The present invention relates to methods to identify antibiotics which affect a previously unknown essential factor in gram negative bacterial cell division Loopine 1 (LOPl). The present invention also relates to materials which affect the expression or activity of LOPl, such as anti-LOPl antibodies or aptamers and iRNA or derivatives of LOPl .

The life and survival of all organisms is dependent on their ability to divide. Understanding the cell division process requires accurate knowledge of cell enlargement, location and timing of the division, which include complex biological processes and requires careful coordination to initiate and complete this event. This process is particularly a challenge for prokaryotic cells, which are devoid of organelles or centrosomes. Until now, more is known about the start of this process than the end. FtsZ was the first protein identified to localise at the midcell furrow during bacterial division (Bi and Lutkenhaus, 1991). FtsZ is a GTPase functionally and structurally homologous to eucaryotic tubulin. FtsZ polymerises at midcell forming a large ring-like network at the cell membrane, known as the Z-ring (Bi and Lutkenhaus, 1991 ; Chen et al., 1999). The formation and subsequent constriction of the Z-ring leads to the recruitment of other essential proteins forming a mature divisome. This mature complex contains all the proteins recruited for lateral cell wall biosynthesis and completion of septation.

The divisome is composed of at least 9 essential proteins each of which plays a direct role in the cell division process (ZipA, FtsA, FtsK, FtsQ, FtsL, FtsB, FtsW, Ftsl and FtsN) (reviewed in (de Boer, 2010)). The functions of these proteins and the dynamics of their interaction in the division cycle are far from understood. In addition to the essential division factors, there are an increasing number of accessory proteins recruited to the divisome, some of which are conditionally essential depending upon the environment the bacteria are replicating in. In E. coli, midcell localization of the Z-ring is mediated by the oscillating Min system (MinC, MinD and MinE) ((Raskin and de Boer, 1997) and reviewed in (Rothfield et al., 2005; de Boer, 2010). Thus, septation of the two daughter-cell is mediated through FtsZ positioning. The formation of which dictates the position of the mature divisome and thus the site of septation. (Adams and Errington, 2009).

FtsZ assembly into filaments depends on GTP binding but not hydrolysis (Mukherjee and Lutkenhaus, 1994). Compared to tubulin FtsZ polymers contain a FtsZ-GTP and FtsZ-GDP mixture (Oliva et al., 2004) (Bi and Lutkenhaus, 1991 ; Romberg and Mitchison, 2004). The highly dynamic nature of FtsZ polymers is mediated by GTP hydrolysis leading to the disassembly and reduction in length of the protofilaments (Bi and Lutkenhaus, 1991 ; Mukherjee and Lutkenhaus, 1998; Chen et al., 1999) (Strieker et al, 2002; de Boer, 2010). In eukaryotic cells, MAPs (microtubule associated proteins) control the stability, bundling and disassembly of tubulin polymers. To date only FtsZ polymerization inhibitors have been identified, including SulA (Bi and Lutkenhaus, 1991 ; Trusca et al., 1998) (Bi and Lutkenhaus, 1991 ; Mukherjee et al., 1998; Chen et al., 1999) and MinC (Hu et al., 1999; de Boer, 2010). ZipA was shown to protect FtsZ from ClpXP-degradation (Raskin and de Boer, 1997; Pazos et al., 2012). However, the mechanisms of constriction of the Z-ring and its control remains unknown.

The inventors have now elucidated a key aspect of cell division in gram-negative bacteria and in relevant part have identified a novel cell division protein called Loopin 1 (Lopl), that is conserved among Gram-negative bacteria and that plays a key role in the disassembly of the Z ring at the final stages of cell septation. Lop 1 is an ATP-dependent serine protease that is transiently recruited to the Z-ring at the onset of the mother cell constriction to trigger the ATP-dependent proteolysis of FtsZ and Z-ring constriction leading to the physical separation of the two daughter cells.

In accordance with a first aspect of the present invention there is provided a method to identify substances which affect bacterial cell division by interfering with the function of LOP 1 , comprising the steps:

a) Bringing into contact a purified protein selected from the group:

FtsZ, FtsQ, FtsL, Ftsl and FtsN; with purified LOP1 protein;

b) Assaying the formation of complexes between LOP1 and the selected other protein in the presence and absence of a substance to be tested; c) Selecting substances from step b) which affect the formation of complexes when present.

The inventors have shown for the first time that LOPl plays an essential role in cell division in gram-negative bacteria. The inventors have shown via disrupting the function or expression of the LOPl gene that the resulting bacteria show very aberrant cell division phenotypes. The inventors have characterised the parts of the divisome with which LOPl interacts namely FtsZ, FtsQ, FtsL, Ftsl and FtsN.

According to this aspect of the present invention there is provided a method to look for substances which affect the interaction of LOPl with one or more of these portions of the divisome. Examples of substances include inorganic or organic chemical molecules, as well as substances such as antibodies or aptamers which specifically bind to LOPl or one of its partners.

The formation of complexes between LOPl and one or more its target proteins can be monitored via a number of different means, for instance the detection of direct protein-protein interactions using conventional direct observational means such as spectroscopy or via indirect measurements such Surface plasmon resonance (SPR). In addition one or both of the proteins maybe labelled using a tag and then measurements made of their interaction using Fluorescence resonance energy transfer (FRET) or resonance energy transfer (RET). Such assay methods also include radioimmunoassays, competitive-binding assays, co-immunoprecipitation, pulldown assay, Western Blot analysis, antibody sandwich assays, antibody detection and ELISA assays.

In addition means of monitoring the formation of complexes between LOP l and one its partners can also be made based upon the alteration the partner as a consequence of its interaction with LOPl . For instance the inventors have shown that FtsZ polymers are degraded by LOPl and more specifically the self- proteolysed fragment LOPlAi. ₅₉ of LOPl . In accordance with this aspect of the invention therefore the measurement of complex formation may also be made by determining the degradation of for instance FtsZ. An example of a substance which would inhibit this degradation is benzamidin, a serine protease inhibitor shown in the examples below by the inventors to prevent FtsZ polymer degradation when in the presence of LOP l Δι- _> <>.

In accordance with this aspect of the present invention there is provided a method to screen for a substance affecting cell division in a gram negative bacteria comprising the steps:

a) Incubating FtsZ polymers with LOPl in the presence and absence of a substance to be tested;

b) Assaying the degradation of said FtsZ polymers in the presence and absence of a substance to be tested;

c) Selecting substances which when present in step b) affect the degradation of FtsZ polymers.

The inventors have carefully characterised the effect of LOPl upon its partner FtsZ, which is that it is transiently recruited to the Z-ring at the onset of the mother cell constriction to trigger the ATP-dependent proteolysis of FtsZ and Z-ring constriction leading to the physical separation of the two daughter cells.

Substances which affect either the recruitment of LOP l to the Z-ring and/or its proteolytic activity upon the FtsZ polymers comprised in the Z-ring, would affect cell division and hence represent a new class of antibiotic.

In accordance with this aspect of the present invention the inventors have developed a novel fluorescence based assay which measures FtsZ polymer proteolysis by monitoring the fluorescence of a solution comprising a FtsZ/FtsZ-GFP mixture. Polymerisation of the FtsZ/FtsZ-GFP mixture leads to an increase of the solution fluorescence. The further addition of LOP 1 Δ 1. leads to a degradation of the FtsZ/FtsZ-GFP polymers and hence a reduction in fluorescence. In accordance with the present invention either full length LOPl (SEQ ID NO: 25) or a truncated version LOPlAi-59 (SEQ ID NO: 26) comprising the N-terminal portion may be used in the methods according to the present Patent Application.

