FtsZ-GFP had not yet condensed to a single ring, membrane invagination was not initiated. (b) Cells with FtsZ-GFP accumulated at the midcell, but without visible constrictions. (c,d) Cells with FtsZ-GFP accumulated at the midcell that also showed visible constrictions, indicative of a later stage during division. All cells in this stage had uniform inner and outer membrane invaginations. FtsZ-GFP was also observed in deeply constricted cells (Supplementary Fig. 3), but was not observed in cells that had completed division. Cells expressing FtsZ-GFP exhibited no apparent growth phenotype (Supplementary Figs 4 and 6) and the amount of FtsZ-GFP was less than 20% of the total cellular FtsZ (Supplementary Fig. 5). The total number cells examined by cryo-CLEM during early FtsZ-GFP accumulation at midcell was 127 (the total number of cells for all stages was >200). (e,f) The membrane tethers FtsA and ZipA localize to midcell together with FtsZ, shown by dual color fluorescence microscopy imaging on live cells simultaneously expressing (e) FtsZ-GFP and ZipA-mCherry or (f) FtsZ-mCherry and FtsA-GFP. n > 100. Scale bars = 2 μm. Images are best viewed on a digital screen.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5016829&req=5

f1: The presence of FtsZ at the division septum is not enough to initiate inner membrane constriction.E. coli cells expressing chromosomally encoded FtsZ-GFP were analysed by cryo-CLEM. (a–d) Upper row, cryo-fluorescence image. Lower row, cryo-electron microscopy image of the same cells as above. (a) In cells where FtsZ-GFP had not yet condensed to a single ring, membrane invagination was not initiated. (b) Cells with FtsZ-GFP accumulated at the midcell, but without visible constrictions. (c,d) Cells with FtsZ-GFP accumulated at the midcell that also showed visible constrictions, indicative of a later stage during division. All cells in this stage had uniform inner and outer membrane invaginations. FtsZ-GFP was also observed in deeply constricted cells (Supplementary Fig. 3), but was not observed in cells that had completed division. Cells expressing FtsZ-GFP exhibited no apparent growth phenotype (Supplementary Figs 4 and 6) and the amount of FtsZ-GFP was less than 20% of the total cellular FtsZ (Supplementary Fig. 5). The total number cells examined by cryo-CLEM during early FtsZ-GFP accumulation at midcell was 127 (the total number of cells for all stages was >200). (e,f) The membrane tethers FtsA and ZipA localize to midcell together with FtsZ, shown by dual color fluorescence microscopy imaging on live cells simultaneously expressing (e) FtsZ-GFP and ZipA-mCherry or (f) FtsZ-mCherry and FtsA-GFP. n > 100. Scale bars = 2 μm. Images are best viewed on a digital screen.
Mentions: Our initial goal was to determine if FtsZ was sufficient to generate a contractile force in vivo. Our working hypothesis was that, if FtsZ was sufficient we should see deformations of the inner membrane when only FtsZ and its membrane tethers FtsA23 and ZipA24 (aka the proto-ring) are present at midcell (but PG synthesizing enzymes are not). We cryogenically preserved Escherichia coli cells expressing a chromosomal copy of FtsZ-GFP by vitrification and imaged them by cryo-CLEM (Supplementary Fig. 2). When examining cells in a pre-divisional stage, indicated by the typical helical arrangement of FtsZ-GFP2526, we never observed membrane invagination (Fig. 1a). In cells where FtsZ-GFP had condensed to a single band (as judged from the cryo-fluorescence images) we noted that 27% lacked a visible invagination (as judged from the cryo-EM images) (Fig. 1b). The cells that had visible membrane invaginations, could be further classified as having either minor (~17%) (Fig. 1c) or major invaginations (~56%) (Fig. 1d). In all cells where membrane invagination was observed we noted that both the inner and outer membranes were equally deformed (see also Supplementary Fig. 3). These observations suggest that the presence of FtsZ-GFP at the division septum is not sufficient to deform the inner membrane in vivo.

View Article: PubMed Central - PubMed

ABSTRACT

The source of constriction required for division of a bacterial cell remains enigmatic. FtsZ is widely believed to be a key player, because in vitro experiments indicate that it can deform liposomes when membrane tethered. However in vivo evidence for such a role has remained elusive as it has been challenging to distinguish the contribution of FtsZ from that of peptidoglycan-ingrowth. To differentiate between these two possibilities we studied the early stages of division in Escherichia coli, when FtsZ is present at the division site but peptidoglycan synthesizing enzymes such as FtsI and FtsN are not. Our approach was to use correlative cryo-fluorescence and cryo-electron microscopy (cryo-CLEM) to monitor the localization of fluorescently labeled FtsZ, FtsI or FtsN correlated with the septal ultra-structural geometry in the same cell. We noted that the presence of FtsZ at the division septum is not sufficient to deform membranes. This observation suggests that, although FtsZ can provide a constrictive force, the force is not substantial at the onset of division. Conversely, the presence of FtsN always correlated with membrane invagination, indicating that allosteric activation of peptidoglycan ingrowth is the trigger for constriction of the cell envelope during cell division in E. coli.

No MeSH data available.


Related in: MedlinePlus