![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| GUEST COMMENTARY |
Centro Nacional de Biotecnología, Consejo Superior de Investigaciones Científicas Campus de Cantoblanco, C/ Darwin 3, 28049 Madrid, Spain
Like the cells of other bacteria that are not rod shaped, Streptococcus pneumoniae cells manage to avoid potentially distressing changes in their surface-to-volume ratio as they grow. The work reported by Fadda et al. in this issue (9) on the streptococcal DivIVA protein suggests new ideas on how this protein may contribute to transformation of the midcell into two pointed poles that result in duplication of the cell volume with no net increase in cell diameter, a mechanism that prevents changes in the surface-to-volume ratio during growth and division.
Some nearly spherical bacteria, like Staphylococcus aureus, have circumvented the issue of surface-to-volume ratio maintenance by placing a sagittal septum at midcell while maintaining a nearly constant diameter. Once finished, the S. aureus septum splits in the middle, and two polar, hemispherical caps are gradually reshaped from the resultant almost flat moieties (33). In contrast, the ovoid S. pneumoniae cells grow and divide simultaneously by recreating two pointed polar caps at midcell. Although a septum is formed, it does not have continuity in the sagittal plane until the very end of division. In some respects the S. pneumoniae septation process resembles constriction, like that observed in Escherichia coli dividing cells, except that the resultant polar surface contributes a larger portion of the cell surface in the case of the streptococcal cell than in the case of the rod-shaped E. coli cell.
divIVA COLOCALIZES IN THE STREPTOCOCCAL CHROMOSOME WITH OTHER CELL DIVISION GENES
The divIVA gene, which for the most part is confined to gram-positive bacteria, was first identified in Bacillus subtilis. Cells with a mutation in this gene have a reduced septation frequency and undergo aberrant polar division, leading to the formation of anucleate minicells (17, 27). In S. pneumoniae a gene coding for a protein with a high level of homology to B. subtilis DivIVA is located in a region downstream from the cell division ftsZ gene (8, 16), forming part of one of the three portions dispersed from the dcw cluster that in rod-shaped bacteria contains most of the tightly grouped cell division and cell wall synthesis genes (18, 31).
The divIVA gene codes for a protein that has been implicated in selection of septum positioning at midcell in vegetative division of B. subtilis, where it has been proposed to play a role similar to that of the E. coli MinE topological specificity component of the MinCDE division site selection system (6, 34). In B. subtilis the DivIVA protein has an additional role during sporulation; together with Spo0J, RacA, and Soj, it is required for relocation of chromosomes and their attachment to the cell poles (2, 23, 32). In gram-positive bacteria, besides S. pneumoniae, the divIVA gene has been studied in Enterococcus faecalis (25) and Staphylococcus aureus (24). A divIVA gene is also present in Streptomyces coelicolor (10) and in other actinomycetes, like Mycobacterium tuberculosis, where Wag31 (antigen 84), a protein proposed to be involved in cell shape maintenance (10, 13), exhibits some sequence similarity with DivIVA from Brevibacterium lactofermentum (26). While many gram-positive bacteria may contain divIVA but lack minE and even the full minCDE system, many gram-negative bacteria have minE but no divIVA. However, there are some exceptions to this. Myxococcus xanthus, a gliding gram-negative bacterium in which peptidoglycan is discontinuously distributed along the wall, contains a gene, fruD, with homology to divIVA (1), but no minE gne has been identified. In some of the clostridia (Clostridium perfringens, Clostridium beijerinckii, Clostridium phytofermentans, and Clostridium thermocellum), which are gram-positive bacteria, divIVA and minE are both present (30). Synechococcus elongatus, a cyanobacterium in which the envelope contains multiple complex layers (28), contains both divIVA and minE (19). In their paper, Fadda et al. propose that in S. pneumoniae, a species without the MinCDE system, DivIVA plays a role in the cell division machinery (9).
PEPTIDOGLYCAN SYNTHESIS AND DivIVA LOCALIZATION OCCUR AT COINCIDENT SITES IN STREPTOCOCCI AND OTHER BACTERIA
Using immunofluorescence and immunogold labeling, Fadda et al. (9) found that streptococcal DivIVA localizes at sites coincident with sites at which peptidoglycan synthesis takes place (Fig. 1). Similar positions are occupied in the hyphal tip by the S. coelicolor homolog, where it is thought to promote hyphal growth and morphogenesis (10), and in the ends of B. lactofermentum cells, where it promotes apical growth (26). B. subtilis DivIVA, when present in E. coli as a heterologous protein fused to green fluorescent protein, also localizes at midcell (depending on the presence of a Z ring) and at the poles; even more remarkably, in Schizosaccharomyces pombe the DivIVA-green fluorescent protein fusion associates with the division septum and the growth zones (7). Interestingly, the B. subtilis DivIVA protein localizes in the fission yeast in the same manner as the endogenous Cdc8p tropomyosin-like protein, suggesting that a possible targeting signal for DivIVA localization is conserved in prokaryotic and eukaryotic organisms.
|
STREPTOCOCCAL DivIVA INTERACTION NETWORK
Although mostly shown by indirect procedures (4, 14), there seem to be numerous interactions between the E. coli and B. subtilis components of the division and the cell wall synthesis machinery (for reviews, see references 11 and 36). If, as suggested by the evidence discussed above, DivIVA forms part of this machinery in S. pneumoniae, it should be possible to detect a number of interactions with its other components.