The inventors have shown that LOPl undergoes ATP-dependent self-proteolysis leading to an active N-terminal portion comprising 59 residues which has serine protease activity, this active fragment is referred to as LOPl Ai_59 (SEQ ID NO: 26). In accordance with the present invention there is provided a further method to identify substances which affect either the auto-proteolysis and/or ATP hydrolysis of LOPl , comprising the steps:

a) Incubating LOPl with a substance to be tested in the presence and absence of ATP;

b) Monitoring the formation of LOP 1 Aj ,59;

c) Selecting substances which when present in step b) decrease the formation of LOP 1 Ai .59.

In accordance with a further aspect of the present invention there is provided a method to identify substances which affect the serine protease activity of LOP 1Δ i _-5 , comprising the steps:

a) Incubating LOPl Aj-sg with a target protein comprising at least one serine protease target site, in the presence and absence of a substance to be tested;

b) Monitoring the cleavage of the target protein;

c) Selecting substances which when present in step b) decrease the cleavage of said target protein.

The serine protease activity of LOPlAi _-5 9 on FtsZ polymers in the Z- ring is a mechanism is associated with the constriction of the Z-ring and hence cell division. Substances which affect the serine protease activity of LOPl Δι _-59 therefore will disrupt cell division and hence represent a new class of antibiotic.

The detection and quantification of serine protease activity is well known in the art and several established methods exist such as Colorimetric or Fluorescent Detection methods (Sigma PC0100& PF0100, Twining, 1984).

In accordance with a preferred embodiment of the present invention the target protein is a FtsZ polymer.

In accordance with a further aspect of the present invention there is provided an inhibitor of the activity or expression of LOPl or an active derivative thereof selected from the group antibodies, aptamers, anti sense RNA or antisense DNA molecules or ribozymes.

Given the essential role of LOPl in cell division in gram negative bacteria an inhibitor of the activity of LOPl such as an anti-LOPl antibody or aptamer; or an inhibitor of the expression of LOPl such as an interference polynucleotide such as iRNA or siRNA, iDNA, siRNA or ribozymes; would be useful in interfering with the division of bacteria and hence represents a new class of antibiotic or research material.

In addition to the listed materials, any other means of affecting the activity or expression of LOPl are also comprised within the present invention for instance alternative antibody replacement technologies such as nanofitins or alternative si-nucleotide systems such as shRNA.

For a better understanding of the invention and to show how the same may be carried into effect, there will now be shown by way of example only, specific embodiments, methods and processes according to the present invention with reference to the accompanying drawings in which:

Figure 1. Lopl encodes for an ATPase. lopl inaetivation leads to a temperature-dependent elongated phenotype of E. coli K12 and Shigella flexneri (M90T).

(A) Partial primary sequence alignment of Lopl proteins from different species. Alignment was made using the ClustalW software and identification of putative Walker A and Walker B consensus sequences are highlighted. Alignment shows in Lopl homologues identified in E. coli K12 MG1655 (B3232, YhcM), Shigella flexneri 5A M90T (S3487), Salmonella typhi (STY3526), Yersinia pestis (YP03564), Candida albicans (AFG11) and Homo sapiens (LACE1). Sequence ID, % of homology and % of identity and P-loop sequence are detailed in Table 3. Lopl ATPase activity was demonstrated using a silica layer chromatography technic (see Figure 9B for Lopl 84A control).

(B and C) Temperature-dependent phenotypic analysis of E. coli (K12) and Shigella (M90T) wild-type and Alopl mutants {K\2::Alopl (b3232, yhcM), M90T::Alopl (s3487)), avirulent M90T V?-::Alopl and the complemented strains (E. coli Kl2: Alopl /plopl-GFP, M90T::A/opi/p/op/-GFP). Bacteria were grown in a rich liquid media at the indicated temperature until an OD ₆ oo=0.5 was reached. Scale bars are 10 μη

(D) Bacteria length measurement performed on each strains and conditions described in IB and 1C panels (see also Figure 8D) using the MicrobeTracker software (Sliusarenko et al.» 201 1). Three independent bacterial cultures were imaged for each strain in each growth condition, 'n' indicates the total number of measured bacteria per condition. *** indicates statistical significance between highlighted conditions, < 0.001 and ** indicates < 0.01 (Student's T test).

(E) Protein stability assay. E. coli K12 Lop l -H,, or LoplA] .59-H ₆ (80μg) were incubated at 37°C during 10 min in a TrisHCl 50 niM pH=7.4 buffer containing 5mM MgCl ₂ in the presence of indicated concentration of ATP. SDS- PAGE gel with Coomassie staining.

(F) Transmission electron microscopy (TEM) analysis of Lopl-H ₆ or LoplAi-59-H ₆ polymer formation on samples described in (E) in the absence or presence of ATP (ImM). Samples were stained with 1% uranyl acetate. Bars are 200 nm.

(G) The solubility of Lop l -I I ₆ was assessed as described in (C) with 200 μg protein after ultracentrifugation (80000 rpm, l lmin, 4°C). SDS-PAGE gel with Coomassie staining.

Figure 2. Lopl is a cytoplasmic protein, which interacts with

FtsZ and is required for Z-ring shape stabilization.

(A) A bacterial two-hybrid assays were performed using the T25- lopl 12 (E. coli K12) or Ί25-Ιορ1 M90T {Shigella M90T) versus T18- zipAlftsAlftsKlftsQIftsLlftsIlftsN plasmid constructs. Results are expressed in Miller Units and averaged from three independent experiments. Error bars show the S.D.. Comparing average activity to the T18 negative control, ** indicates p < 0.01 and *** indicates p < 0.001 (Student's T-test)

(B) The interaction between E. coli FtsZ-GFP and E. coli Lopl-H6, Lopl _K84A -H ₆ and LoplAi.sg-FL; respectively was analysed using an His-pullown assay. Schematic representation of Lopl, Lopl _K 8 _4A and LoplAi- ₅₉ key amino acids involved in ATP binding site and cleavage site.

(C) The interaction between K12 Lopl-H ₆ and ZipA-GFP (pDSW242), FtsQ-GFP (pDSW240), Ftsl-GFP (pDSW234), FtsL-GFP (pDSW326) and FtsN-GFP (pDSW238) was analysed using an His-pulldown approach.

(D) Localization of the FtsZ-GFP (pDSW230, represented in the upper panel) protein fusion in K12 wild-type (wt) and Κ12::Δ/ορ7 strains grown in minimum media at 37°C in the absence of IPTG until an OD ₆ oo ⁼ 0.5 was reached. Results are representative of three independent experiments. Bars are 2 μιη.

(E) Localization of the FtsZ-GFP in K12 and K\2::Alopl strains grown in rich media (LB) at 37°C, until an OD ₆₀ o=0.5 was reached. Results are representative of three independent experiments. Bars are 5 μηι.

(F) Western blotting of FtsZ and Lopl in 12 and Κ12::Δ/ορ7 strains using rabbit polyclonal antibodies on supernatant (Sup.) and pellet fractions.

(G) Lopl localization was performed in K12 and MvAlopl strains by electron microscopy using immunogold staining with a polyclonal a-Lopl antibody (1 : 1000) and Protein A gold labelled on cryosections. Bars are 200 nm.

Figure 3. Lopl co-localises with constricting Z-ring. The N- terminal fragment of Lopl is required for the Z-ring association.

