1 Dipartimento di Biologia, Università "Tor Vergata", Via della Ricerca Scientifica, I-00133 Rome, Italy
2 Istituto di Biologia e Patologia Molecolari del CNR, Rome, Italy
Correspondence
L. Paolozzi
Paolozzi{at}bio.uniroma2.it
![]() |
ABSTRACT |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
INTRODUCTION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
An example of complex machinery in prokaryotes is the septosome, a complex structure formed by many different proteins responsible for cytokinesis. In Escherichia coli, and in other rod-shaped bacteria, the septosome differentiates in the cell centre at a particular time of the bacterial cycle, resulting in the formation of two daughter cells of the same size.
FtsZ is a key protein in the targeting and initiation of cytokinesis. Conditional mutants of ftsZ in E. coli fail to divide, yielding long filamentous cells that replicate and segregate their chromosomes but show no sign of division septa or cellular constrictions (hence the term fts, standing for filamentation temperature sensitive). Formation of long FtsZ polymers, starting from single subunits extending into the cytoplasm, is the first step towards septosome differentiation. This polymer forms a ring bound to the inner membrane (Bi & Lutkenhaus, 1991) and represents a sort of scaffold on which the other components of the division machinery are assembled. At least nine more proteins are currently known to participate in septum formation: FtsA, FtsK, FtsW, FtsQ, FtsI, FtsL, FtsN and ZipA (for reviews, see Rothfield et al., 1999
; Margolin, 2000
; Donachie, 2001
) and YgbQ (or FtsB) (Buddelmeijer, 2002
).
Cytological experiments have established that all these proteins co-localize at the septum level. In addition, it has been shown that ts mutations in the ftsZ gene result in the loss of co-localization of the eight remaining proteins, indicating that FtsZ is the first protein involved in the cell division process. These experiments demonstrated that FtsA localization to the Z-ring depends on FtsZ, whereas Z-ring localization does not depend on FtsA (Ma et al., 1996, 1997
; Addinal et al., 1996
; Addinal & Lutkenhaus, 1996
). Evidence of a direct interaction between FtsZ and FtsA comes from yeast two-hybrid experiments (Wang et al., 1997
) and from co-localization experiments with GFP-tagged FtsA (Ma et al., 1996
, 1997
; Addinal & Lutkenhaus, 1996
).
Various authors have proposed models for the assembly of the septosome, based on their migration sequence by co-localization in the septum of the division proteins fused with GFP. The latter proteins are recruited in the septum soon after Z-ring formation in a linear sequence (for a review, see Rothfield et al., 1999) with two possible scenarios: (a) FtsA, ZipA, FtsW, FtsK, FtsQ, FtsL, FtsI, FtsN (Margolin, 2000
); and (b) FtsA, ZipA, FtsK, FtsQ, FtsL, FtsW, FtsI, FtsN (Chen & Beckwith, 2001
). These scenarios were presented without taking into account the molecular interactions between the various components of the septosome. It is not yet known whether the sequence of protein recruitment in the septosome is also representative of the order in which they assemble with each other.
The two-hybrid system is an effective and quick tool for the in vivo study of proteinprotein interaction which has been successfully employed in the study of many proteins both in prokaryotes and eukaryotes (for a review, see Hu et al., 2000). In this paper, we present an application of this method to the interaction between the nine division proteins and to 43 possible combinations of interaction between pairs of proteins. In addition, we have developed a model for their presumed sequence of interaction.
![]() |
METHODS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
-Galactosidase assay and two-hybrid assay.
-Galactosidase activity was assayed as described by Miller (1972)
. The two-hybrid assay was performed on bacterial cultures grown at 34 °C in B medium supplemented with 1x10-4 M IPTG to OD600 0·5, as described previously (Di Lallo et al., 2001
).
![]() |
RESULTS |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Pairs of these plasmids (one of which harboured the 434 hybrid repressor, and the other the P22 hybrid repressor) were transformed into the E. coli strain R721, carrying the 434P22 chimeric operator. We obtained 45 bacterial strains, with which 43 possible combinations of division protein couples were studied.
Interactions of 43 combination pairs among the nine division proteins
These combinations refer to only one of the two genetic configurations; i.e. where two different proteins (X and Y) were examined, the data reported in the table refer to cI434XcIP22Y only. However, we found that in most cases, the interactions were observed in both genetic configurations: the cI434XcIP22Y pair reduced -galactose as well as the cI434YcIP22X pair (data not shown).
