A Bacterium Learns Long Division

by Nanne Nanninga

The common picture of a dividing rod-shaped bacterium encompasses the positioning of the divisome, including an FtsZ-ring, in the cell center. This occurs after the cell has doubled its length without increasing its diameter. Conversely, increase in diameter without cell elongation would seem highly unlikely in a rod-shaped organism. Yet, this happens.

Figure 1: SEM of tightly apposed ectosymbionts on the surface of L. oneistus. Electron micrograph by N. Leisch.

In fact, this is the normal condition for an ectosymbiotic bacterium that lines the surface of the marine nematode Laxus oneistus. The symbiont is a g-proteobacterium like E. coli, but unlike E. coli it has not been cultured outside its natural habitat. As originally described by Polz et al. in the early nineties, the bacteria are positioned perpendicular to the surface of L. oneistus. They are glued to the surface of the nematode by a C-type lectin. The original paper already indicated that division takes place longitudinally, with one of the daughters presumed to remain attached to the nematode. This makes sense because otherwise the daughter cells would get lost to the environment. Recall that this phenomenon lies at the basis of the Helmstetter-Cooper baby machine, whereby one of a pair of newborn cells is selected to start a synchronous culture (Helmstetter et al.)

Recently, Leisch and colleagues have extended these observations by carefully determining cellular dimensions and visualizing the FtsZ division protein with fluorescent E. coli monoclonal antibodies. The results can be compared with E. coli data (Figure 2A, B). Whereas E. coli elongates as expected, this is not what happens with the symbiont. The symbiont does not change its length (Figure 2A) but increases its diameter (Figure 2B). Consequently, the symbiont divides longitudinally, (Fig. 2 C-H). Immunostaining of FtsZ reveals that FtsZ positioning correlates with the cell constriction. In fact, an ellipsoidal FtsZ-ring is observed stretched along the length of the endosymbiont.

Figure 2: A. Cell length compared to cell volume in L. oneistus symbiont en E. coli. B. Cell width in L. oneistus en E. coli compared to cell volume. C-D. Cells arranged according to division stage and imaged by differential contrast images (upper row) and confocal fluorescent microscopy (lower row). FtsZ is visible as green spots in the lower panel. Source.

These remarkable observations provide more questions than answers. For instance, how does the diameter of the symbiont increase? Is the increase the same in all directions or is it polarized? Though, the authors do not discuss this point, it would seem that the bacterium has the shape of shoebox. Thus, increase in width implies widening of the shoebox without altering its height. How is the shoebox-shape, if applicable, maintained? Furthermore, how is DNA segregation carried out in the absence of cell elongation? How are the nucleoids arranged? It would seem that widening of the shoebox is sufficient. Again the spatial mechanism is not known. As pointed out by the authors, genes encoding Min, C, D and E are also present. Do they function in the symbiont, and if so how do they conform to different geometry? Answers to these questions are likely to elucidate the degrees of freedom in bacterial cell division. But that’s what we learned in our arithmetic class about the long division. Unless one deals with a whole number, there is always a remainder.

Prof. Nanninga is Emeritus Professor of Molecular Cytology at the University of Amsterdam Swammerdam Institute for Life Sciences

Comments

[Anthony Vecchiarelli]

This is fascinating. Current Min models propose that the system can "find" the long-axis of the cell, and use it as a spatial cue for mid-cell positioning of the divisome. Cell-free reconstitution suggests "geometry sensing" is inherent in the biochemistry, while other models show cardiolipin and curvature at the cell poles as "ultimate" cues for long-axis positioning. Short axis cell-division (and nucleoid segregation?!) has significant implications in understanding how stuff is spatially organized in different bacteria. There are so many questions to ask and I can't wait to see how the story unfolds.

[Rebin Balachandran]

a good knowledge for all microbiologist

[Paul Orwin]

Funny,
I just used this example (the laxus symbiont) in a seminar about nematode bacterial symbiosis. Wonderful to see that the basic biology is being probed :)

[Kelly]

Speaking of FtsZ, this study from the Errington Lab (http://www.cell.com/cell-reports/abstract/S2211-1247(12)00093-9?switch=standard) shows B. subtilis mutants without FtsZ generate L-forms, but not without BCFA genes. Is this antibiotic resistance when targeting cytoskeletal proteins yet illness continues?
"L-forms provide a simple biological model that might be representative of primitive cell proliferating systems...independent of known cytoskeletal proteins"