(A) FtsZ-GFP and Lopl-mCherry time-dependent expression and localization during a cell division process observed in a K\2::Alopl /pDSW230/p/o - mCherry strain. Time-lapse observation was performed on a LB-agar pad at 30°C, using a 200M Axiovert epifuorescent microscope (Zeiss). Image acquisition was performed every 3 min. This result is representative of five individual observations from three independent experiments. Bars are 2 μπι.

(B) FtsZ-GFP and Lop 1 -mCherry fluorescent signals (AU) were quantified in relation with the distance from the Z-ring center (as indicated on the left- hand scheme). Measurements were performed on images acquired at the maximal constriction (Max. constriction) and respectively 50 min, 20 min, 15 min, 10 min, 5 min before. n=5 independent observations, error bars show the S.D.

(C) Lopl -mCherry mean signal (AU) and Z-ring diameter (μιη) were calculated for each time-point described in (B). n=5 independent observations, error bars show the S.D.

(D) Representation of the Lopl -mCherry mean signal (AU) in relation with the Z-ring diameter represented in (C). n=5 independent observations, error bars show the S.D. *** indicates statistical significance p< 0.001 , (Student's T test). (E) FtsZ-GFP and Lop l Ai _-59 -mCherry time-dependent expression and localization during a cell division process observed in a K12: :A/o i/pDSW230/p/op7Ai.59-mCherry strain. Time-lapse observation was performed as described in (A). This result is representative of three individual observations from three independent experiments. Timing is indicated in min. Bars are 2 μιτι.

Figure 4. Lopl overexpression leads to cell-shape modification. The ATP binding-site and the N-terminal 1-59 aa fragment are required for Lopl function

(A) Schematic representation of plop I -I I<„

¾ and plopl i _-59 -H ₆ .

(B) Cell-shape modifications of K12: :Alopl /plopl -H ,

¥Α2νΑΙορ1/ρΙορ1κ84Α-Ή-6, Κ12::Δ/ορ7/ρ/ορ Δ _{1 -5} 9-Η ₆ and K\ 2: :Alopl/plopl sg-He strains grown in LB liquid media at 37°C in the presence of IPTG, as indicated until the OD ₆ oo ⁼ 0.5 was reached. These results are representative of at least three three independent experiments. Bars are 2 μη .

(C) In order to analyse the turn-over of Lop l -I l„. Lopl K84A-¾ and Lo l Ai_59-H ₆ proteins fusions overexpression, the constructs described in (B) were grown in LB liquid media at 37°C in the presence of IPTG 0.1M until the OD ₆ oo ⁼ 0.5 was reached before growing them in a fresh LB media without IPTG (time t=0). Bars are 2 μηι.

(D) Western blotting of Lopl -H,,. Lopl _K 84A-H6 and ίορ1 Δι _{- 9} -Η ₆ (a- His) performed on samples described in (C) in supernatant (Sup.) and pellet fractions at time 0, 30, 45 and 60 min. These results are representative of three independent experiments. Lo l A _{1 -5} 9 is indicated with a white arrow.

(E) Western blot analysis (a-Lopl and -FtsZ) performed on supernatant (Sup.) and pellet fractions from K 12: :AlopI/plop]-\ I,,. Kl 2 :Alopl/plopl KS4A-VL6, K\ 2: :Alopl/plopl AI. ₅₉ -K(, described in (C) at time 0, 30 and 60 min. These results are representative of three independent experiments. Lopl Ai_59 is indicated with a white arrow. Figure 5. Lopl overexpression leads to the Z-ring destabilisation.

(A) FtsZ-GFP expression using the p.IC 104 vector (Mukherjee et al., 2001) in the K12 and \ 2: Alopl strains grown in LB liquid media at 37°C in the absence of arabinose, until the OD ₆₀ o=0.5 was reached. Bars are 2 μιη

(B) FtsZ-GFP expression (pJC104) and localization upon Lopl- mCherry ( pSU-/ /?/-mCherry) and mutated forms (pSU-/op/K84A-mCherry, pSU- /op A] .59-mCheny) overexpression. Bacteria were grown in LB media at 37°C in the presence of IPTG, as indicated until an ODgoo^O.S was reached. These results are representative of two independent experiments. Bars are 2 μιη.

Figure 6. In vitro, Lo lAi^ catalyses the FtsZ proteolysis in an ATP-independent manner.

(A) The stability of FtsZ polymer (produced as described in Figure 16) was assessed in a reaction mixture containing 50 mM Hepes, 50 mM KCl, 5 mM MgCl ₂ and equimolar quantities of Lopl-¾, Lopl _K 84A-H6 or LoplAi _-5 9-H ₆ and 1 mM ATP when indicated, for 3 min at 30°C. FtsZ polymers containing pellet fraction (P) was separated from the soluble fraction (S) by ultracentrifugation. Lop 1 -1 L (and mutated forms) and FtsZ were detected in both fractions by western blot using rabbit polyclonal antibodies. The results are representative of four independent experiments.

(B) TEM observation of Z-ring formation was performed by negative staining in a reaction mixture containing 50 mM Hepes, 50 mM KCl, 5 mM MgC - in the presence of 10 mM of CaCl ₂ , as indicated previously (Yu and Margolin, 1997b). The reaction occurred at 30°C during 3 min, in the presence of 1 mM GTP, 30 μg/mL purified FtsZ, and equimolar quantities of purified Lop l -H ₆ , Lo l i 84A-H ₆ or LoplA _{1 -5} 9-H6 and 1 mM ATP or 5 mM EDTA or 1 mM PMSF when indicated. The results are representative of three independent experiments.

(C) Co-expression of FtsZ-GFP and Lop l -mCherry in the K 12 : : Δ/ο/? / /p I ) S W 230· ^! ρΙο J -m C herry strain grown in LB liquid media at 37°C in the absence of IPTG, until the 00(500^0.5 was reached. Bars are 2 μηι. This result is representative of ten individual observations from four independent experiments. Bar is 2 μπι. In graph, n represents the total number of measured bacteria. Error bars show the S.D., ** * indicates p < 0.001 (Student's T-test)

(D) K 12 : : / /p D S W 230/plop / - m C licrry strain was grown on a LB-agar pad at 30°C in the absence of IPTG. Imaging was performed from 0 to 120 min, as indicated, using a 200M Axiovert epifuorescent microscope (Zeiss). Bars are 3 μηι.

(E) FtsZ polymer proteolysis by Lopl Ai.59-H ₆ was assessed as described in (A) in the presence of 1 niM ATP and 5 niM EDTA, 2.5 μg/mL pepstatin, 1 mM PMSF or 10 μg/mL leupeptin. Ι_.ορ1 Δ _ΐ 9-Η ₍ , and FtsZ were detected in the pellet fraction by western blot using rabbit polyclonal antibodies. The results are representative of four independent experiments.

( F) FtsZ polymer proteolysis by H ₆ -LoplAi. ₅ 9 was assessed during 1 min at 30°C as described in (A) in the presence of H6-Lopl A _1-59 or H ₆ -Lopl Ai _-5 9A _303- 375. Proteins were detected in the pellet fraction by western blot using rabbit polyclonal antibodies. The results are representative of three independent experiments.

Figure 7. Graphical abstract. Schematic representation of the proposed model for Lopl-promoted Z-ring proteolysis, leading to its constriction.