Before analysing the results of this study, we need to evaluate the sensitivity of the assay. The assay sensitivity has already been analysed in previous work (Di Lallo et al., 2001). We have observed, for example, a residual
-galactosidase synthesis of 5 % (standard deviation±0·25) for the pairs of reconstructed repressors (434-
and P22-
). This value represents the lower limit of the assay. In addition, the study of various prokaryotic and eukaryotic proteins which are positive with the Saccharomyces cerevisiae two-hybrid assay, or have been shown to interact by other methods, showed that the residual
-galactosidase synthesis varies from 13 % to 43 %, depending on the pair of proteins analysed (data not shown). This range could be explained either by the interaction strength, which differs from one pair of proteins to another, or by the different experimental conditions necessary to optimize the interactions, e.g. the protein concentration may be different. Theoretically, in the case of non-interaction, the
-galactosidase synthesis should be similar to that obtained in the parental strain without plasmids. However, we cannot exclude that an interaction, although transient, between the N-terminal domain of the fusion proteins and the operator site could interfere with the assay by slightly repressing the
-galactosidase synthesis.
What is the range of the residual -galactosidase synthesis that determines whether or not two proteins interact with each other? On the basis of a series of results obtained with this assay, we will assume that the residual level of
-galactosidase synthesis between 13 % and 50 % represents positive interaction whereas values beyond this range are considered negative interactions. Although this is an empirical determination, we submitted the data reported in Table 3
to the statistical program SPSS for cluster analysis. The program showed two clearly distinct clusters: cluster I from 1317 % (FtsZFtsZ, FtsZFtsA, ZipA29328FtsZ, true positives in accordance with results of other authors), to 45 % (ZipAZipA interaction); and cluster II, from 80 % (ZipAFtsA) to 90100 % (FtsKFtsA, FtsKFtsN, etc.). Cluster I can be subdivided into three groups, according to the interaction strength hypothesis, whereas cluster II is homogeneous.
|
![]() |
DISCUSSION |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
FtsZ and FtsA are able to interact both with themselves and with each other, according to the observations reported in the literature on the S. cerevisiae two-hybrid system (Wang et al., 1997). Membrane protein ZipA also belongs to this group since is it able to form heterodimers with FtsZ (ZipAFtsZ) as also reported by Haney et al. (2001)
. Furthermore, the two-hybrid assay reveals a ZipA homodimerization not described by other authors. Interestingly, these ZipA interactions are observed despite the fact that ZipA is a membrane-anchored protein. ZipA is, in fact, formed by five domains: amino acids (aa) 16 are a charged periplasmic domain; aa 728 are a transmembrane segment; aa 86185 are called the P/Q domain; and aa 186328 are a globular domain. The globular domain of ZipA has been shown to bind the conserved C-terminal peptide of FtsZ (Hale & de Boer, 1997
; Hale et al., 2000
; Haney et al., 2001
; Liu et al., 1999
; Moy et al., 2000
; Ohashi et al., 2002
). The interactions described in Table 3
can be explained by the hypothesis that a portion of the fusion may not be correctly inserted in the membrane as well as by the fact that the chimeric protein cIZipA could show affinity for both the cytoplasmic membrane and the operator site. The observed interactions could therefore be the resultant of these two effects. According to this interpretation, we observed that the residual
-galactosidase activity obtained with a pair of ZipA proteins lacking the periplasmic domain and transmembrane segment (aa 128) becomes 15 % (±6·1 %) instead of 46 % (±5·4 %) obtained with the whole protein.
In the second group we find three division proteins. FtsI, FtsN and the polytopic membrane protein FtsW are able to interact among themselves; FtsI and FtsN are also able to homodimerize; in addition, FtsW also interacts with FtsL. The third group is made up by FtsL and FtsQ proteins, which could have a binding function between the two previous groups. As a matter of fact, FtsL heterodimerizes with FtsK (belonging to the first group), FtsW (belonging to the second group) and FtsQ (the other connection protein). The FtsQFtsN interaction is supported by genetic data consisting in the fact that the overexpression of FtsN partially suppresses FtsQ1(ts) (Dai et al., 1993). The interaction between FtsQ and FtsK, and between FtsI and FtsL, has not been so far described. The reason why these interactions have not been identified could be that the double hybrid S. cerevisiae, much exploited for these investigations, may not be an appropriate tool, as already hypothesized by Guzman et al. (1997)
. In fact, these authors suggest adopting a prokaryotic double hybrid in order to detect the interactions between division proteins. In any case, the present work aims at identifying the subdomains of the FtsQ protein specifically involved in the interaction with the various partners identified during this study.