Fadda et al. (9) used the E. coli two-hybrid assay developed by Di Lallo et al. (4, 5) to probe DivIVA interactions with other components of the streptococcal divisome. They found that streptococcal DivIVA interacts with itself, which agrees with the interactions of B. subtilis and Enterococcus faecalis DivIVA in vivo and in vitro, respectively (20, 25). As expected, streptococcal DivIVA also interacts with the streptococcal homolog of the chromosome segregation protein Spo0J. In addition, Fadda et al. found that DivIVA interacts with most of the molecules involved in the early and late stages of the S. pneumoniae septation process, including FtsZ, FtsA, FtsK, DivIVB (FtsQ in E. coli), DivIC (FtsB in E. coli), FtsL, and FtsW. It is even able to interact with homologs of the ErzA and ZapA proteins that regulate Z-ring stability in other bacteria. Moreover, interactions were also found with proteins involved in cell wall rearrangement, like LytB, a protein with polar localization involved in daughter cell separation (3), and PcsB, a putative murein hydrolase reported to be a morphogenetic determinant (21). On the other hand, no interaction between DivIVA and two septal enzymes involved in murein biosynthesis, PBP2X and PBP1A, or between DivIVA and PrfA, an orthologue of a B. subtilis DNA-binding protein implicated in cell wall synthesis, chromosome segregation, and DNA recombination and repair (22), was detected. Perry and Edwards (23) performed coimmunoprecipitation assays demonstrating the association of B. subtilis DivIVA with FtsZ and MinD during vegetative growth and with Spo0J during sporulation; Fadda et al. (9) similarly found that streptococcal DivIVA coprecipitates with FtsZ and Spo0J but, as expected, not with PrfA.
DivIVA ROLE: POINTING AT THE SEPTUM AND SHARPENING THE POLE
Although the absence of DivIVA in S. pneumoniae does not result in complete cessation of cell division, it does decrease the proportion of productive FtsZ rings able to terminate septation, although not their total number, suggesting that DivIVA may contribute to Z-ring progression. This lack of closing of Z rings gives rise to 15% anucleate dead cells. As reported by Fadda et al. (9), introduction of an A78T mutation into streptococcal DivIVA in a region shown in B. subtilis to be related to cell division defects resulted in a partially active protein that was not fully efficient in localizing at the septal and polar positions occupied by the wild-type protein. The A78T protein was able to interact with itself and with the Spo0J and PcsB proteins involved in segregation, and therefore, the mutant showed no salient segregation defects; however, there was no interaction with LytB, and so the mutant was not able to effect cell separation, resulting in a chainy phenotype.
Curiously, neither the morphology nor other identifiable cell growth and division processes of divIVA mutants of the spherical organism S. aureus are affected (24), while the S. pneumoniae divIVA null mutant described by Fadda et al. (9) is hemispherical rather than ovoid. It is possible that DivIVA is attracted to regions in which the cell envelope bears a stress associated with growth of a thick or complex peptidoglycan layer during the formation of either a lateral (elongation) or transverse (septation) wall. Once there, DivIVA may recruit a variety of other proteins involved in septation and even in DNA partition.
Is it possible that the DivIVA protein is used by gram-positive bacteria primarily as part of a cytoskeletal stress-bearing mechanism related to cell wall growth? Is it possible that other proteins took advantage of the DivIVA localization to add a topological determinant to their own function by simply establishing an interaction with DivIVA? Considering the large number of interactions with other proteins found for DivIVA, this is a possibility. If all this were correct, the acquisition of a DivIVA interacting domain could have been used frequently by proteins to acquire topological specificity. Such acquisition should have left an evolutionary marker in their sequences.
We do not know if the ancestral gram-positive bacteria (15) contained both MinE and DivIVA as topological determinants as some present-day clostridia do. In bacteria that evolved an outer membrane and consequently could streamline their peptidoglycan layer (e.g., gram-negative bacteria), the stress-bearing DivIVA protein may have become dispensable and eventually lost together with its topological determination for some activities. In the case of MinC and MinD, being attracted to the poles is a requirement, as they block FtsZ polymerization; therefore, their presence prevents abnormal polar septation. Eventual loss of MinE could be tolerated by gram-positive bacteria if MinC and MinD could be directed to the poles by an alternative interaction with DivIVA. On the other hand, gram-negative bacteria, if they were deprived of DivIVA, would be forced to retain or otherwise acquire MinE as their sole topological specifier of the septum-positioning mechanism.
 |
FOOTNOTES |
|---|
FOOTNOTES
Published ahead of print on 17 November 2006.
The views expressed in this Commentary do not necessarily reflect the views of the journal or of ASM.
REFERENCES
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Appl. Environ. Microbiol. | Infect. Immun. | Eukaryot. Cell |
|---|---|---|
| Mol. Cell. Biol. | J. Virol. | Microbiol. Mol. Biol. Rev. |
| ALL ASM JOURNALS |