In this study, the inventors demonstrate that Z-ring dynamic constriction is promoted by a novel ATPase named Loopin 1 (Lopl ), according to its function. The N-terminal extremity of Lopl is required for its interaction with FtsZ in vitro and in the bacteria. In an ATP-dependent manner, this N-terminal (1-59) fragment is cleaved by autoproteolysis. Lo l Ai_59 is a serine protease catalyzing the proteolytic cleavage of FtsZ -polymers. Taken together, these results suggest that the Z-ring constriction process is the consequence of an active proteolysis promoted by Lopl Ai _-5 9 on the Z-ring. The autoproteolyis activity of LoplAi. ₅₉ observed in vitro is a first element of an auto-regulation of this process, allowing a new cycle of division to be initiated. Figure 8. The Shigella flexneri 5A Μ90ΤΔ/ορΙ mutant is attenuated in vivo.

(A) Sequence comparison between the Shigella flexneri and E. coli Lopl protein sequences (SFV3259 and B3232 (YhcM) respectively) using the ClustalW software. The GGXGVXKT ATP-binding site is highlighted in purple.

(B) Competitive index (C.I.) of Shigella flexneri 5A lopl tansposon mutant (M90Tmut6), lopl mutant (Μ90Τ::Δ/<7/> / ) and complemented strain (M90T: : Δ/ / / 1 plop I -G F P M90T) in vivo. The C.I. assessed the ability of each mutant to colonize the rabbit ileal loop in comparison with the wild-type strain. A C.I. of 1 indicates no attenuation. The results are an average of at least three independent experiments.

(C) Histo-pathological analysis of rabbit ileal loops infected by M90T, M9Q .:Mopl and M90T VP- (BS 176). Paraffin embedded tissues were stained using haematoxylin-eosin. Bars are 50 μηι.

(D) Immunodetection of the M90T and Μ90Τ::Δ/ορ7 strains in the rabbit ileal loop model. DNA is stained with Dapi (blue), actin with RRX-Phalloidin (Red). Shigella strains are labelled using a rabbit polyclonal a-LPS antibody (green). Image acquisition was performed using a confocal microscope. Bars are 5 μηι.

Figure 9. E. coli and Shigella Lopl sequence alignment. Biochemical properties of E. coli Lopl and Lo lAi^.

(A) Bacteria length measurement performed on each strains and conditions described in Figures IB and 1 C panels using the MicrobeTracker software (Sliusarenko et al., 2011). Three independent bacterial cultures were performed for each strain in each growth condition, n indicates the total number of measured bacteria per condition. *** indicates statistical significance < 0.001 , ** p<0,05 and * pO.01 respectively (Student's T test).

(B) Lopl _84A ATPase activity was analysed using a silica layer chromatography technique. The reaction was performed in a TrisHCl 50 mM pH7.4 buffer containing 10 mCi of radiolabeled ΑΤΡγ32 (or GTPy32), 10 mM ATP and 2.5 mM MgCl ₂ in the presence of various Lopl _K 84A-¾ quantities, as indicated. The reaction occurred during 10 min at 30°C. This result is representative of three independent experiments.

(C) Protein stability assay. Lopl -1 1 ₆ or Lopl K _84A -¾ (80μg) were incubated at 37°C during 10 min in a TrisHCl 50 mM pH=7.4 buffer containing 5mM MgClo in the presence of indicated concentration of ATP and when indicated in the presence of 5mM EDTA, 1 mM AMP-PNP. SDS-PAGE gel with Coomassie staining.

(D) Enzymatic parameters (Vmax, Km) of LoplA _{1 -5} 9-H ₆ calculation using 5μΜ of purified enzyme in the presence of various ATP concentrations (0.1 , 0.5, 1 and 10 mM). The initial rates (μΜ Pi. min ^"1 ) were averaged from three independent experiment performed in duplicate.

Figure 10. FtsQ-GFP, FtsL-GFP and FtsN-GFP localization in E. coli K12 wild-type and AIopl strains.

(A) Localization of FtsQ-GFP (pDSW240), FtsL-GFP (pDSW326) and FtsN-GFP (pDSW238) protein fusions in K12 wild-type (wt) and Mopl strains grown in minimum media at 37°C in the absence of IPTG (except FtsL-GFP expression with 10 μΜ IPTG), until an OD ₆ oo=0.5 was reached. Results are representative of three independent experiments. Bacteria were observed using a Nikon Eclipse 80i epifluorescent microscope. Bars are 2 μπι.

Figure 11. FtsZ-GFP expression in 12 and Kl2::Alopl strains. (A) FtsZ-GFP (pDSW230) localization in K12 and \2: Mopl strains during stationary phase performed in LB rich media at 37°C or 42°C. These observations are representative of at least three independent experiments. Bars are 2 μιη.

(B) FtsZ-GFP (pDSW231) time-dependent expression and localization in E. coli K12 wild-type and Δΐορΐ strains during a cell division process.

White arrows indicate normal Z-rings. Time-lapse observation was performed on strains grown on a LB-agar pad at 30°C in the absence of IPTG, using a 200M Axiovert epifuorescent microscope (Zeiss). Image acquisition was performed every 3 min. These results are representative of at least three independent observations. Bars are 2 μηι. Figure 12. pSUC plasmid description.

Schematic representation of the pSUC plasmid map in addition with its multiple cloning site sequence ( Hindi II and Xbal restriction sites are underligned).

Figure 13. Inducible expression of Lopl-mCherry, Lopl _K s- - ni Cherry and LoplAi^-mCherry in E. coli K12.

Inducible overexpression of Lopl-mCherry and mutated forms in K12 (pSU/o/?i-mCherry, and pSU/ p A/_59-mCherry, representative scheme). Bacteria were grown in LB media at 37°C in the presence of IPTG, as indicated until an OD ₆ oo ⁼ 0.5 was reached. The observations were performed using a Nikon Eclipse 80i epifluorescent microscope. These results are representative of two independent experiments. Bars are 2 μηι.

Figure 14. FtsZ polymerization assay.

FtsZ polymerization assay was performed in a reaction mixture containing 50 mM Hepes, 50 niM C1, 5 mM MgCL in the presence of 0.5 ^ig/ml . purified FtsZ and 1 mM GTP when indicated. The reaction was performed at 30°C during 3 min. FtsZ polymers containing pellet fraction (P) was separated from the soluble fraction (S) by ultracentrifugation. FtsZ was detected in both fractions by western blot using a rabbit polyclonal antibody. The results are representative of four independent experiments.

Figure 15. Lopl _k s- and LoplAi.sy expression do not prevent Z- ring formation

Lopl-mCherry and mutative forms expression and Z-ring formation observation in K12::A/o/?7/pDSW230/p/ p7-mCherry,

K 12::A½//pDSW230/p /^ ,-mCherry and Kl2::Ahpl/pOSW230/plopl Δ,. ₅₉ - mCherry strains (schematic representation) grown in LB media without IPTG in a LB rich media at 37°C.

Figure 16. Lopl do not interact with ClpP // vitro.