As for the pair FtsAFtsI, results were obtained only with the combination cI434FtsAcIP22FtsI (i.e. with FtsA cloned in a low-copy-number plasmid) and with an IPTG induction of 90 min. The bacterial strain carrying the two plasmids does not grow under other experimental conditions, suggesting that a precise ratio of these proteins is necessary for cell physiology.
A new protein, ZapA, has recently been discovered in Bacillus subtilis which localizes to the Z-ring and stabilizes it, probably by promoting the bundling of FtsZ protofilaments (Gueiros-Filho & Losick, 2002; for a review, see Margolin, 2003
). We studied the interaction between the orthologous E. coli ZapA protein, inferred from the B. subtilis zapA sequence, and the division gene products. We observed that ZapA strongly interacts with both FtsZ and FtsA (19 % and 25 % of residual
-galactosidase activity, respectively) whereas no interaction was detected with ZipA. These results suggest that this protein may act early in E. coli, perhaps stabilizing the newly formed Z-ring, as hypothesized by Gueiros-Filho & Losick (2002)
in B. subtilis. Further investigations will be necessary to elucidate the possible role of this protein in E. coli cell division.
A model for septosome assembly
It was possible to develop a model of the temporal sequence of assembly according to the following premise. If protein A is the first to be recruited in the division site and interacts with protein B but not with C, while B and C interact with each other, we may conclude that the protein assembly sequence is A/B/C. Taking this observation into account, the interaction groups described in Table 3 can be subdivided into early interactions, in which proteins FtsZ, ZipA, FtsA, FtsK participate, and late interactions, in which FtsW, FtsN, FtsI participate. It seems that, out of these two groups, FtsL and FtsQ proteins could have a binding role.
These data, together with the data from the literature on Z-ring differentiation, suggest the following model of localization and assembly of division proteins. (a) FtsZ polymerization and Z-ring formation are the first steps towards septosome differentiation (Bi & Lutkenhaus, 1991); since both FtsA and ZipA interact with FtsZ but not with each other, they could interact with FtsZ early but independently from each other as described by Liu et al. (1999)
and by Hale & de Boer (1999)
. This result agrees with previous data on FtsA and ZipA recruitment at the Z-ring level (Ma et al., 1996
; Addinal et al., 1996
; Addinal & Lutkenhaus, 1996
; Ma et al., 1997
). (b) Soon after, FtsK, the membrane protein able to interact with the cytoplasmic FtsZ, is recruited since localization data show FtsK to be dependent upon FtsA and ZipA (Hale & de Boer, 2002
). (c) At this point, two proteins follow: FtsL, by interacting with both FtsK and FtsW, and FtsQ, by interacting with FtsK and FtsL. (d) FtsW, in turn, interacts with FtsI and FtsN, which also interact with each other. In addition, FtsI interacts with FtsK but its localization to the septum depends on the presence of FtsW. FtsI also interacts with FtsQ and FtsA. Before interacting with the other proteins, FtsI and FtsN probably dimerize or polymerize, since the two-hybrid assay indicated their ability to homodimerize. Since FtsL, FtsW, FtsI and FtsN recruitment depends on FtsQ (Addinall et al., 1997
; Ghigo et al., 1999
; Weiss et al., 1999
), it is plausible to believe that it localizes after FtsK and that, at a later stage, the other proteins could interact with FtsQ for septosome assembly.
The assembly sequence of the division proteins can thus be represented by the linear sequence FtsZ, FtsA, ZipA, FtsK, FtsQ, FtsL, FtsW, FtsI, FtsN. Our model agrees with that reported by Chen & Beckwith (2001) and Mercer & Weiss (2002)
.
Potential importance and limitations of the findings
It is often desirable to be able to assay the ability of a particular protein domain to associate with one or several alternative partners, or various proteins which, by interacting with each other, can assemble in a complex machinery. In this context, various two-hybrid assays have been developed, in both prokaryotes and eukaryotes, to study proteinprotein interactions.