Gel filtration analysis of the Lopl-H6 and ClpP-H ₆ interaction. (A) 2mg of each His-tagged protein was incubated in Buffer A prior gel filtration analysis. (B) SDS-PAGE analysis of each detected peak between 87 and 93 mL corresponding to the Lopl-H ₆ K12 elution fraction and 1 1 1 and 1 17 mL corresponding to the ClpP- ¾ elution fraction. SDS-PAGE gel was stained with a Coomassie staining. Figure 17: In vitro fluorescence-based assay

As Lo l overexpression seemed to perturb the Z-ring constriction, we aimed at deciphering whether Lopl and Lop 1 Δ 1. 9 act directly on FtsZ polymers in vitro. We designed a fluorescence-based assay as polymerization of a FtsZ/FtsZ-GFP mixture leads to an increase of the solution fluorescence (Trusca and Bramhill, 2002). In the presence of GTP, purified FtsZ and FtsZ-GFP form polymers, as described previously (Yu and Margolin, 1997b); however the level of the detected fluorescence remains low and the polymers length was reduced (Figures 17A and 17B). Indeed upon equimolar addition of Lopl or LoplAj.sg we could observe a significant increase of the solution fluorescence (Figure 17A). Full-length Lopl addition promoted FtsZ/FtsZ-GFP bundling and the formation of large helical three-dimensional polymerized structures (Figure 17B t=0). These structures remained stable over time in the absence of ATP (t=T0min), while they remain no longer stable in the presence of ATP (Figure 17B, t=10min). This observation was correlated with a decrease of the solution fluorescence in in the fluorescence-based assay (Figure 17 A, Student's T test p<0.01), which was consistent with the ATP-dependent autoproteolytic maturation of Lopl described previously (Figure IE). Alternatively, in the presence of LoplAi_59, while we similarly observed a rapid and significant increase of the solution fluorescence level (Figure 17A), we could not observe the FtsZ/FtsZ-GFP helical structures formation, however we could visualize the formation of a homogeneous and dense FtsZ/FtsZ-GFP polymer network (Figure 17B, t=0). Interestingly, these polymers were rapidly degraded (Figure 17B, t=10min), which was correlated with a significant decrease of the fluorescence level (Figure 17C, Student's T test p<0.01). As a control, the simultaneous addition of benzamidin with LoplA] _ ₅ 9 did not impair the fluorescence increase associated to the formation of the FtsZ/FtsZ-GFP polymer network (Figure 17A and 17B), although preventing its degradation (Figure 17B), in association with a stable level of the fluorescent signal (Figure 17A). This experiment showing an inhibition of the proteolytic activity of Lop 1Δ 1.59 by benzamidin suggested that this protein has a serine protease activity.

There will now be described by way of example a specific mode contemplated by the Inventors. In the following description numerous specific details are set forth in order to provide a thorough understanding. It will be apparent however, to one skilled in the art, that the present invention may be practiced without limitation to these specific details. In other instances, well known methods and structures have not been described so as not to unnecessarily obscure the description.

Example 1 - EXPERIMENTAL PROCEDURES Expression plasniid construction

The pSUC vector construction was made by amplifying the mCherry fusion from the pmCherry-Nl vector using the SG150 (SEQ ID NO: 1 1) and SG151 (SEQ ID NO: 12) primer pair (Table 2), introducing the Bam HI and EcoRJ restriction sites. The pSU19 vector was digested with BamHI and EcoRI restriction enzymes prior ligation of the digested mCherry amplified fragment, leading to the generation of the pSUC vector. This expression vector allows the expression of mCherry protein fusions in C-terminal under the control of the gene of interest promoter in E. coli and in Shigella.

This plasmid allowed the expression of Lop 1 -mCherry fusion under the control of the lopl promoter (see below).

The expression of the FtsZ-GFP fusion under the control of a lad promoter was performed using either the pDSW230 and pDSW231 constructs or p.IC 104 (Mukherjee et al., 2001) (kindly provided by Pr. Lutkenhaus) (described in Table 1).

Table 1.

Table 2

Table 3.

Proteins overexpression and purification

K12 Lopl-6xHis, Lopli84A-6xHis, LoplAi-59-6xHis and GpP-6xI lis protein fusions were expressed using the p/o?7-6xHis K12, K12, p/o/?/zl/-j9-6xHis K12 and pNB140 constructs (see Table 1) expressed in an E. coli BL21DE3 strain. Proteins purifications are described below.

The Topi L59/W60 cleavage site identification was performed by automated N-terminal sequence analysis on a Procise ABI 470 (Applied Biosystems).

Native proteins or protein fusions were overexpressed in an E. coli

BL21DE3 strain. Overnight cultures were subcultured in fresh LB media (1 :100) and grown at 37°C until the OD ₆ oo ⁼ 0.5 was reached. Overexpression was induced by the addition of 0.5 mM IPTG and was performed overnight at RT. Lopl-¾, Lopl _K84A -¾, LoplAi-59-H6 His-Tagged proteins were purified on Talon beads (Clontech) and further purified by gel filtration using an Hiload 16/60 Superdex 200 column (GE) in a Tris 50mM pH7.5 buffer containing 5 mM MgCl ₂ , 1 mM EDTA and 0.1M NaCl. FtsZ and FtsZ-GFP were purified by ion exchange on a Hiload 16/10 DEAE column using a Tris 50mM pH7.5 buffer containing 5 mM MgCl ₂ , 1 mM EDTA and 0.1M NaCl (Buffer 1) and Tris 50mM pH7.5 buffer containing 5 mM MgCl ₂ , 1 mM EDTA and 1 M KC1 (Buffer2) as described previously (Yu et al, 1997), followed by a gel filtration, as described above.

FtsZ polymers proteolysis assay

FtsZ polymers (P) were generated as described below during 3 min at 30°C and collected by ultracentrifugation (1 1 min, 80K) at 4°C (Beckman, TL-100 Ultracentrifuge). Then, the reactive buffer was discarded and replaced by a reaction mixture containing 50 mM Hepes, 50 mM KC1, 5 mM MgCl ₂ in addition with 1 mM ATP and 0.5 μg mL of purified Lopl -6xHis, Lopl 8 _4A -6xHis or Lopl Ai.j ₉ -6xHis when indicated in a final volume of 100 The reaction was stopped after 0, 1 or 3 min, as indicated. PMSF (Sigma-Aldrich), EDTA (Sigma-Aldrich), peptstatin (Sigma- Aldrich), leupeptin (Calbiochem) or PMSF (Roche) were added when indicated. FtsZ polymers containing pellet fraction (P) was separated from the soluble fraction (S) by ultracentrifugation (1 1 min, 80K) at 4°C (Beckman, TL-100 Ultracentrifuge). Samples were re-suspended in a Laemli buffer IX final and subsequently subjected to SDS- PAGE gel analysis and transfer onto a nitrocellulose membrane. FtsZ and Lopl- 6x1 lis. Lopl 84A-6xHis or LoplAi. ₅₉ -6xHis were detected in both fractions by Western blot using rabbit polyclonal antibodies (see below). Fluorescent FtsZ/FtsZ-GFP polymers were generated in a buffer containing 50 mM Hepes, 50 mM KC1, 5 mM MgCl ₂ , 10 mM CaCl ₂ in addition with 1 mM GTP. Polymerization of FtsZ (100 μΜ) and FtsZ-GFP (50 μΜ) occurred during 3 min at 30°C in 96-well plates (Greiner Bio One).

Then, Lopl-H ₆ , Lopl _K 84A-¾ or Lopl Ai-59-H ₆ (100 μΜ) was added to reach a in a final volume of 100 μΐ,. The fluorescence was quantified over the time (lOmin, acquisition every 45s) using a SLM 8000C fluorimeter (SLM Instruments). The experiments were performed in triplicate on three independent occasions. As a negative control lmg/mL benzamidine was added at the initial step, when indicated.

Additionally, a similar experiment was performed on glass slides to visualise the formation and proteolysis of FtsZ/FtsZ-GFP polymers using a TCS SP5 confocal microscope (Leica).