In this paper, we have shown that our assay, which has already proved to be very versatile in the study of prokaryotic and eukaryotic protein interactions, is also a powerful instrument for an in vivo study of the interaction and assembly of proteins, as in the case of septum division formation.
As far as the reliability of the assay is concerned, we addressed the problem regarding the ability to discriminate between direct interactions and false positives or negatives according to the reasoning here described. False positives could be due to bridging proteins. As septum proteins are likely to be present in a large complex, bridging proteins could contribute to at least some of the positive results of the assay. Although it would be rather difficult to rule out this hypothesis (which would require a comparison between these results and those obtained by other groups through other methodologies), as regards the present work, recent genetic data seem to minimize the relevance of false molecular interactions.
We selected FtsZ mutants impaired in homodimerization. One of these mutants, ZD4 (for which fusion with GFP shows that the protein is diffused in the cytoplasm and that it does not localize in the septum), is unable to homodimerize by our bacterial two-hybrid assay although it still interacts with FtsA (unpublished data). FtsA normally interacts with FtsZ and, if we follow the reasoning applied to false positives, then it should behave as a bridging protein; this possibility should have revealed itself through our assay as a false positive interaction for FtsZ. Instead, this was not observed even when the intracellular level of FtsA was increased. Formally, such an approach could potentially be extended in order to exclude possible false positives from the assay.
Negative results could be due to the fact that one or both the proteins to be tested are not present at all in the cell (due to a defect in expression or translation of the coding gene or to protein degradation) or are present in insufficient amount. However, if a protein pair does not show any interaction but each of the two proteins is able to interact with another protein, the negative result could be significant. Moreover, one cannot exclude that a particular fusion might fold so as to expose the surface for interaction with one partner but not another.
In conclusion, although the data here reported need confirmation through other methodologies, they represent an important step in obtaining an accurate dissection of E. coli septum differentiation.
![]() |
ACKNOWLEDGEMENTS |
---|
![]() |
REFERENCES |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Addinall, S. G., Bi, E. & Lutkenhaus, J. (1996). FtsZ ring formation in fts mutants. J Bacteriol 178, 38773884.[Abstract]
Addinall, S. G., Cao, C. & Lutkenhaus, J. (1997). FtsN, a late recruit to the septum in Escherichia coli. Mol Microbiol 25, 303309.[Medline]
Alberts, B. (1998). The cell as a collection of protein machines: preparing the next generation of molecular biologists. Cell 92, 291294.[Medline]
Bi, E. & Lutkenhaus, J. (1991). FtsZ ring structure associated with division in Escherichia coli. Nature 354, 161164.[CrossRef][Medline]
Buddelmeijer, N., Judson, N., Boyd, D., Mekalanos, J. J. & Beckwith, J. (2002). YgbQ, a cell division protein in Escherichia coli and Vibrio cholerae, localizes in codependent fashion with FtsL to the division site. Proc Natl Acad Sci U S A 99, 63166321.
Chen, J. C. & Beckwith, J. (2001). FtsQ, FtsL and FtsI require FtsK but not FtsN for co-localization with FtsZ during Escherichia coli cell division. Mol Microbiol 42, 395413.[CrossRef][Medline]
Dai, K., Xu, Y. & Lutkenhaus, J. (1993). Cloning and characterization of ftsN, an essential cell division gene in Escherichia coli isolated as a multicopy suppressor of ftsA12(Ts). J Bacteriol 175, 37903797.[Abstract]
Dente, L., Cesareni, G. & Cortese, R. (1993). pEMBL: a new family of single stranded plasmids. Nucleic Acids Res 11, 16451655.
Di Lallo, G., Castagnoli, L., Ghelardini, P. & Paolozzi, L. (2001). A two-hybrid system based on chimeric operator recognition for studying protein homo/heterodimerization in Escherichia coli. Microbiology 147, 16511656.
Donachie, W. D. (2001). Co-ordinate regulation of the Escherichia coli cell cycle or The Cloud of Unknowing. Mol Microbiol 40, 779785.[CrossRef][Medline]
Ghigo, J. M., Weiss, D. S., Chen, J. C., Yarrow, J. C. & Beckwith, J. (1999). Localization of FtsL to the Escherichia coli septal ring. Mol Microbiol 31, 725737.[CrossRef][Medline]
Gueiros-Filho, F. J. & Losick, R. (2002). A widely conserved bacterial cell division protein that promotes assembly of the tubulin-like protein FtsZ. Genes Dev 16, 25442536.