FtsZ/FtsZ-GFP polymers containing pellet fraction (P) was separated from the soluble fraction (S) by ultracentrifugation (1 1 min, 80K) at 4°C (Beckman, TL-100 Ultracentrifuge). Samples were re-suspended in a Laemli buffer IX final and subsequently subjected to SDS-PAGE gel analysis and transfer onto a nitrocellulose membrane. FtsZ and Lopl-H ₆ , Lopl _84A -¾ or LoplAi. ₅₉ -H6 were detected in both fractions by Western blot using rabbit polyclonal antibodies FtsZGFP was detected using an anti-GFP antibody (Sigma-Aldrich).

EM observation of Z-ring formation in vitro

In order to allow FtsZ to polymerize as a proper ring (Z-ring), 10 mM of CaCl ₂ were added into a reaction mixture containing 50 mM Hepes, 50 mM KC1, 5 mM MgCl ₂ , as described previously (Yu and Margolin, 1997b; Camberg et al., 2009) (Mateos-Gil et al., 2012). The reaction occurred onto during 5 min at 30°C in the presence of 30 μg/mL of purified FtsZ and of an equimolar quantity of purified Lopl-6xHis, or Lop l Ai. y-6xI lis and 1 mM ATP as indicated. As polymerized proteins might be unstable, the reaction occurred directly on glow discharged copper grids and the reaction was stopped by immersion of the grids in a 2% uranyl acetate solution (see below).

Bacterial strains and growth conditions

The bacterial strains and plasmids used in this study are described in Table S 1. Shigella strains (including S. flexneri) were grown in trypticase soy (TCS) broth or on TCS agar plates supplemented with 0.01% Congo Red (Sigma), when necessary. E. coli strains, as well as Salmonella thyphi were grown in LB media.

DNA manipulations

The initial transposon insertion in S. flexneri M90T in lopl was performed as described previously (West et al., 2005). The construction of inactivated lopl mutants was then performed in E. coli and 5 ^* . flexneri.

Alopl mutants construction. In MG1655 E. coli K12 Plvir page lysate was prepared on the donor strain JW3201 from the Keio collection (Baba et al, 2006) as described (Miller J.H., 1992). In JW3201 strain, the lopl ORF (open reading frame) is substituted by the kanamycin-resistance marker (Alopl ::Kam) (Baba et al, 2006). The cassette Alopl ::Kam was introduced into MG1655 by PI transduction (Miller, J. H. 1992) and selection for kanamycin-resistant (Km ¹ ) colonies was made on LB plates containing kanamycin (50 μg/ml). After re-isolation, several clones were verified by PGR to confirm the right chromosomal structure of the deletion. One clone was chosen and named K12A lopi. :Km.

ΥΛΙν.ΑΙορΙ was then obtained from K12A/o/?7:;Km by removing the kanamycin-resistance marker from the Alop l cassette. In this Alopl ::Km cassette, the antibiotic-resistance marker is flanked by two direct frt repeats, which are the recognition targets for the site specific recombinase FLP (Baba et al, 2006). Therefore, to get rid of the resistance marker from the chromosome, a temperature-sensitive plasmid pCP20 that encodes the FLP recombinase was used (Cherepanov & Wackernagel, 1995). Briefly, K\2A opl::Ksa cells were transformed with pCP20, and chloramphenicol-resistant (Cm ¹ ) colonies were selected at 30°C on LB plates containing the corresponding antibiotic (30 μg/ml). Several of these clones were grown overnight on antibiotic-free LB plates at 42°C. Ten independent colonies were selected and after single-colony passage at 30°C, all ten colonies were no longer Cm and Km ^r , indicating simultaneous loss of pCP20 and the kanamycin-resistance marker from the bacterial chromosome. This FLP-catalysed excision created an in- frame deletion of the lopl ORF, leaving behind a 102-bp scar sequence (Alopl v.frf) (Baba et al, 2006). To confirm the correct chromosomal structure of the deletion several Cm ^s and clones were tested by PGR using the NWpr40 and NWpr41 primer pair (Table 2). After confirmation, one clone was chosen and named K\2::ALopl.

In order to inactivate lopl in Shigella flexneri (M90T), a one-step chromosomal inactivation method was used to target homologous region for integration. Therefore, the inventors generated PCR products with much longer flanking sequence using the K\2::Alopl null mutant as the template. The M90T was transformed with PCR products amplified from Κ12::Δ1ορ1 ::Km mutant genomic DNA using primers NWpr23 and NWpr24 (Table 2). The primers NWpr23 (SEQ ID NO: 20) and NWpr24 (SEQ ID NO: 21) were designed to include 50 bp upstream and downstream sequence flanking lopl. This product was transformed into M90T::pKD46 which resulted in all kanamycin resistant colonies containing the 1.5 kb kanamycin resistance gene when analysed by PCR. Thus, a S. flexneri null mutant was successfully generated (Μ90Τ::Δ/ορ7).

Expressing respectively a lopl-GFP and a lopl-mCherry fusion under the control of lopl promoter performed the complementation of the Μ90Τ.·. Δ/ορ7 and Κ12::Δ/ο/?7 mutants. In order to express a Lopl-GFP fusion, the lopl gene of Shigella and E. coli and their promoters (~500bp) were amplified with the SG127 (SEQ ID NO: 3) and SG128 (SEQ ID NO: 4) primer pair (Table 2) and cloned in pFpV25 vector digested with the BamHI and Ndel restriction enzymes. The p/opi-GFP M90T and p/opi-GFP 12 constructs were obtained and sequenced (Table 1).

In order to express lopl-mCherry, the lopl gene and its promoter (~500bp) were amplified with the SGI 54 (SEQ ID NO: 27) and SGI 55 (SEQ ID NO: 14) primer pair (Table 2) and cloned in pSUC vector digested with the Hindlll and Xbal restriction enzymes. The K84A point mutation of lopl was performed using the SGI 14 (SEQ ID NO: 5) and SGI 15 (SEQ ID NO: 6) primer pair (Table 2). The truncation of the N-terminal part of lopl (Δ1-59) was performed using the SG164 (SEQ ID NO: 15) and SG165 (SEQ ID NO: 16) primer pair (Table 2). Both mutated version of lopl were amplified with the SGI 54 (SEQ ID NO: 27) and SGI 55 (SEQ ID NO: 14) primer pair (Table 2) and cloned in pSUC vector digested with the Hindlll and Xbal restriction enzymes. Respectively the p/opi-mCherry, p/op7 _K 84A-mCherry and p/o/? i.59-mCherry constructs were obtained and sequenced (Table 1). In order to control the expression of / pi-mCherry, lopl KM A- mCherry and op Ai-sg-mCherry (from lopl start codon) with a lad promoter, the corresponding fragments were amplified from the p/o/?7-mCherry, p/o/?iK84A-mCherry the SG278/SG329 (SEQ ID NOs: 17 & 19) primer pair (Table 2) and from the p/o/?7Ai. ₅₉ -mCherry constructs with the SG328/SG329 primer pair (SEQ ID NO: 18 & 19) (Table 2) and cloned in pSU19 digested with the BamHI and the Ecorl restriction enzymes. The pSU-/o/?7-mCherry, pSV -I op 1 Λ sv .(-m herry and pSU-loplA] _-5 9- mCherry constructs were obtained and sequenced (Table 1).

In order to overproduce the Lopl-H ₆ , Lopl 84A-H ₆ , Lop l A]. ₅₉ -H ₆ and Lopl _1-59 -H ₆ protein fusions in an IPTG-dependent manner the corresponding lopl DNA fragment were amplified by PCR prior cloning in the pKJl plasmid, digested at the Ncol and BamHI restriction (Table 1). lopl was amplified using the SG90/SG91 (SEQ ID NOs: 7 & 8) primer pair (Table 2), Ιορ1κ8 _4Α was obtained using the SG150/1G151 primer pair (SEQ ID NO: 1 1 & 12) to introduce a single point mutation K84A (Table 2). Ιορ1Α . ₅₉ was amplified using the SG157/SG91 (SEQ ID NO: 9 & 8) primer pair (introducing an additional Methione at the N-terminus) (Table 2) and lopl 1-59 was amplified using the SG90/SG169 primer pair (SEQ ID NO: 7 & 10) (introducing a 5' stop codon in the ORE) (Table 2). The resulting plop 1 -R , ρίορ1κ _84Α - H(„ ρΙορΙΔι-59- and plopl i-59-Rb constructs were analyzed by PCR and sequenced. In order to generate the H ₆ -Lopl A _1-59 and Η6-ίορ1Δι_5 ₉ Δ303-375 constructs, sub-cloning was performed using LIC-cloning methodology, allowing the generation of the pNIC28-Bsa4-/op7 l/-j9 and constructs.

Rabbit ligated ileal loop model

New Zealand White rabbits weighting 2.5-3 kg (Charles River Breeding Laboratories, Wilmington, MA) were used for experimental infections. For each animal, up to 12 intestinal ligated loops, each 5 cm in length, were prepared as described previously (Martinez et al., 1988; West et al., 2005). For the evaluation of the C.I., an equal quantity of the wild-type strain and of the mutant was injected in each loop (corresponding to a total dose of 10 ⁵ CFU per loop). After 16h, animals were sacrificed and the luminal fluid was aspirated and S. flexneri recovered. C.I. was calculated as the proportion of mutant to wild-type bacteria recovered from animals, divided by the proportion of mutant to wild-type in the inoculums, and results are expressed as the mean of at least 4 loops from two independent animal. The experimental protocol was approved by the Ethic committee Paris 1 (number 20070004, December 9th 2007).

For immunohistochemical staining, infected rabbit ileum samples were washed in PBS, incubated at 4°C PBS containing 12% sucrose for 90 min, then in PBS with 18%o sucrose overnight, and frozen in OCT (Sakura) on dry ice. 7 μιη sections were obtained using a cryostat CM-3050 (Leica). Fluorescent staining was performed using a rabbit anti-Shigella LPS primary antibody (1 :200 dilution) (P. Sansonetti, Institut Pasteur) and an anti-rabbit-FITC conjugated secondary antibody (1 : 1000). Epithelium cell nuclei were stained with Dapi (1 :1000) and actin stained with RRX-Phalloidin (1 :1000). Image acquisition was performed using laser-scanning confocal microscopy. Image analysis was performed using Image J software.

Two hybrids screen

The inventors used the BACTH system that is based on the interaction-mediated reconstitution of an adenylate cyclase (AC) enzyme in the otherwise defective E. coli strain DHM1 (Valdivia and Falkow, 1996; Karimova et al., 1998). This system is composed of two replication compatible plasmids, pKT25 and pUT18, respectively, encoding the intrinsically inactive N-terminal T25 domain and C-terminal T18 domain of the AC enzyme. E. coli and Shigella lopl was amplified using the NG1281 and NG1282 primer pair (SEQ ID NO: 1 & 2) and cloned in pK 1 25 vector.

pKT25 and pUT18 plasmids, which were subsequently doubly transformed to 1)1 IM l to search for AC reconstitution that turns on β-galactosidase production leading to the blue colour after 2 days of growth at 30°C on indicator plates containing Xgal (Eurobio, 40 mg ml-1), isopropyl-l-thio-b-D- galactopyranoside (Invitrogen, 0.5 mM), Ap, Km and nalidixic acid, β-galactosidase activity was measured as described before, averaged from three independent experiments and expressed as Miller Unit (Sansonetti et al., 1982; Karimova et al., 1998). Antibody

a-Lopl antibody production. An -Lopl rabbit polyclonal antibody was collected from two New- Zealand rabbits challenged with purified Lopl-H ₆ (2 mg/mL solution) on four occasions with the purified protein separated by 2-weeks rest. The first injection (500 μί., intradermal) was performed with the purified protein (125 μg) in addition with a complete Freund's adjuvant. The second injection was performed following a similar procedure in the presence of incomplete Freund's adjuvant. The third and the last injection were performed with no adjuvant. Final blood collection was performed by cardiac puncture in heparin-free tube. Sera were separated from blood cells by centrifugation (14000 rpm, 30 min). As a note, the a- Lop antibody obtain following this procedure allow the detection of Lopl -¾ but also the Lo l _84A -¾ or LoplAi _-59 -H ₆ mutated versions of Lop 1 -I I,, by western blot.

α-FtsZ rabbit polyclonal antibody was kindly provided by Pr. Kenn Gerdes (Weiss et al., 1999; Galli and Gerdes, 2010).

Thin Layer Chromatography (TLC) analysis

ATPase and GTPase assays were performed in the presence of BSA 1.25 mg/mL (Sigma), ΑΤΡγ32 or GTPy32 30 μΠ (Perkin Emer), ATP or GTP 50 niM (Sigma) and 0.1 to 10 μg of purified Lopl-H ₆ and Lopl _K84A -¾ as indicated. The final reaction mixture volume is 20 μΐ . in a TMD buffer (Tris pH7.4 25 mM, MgCL 10 mM, DTT 1 mM). The reaction was run during 10 min at 37°C and stopped by the addition of 20 μL methanol. When indicated, the chromatography is performed on TLC plates (Thomas Scientific), with a mobile phase containing a mixture of lithium chlorure (LiCl) and of formic acid. After migration, a film is exposed on the plate and further developed. Radiolabelled Pi presence through ATPg32 hydrolysis is then revealed.

ATPase assay

The ATPase assay was performed using the colorimetric Pi ColorLock ALS kit (Innova Biosciences) in the presence of 5 μιηοΐ Lopl at 37°C. A595nm absorbance measurements were performed at t=2 min. The reactions occurred at 37°C during 2 min in a TrisHCl 50 mM pH=7.4 buffer containing 5 mM MgCl ₂ . The experimental data (Vmax, Km) were analyzed with the Michaelis-Menten equation, using a nonlinear regression analysis program (Kaleidagraph, Synergy Software) on three independent experiments performed as duplicate.

Western blot analysis

Western blot analyses were performed either on bacterial extracts or on purified proteins (polymerization and proteolysis assays).

Bacterial extracts were prepared as followed. For FtsZ and Lopl detection in K12 and K \ 2: :&lop] strains, overnight bacterial cultures were subcultured in 100 ml LB liquid media at 37°C until the O.D. A ₆ oo reached 0.3. Bacteria were harvested by centrifugation for 15 min at 3,000 x g, washed, then re- suspended in 10 ml PBS. Cells were spun again for 5 min at 3000 x g, and re- suspended in 10 mL of PBS. Membranes associated (Memb.) and cytosolic (Cyt.) proteins were separated by centrifugation for 20 min at 12,000 x g.