Guzman, L. M., Weiss, D. S. & Beckwith, J. (1997). Domain-swapping analysis of FtsI, FtsL, and FtsQ, bitopic membrane proteins essential for cell division in Escherichia coli. J Bacteriol 179, 50945103.[Abstract]
Hale, C. A. & de Boer, P. A. (1997). Direct binding of FtsZ to ZipA, an essential component of the septal ring structure that mediates cell division in E. coli. Cell 88, 175185.[Medline]
Hale, C. A. & de Boer, P. A. (1999). Recruitment of ZipA to the septal ring in Escherichia coli is dependent of FtsZ and independent of FtsA. J Bacteriol 181, 167176.
Hale, C. A. & de Boer, P. (2002). ZipA is required for recruitment of FtsK, FtsQ, FtsL and FtsN to the septal ring in Escherichia coli. J Bacteriol 184, 25522556.
Hale, C. A., Rhee, A. C. & de Boer, P. A. (2000). ZipA-induced bundling of FtsZ polymers mediated by an interaction between C-terminal domains. J Bacteriol 182, 51535166.
Haney, S. A., Glasfeld, E., Hale, C., Keeney, D., He, Z. & de Boer, P. (2001). Genetic analysis of the Escherichia coli FtsZ-ZipA interaction in the yeast two-hybrid system. Characterization of FtsZ residues essential for the interactions with ZipA and with FtsA. J Biol Chem 13, 1198011987.[CrossRef]
Hu, J. C., Kornacker, M. G. & Hochschild, A. (2000). Escherichia coli one- and two-hybrid systems for the analysis and identification of protein-protein interactions. Methods 20, 8094.[CrossRef][Medline]
Liu, Z., Mukherjee, A. & Lutkenhaus, J. (1999). Recruitment of ZipA to the division site by interaction with FtsZ. Mol Microbiol 31, 18531861.[CrossRef][Medline]
Ma, X., Ehrhardt, D. W. & Margolin, W. (1996). Colocalization of cell division proteins FtsZ and FtsA to cytoskeletal structures in living Escherichia coli cells by using green fluorescent protein. Proc Natl Acad Sci U S A 93, 1299813003.
Ma, X., Suun, Q., Wang, R., Singh, G., Jounietz, E. L. & Margolin, W. (1997). Interaction between heterologous FtsA and FtsZ proteins at the FtsZ ring. J Bacteriol 179, 67886797.[Abstract]
Margolin, W. (2000). Themes and variations in prokaryotic cell division. FEMS Microbiol Rev 24, 531548.[CrossRef][Medline]
Margolin, W. (2003). Bacterial division: the fellowship of the Ring. Curr Biol 13, R16R18.[CrossRef][Medline]
Mercer, K. L. & Weiss, D. S. (2002). The Escherichia coli cell division protein FtsW is required to recruit its cognate transpeptidase, FtsI (PBP3), to the division site. J Bacteriol 184, 904912.
Miller, J. H. (1972). Experiments in Molecular Genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Moy, F. J., Glasfeld, E., Mosyak, L. & Powers, R. (2000). Solution structure of ZipA, a crucial component of Escherichia coli cell division. Biochemistry 39, 91469156.[CrossRef][Medline]
Ohashi, T., Hale, C. A., de Boer, P. A. & Erickson, H. P. (2002). Structural evidence that the P/Q domain of ZipA is an unstructured, flexible tether between the membrane and the C-terminal FtsZ-binding domain. J Bacteriol 184, 43134315.
Rothfield, L., Justice, S. & Garcia-Lara, J. (1999). Bacterial cell division. Annu Rev Genet 33, 423448.[CrossRef][Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Wang, X., Huang, J., Mukherjee, A., Cao, C. & Lutkenhaus, J. (1997). Analysis of the interaction of FtsZ with itself, GTP, and FtsA. J Bacteriol 179, 55515559.[Abstract]
Weiss, D. S., Chen, J. C., Ghigo, J. M., Boyd, D. & Beckwith, J. (1999). Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z ring, FtsA, FtsQ, and FtsL. J Bacteriol 181, 508520.
Received 18 June 2003;
revised 3 September 2003;
accepted 12 September 2003.
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
J MED MICROBIOL | ALL SGM JOURNALS |