For FtsZ and Lopl detection in the KHAlopl strain upon Lopl-H ₆ and mutated versions overexpression (p/opi-H ₆ , plopl _K84A - („ plopl Ai. ₅₉ -R _b ) overnight bacterial cultures were subcultured in 800 ml LB liquid media at 37°C in the presence of IPTG at a final concentration of 1 mM for 3h. As indicated, bacteria were harvested by centrifugation for 15 min at 3,000 x g, washed, then subcultured in an equal volume of fresh LB liquid media. For each time point (0, 30 and 60 min), 100 mL of bacterial culture were harvested by centrifugation for 15 min at 3,000 x g, washed, then re-suspended in 10 mL PBS for sonication. Membranes associated (Memb.) and cytosolic (Cyt.) proteins were separated by centrifugation for 20 min at 12,000 x g. For FtsZ and Lopl detection in polymerization and proteolysis assays, 100 μί of the reaction mixture are loaded on each well.

Total protein concentrations were measured by the method of Bradford (Biorad). Proteins were separated by 16% SDS-PAGE and transferred to nitrocellulose membranes, and incubated with the primary antibodies diluted in PBS/5% milk/0.01% Tween20 (Sigma) overnight. Membranes were washed in PBS three times, then incubated with secondary antibodies for 1 hour before washing. Antibody binding was detected with chemiluminescence (ECL kit, GE Healthcare).

Electron-microscopy analysis

Bacteria. MG1655 E. coli wild-type and Alopl strains were observed by EM for immunodetection of Lopl . For immuno-EM, bacteria were fixed with 4% formaldehyde in 0.1 M phosphate buffer (pH 7.4), and embedded in 12% gelatin. Blocks were infiltrated with 2.3 M sucrose for cryoprotection, mounted on specimen holders and frozen in liquid nitrogen. Cryosections were performed with a Leica EM UC6/FC6 Microtome (Leica Microsystems, Vienna, Austria). A labeling was performed on thawed cryosections using antibody directed against Lopl, which is recognized by protein A gold. Cryosections were labeled first with an -Lopl rabbit polyclonal antibody at 1/1000 dilution, then with protein-A gold-10 nm diluted at 1/60 obtained from Utrecht University (Utrecht, The Netherlands) (Slot et al., 1991 ; West et al., 2005). The grids were viewed on a Jeol JEM 1010 (Japan) transmission electron microscope at 80 kV and Images were taken using a KeenView camera (Soft Imaging System, Lakewood, CO, USA) using ΪΤΕΜ5.0 software (Soft Imaging System GmbH).

Polymerized proteins. FtsZ and Lopl protein extracts were negatively stained with 2% uranyl acetate on glow discharged copper grids. The samples were observed in a Jeol 1200EXII or a JEM 1010 microscope (Jeol Company, Tokyo Japan) at 80-kV with an Eloise Keenview camera. Images were recorded with Analysis Pro Software version 3.1 (Eloise SARL, Roissy, France).

Bioinformatics

Lopl homologous proteins identification among other organisms was performed using BlastP. The Lopl sequence comparisons were performed using the ClustalW software. Bacteria length measurement was performed with the MicrobeTracker suite software (version 0.937) (Weiss et al., 1999; Sliusarenko et al., 201 1) and the data mining was performed using the Matlab computing system (R2012 version with Image Processing Toolbox and Statistics Toolbox). Statistical analyses were performed using the Graphpad Prism 5 software.

Fluorescent protein-fusions imaging

In order to localize FtsZ-GFP and Lo l-mCherry and mutated versions protein fusions in bacteria, the corresponding expression plasmids were transformed in E. coli K12 MG1655 wild-type or ALopl strains, as indicated. The localization was either performed on fixed or living bacteria. The fixation of bacteria was performed by adding 4% PFA followed by a washing in PBS. The observation was performed using a Nikon Eclipse 80i epifluorescent microscope. The live observation of FtsZ-GFP and Lopl-mCherry during the cell division process was performed on LB agar (1%) pad using a 200M Axiovert epifuorescent microscope (Zeiss) equipped with a Lambda LS 300W Xenon lamp and a CoolSnapHQ CCD camera.

Example 2 - RESULTS

Identification of lopl, essential for ileal loop colonization by

Shigella

The inventors identified lopl while screening a library of Shigella flexneri mutants by performing Signature Tagged Mutagenesis (STM, (Hensel et al., 1995; Raskin and de Boer, 1997),(Rothfield et al., 2005; West et al., 2005; de Boer, 2010) (Rothfield et al., 2005; de Boer, 2010; Marteyn et al., 2010)); a transposon insertion located 12 bp upstream of the predicted ORF was defective for gastrointestinal colonization and had a growth defect (not shown). The lopl gene is 1 128 bp in length and the predicted proteins in E. coli K12 (accession number b3232, yhcM) and Shigella M90T (accession number S3484) share 97.6% amino acid identity and putative Walker A and Walker B sites (Figures 1A and 8 A).

Lopl is conserved among Gram-negative bacteria (cocci and bacilli). In addition, homologues with up to 27% amino acid identity can be found among the eukaryotic kingdom such as fungi (C. albicans), yeast (S. cerevisiae) or human (H. sapiens) (Figure 1A and Table 3), which are predicted to be localised in mitochondria, although no experimental demonstration has yet been provided.

Loss of lopl leads to a temperature-dependent filamentous phenotype.

lopl mutants were constructed in E. coli K12 strain MG1655 (Κ12::Δ/ο/?7) and in S. flexneri strain M90T (M90T::Alopl), and complemented (K12::A/op7-p/o/?7-GFP and M90TA/o/?7-p/o/?7-GFP respectively). The S. flexneri lopl mutant was attenuated for GI colonization (Figure 8B) with a reduced tissue destruction compared with the wild-type strain (Figure 8C). Of note, M90T::Alopl had a filamentous phenotype in vivo (Figure 8D).

In vitro, the K\2::Alopl and Μ90Τ::Δ/ο/?7 mutants had a temperature-dependent elongated phenotype as compared to wild-type strains. At 42°C, average bacterial cell length increased significantly from 4.6±0.9 to 5.3±3 μιη in KUy.Alopl and 4.6±0.9to 93±18μηι in Μ90Τ: :Δ/ο 7 (Figures I B, 1 C, I D, Student's T test p<0.01 or pO.001) and was abolished by complementation of the mutants (Figure I B, 1 C and 9A). Interestingly, the conditional elongation phenotype could be complemented by eptotic expression of lopl-GFP in both strains suggesting that the GFP-fusion was functional (Figure I B and 1 C). The avirulent M90T strain TNY- Alopl ('virulence plasmid cured, congo-red negative strain) showed an intermediate filamentous phenotype as the mean length of bacteria was reduced as compared to Μ90Τ: :Δ/ορ7 at 42°C (18±3 μιη vs 93±18 pm, pO.001 , Figures I B and 9A). As a general statement, the temperature-dependent elongated phenotype was observed in all cases, comparing growth at 30°C and 42°C (Student's T test p<0.001) (Figures I D and 9A).

Lopl is an ATPase conserved among Gram-negative bacteria, which undergoes an ATP-dependent autoproteolysis.

To determine the biochemical function of Lopl , due to the presence of a putative nucleotide-binding site (GGVGRG _f uT). the inventors tested its ability to hydrolyse ATP and GTP. They first observed that Lopl is a monomeric protein by gel filtration (data not shown). The inventors demonstrated in vitro that purified K12 Lop l -Hi, hydrolyses ATP but not GTP (Figure ID). Point mutation of the putative Walter-A ATP-binding site (Figure 9A, GGVGRGKT) abolished ATPase activity (Figure 9B). Furthermore, in the presence of ATP, Lopl undergoes an ATP-dependent autolytic cleavage at amino acids L5 ₉ /W ₆ o confirmed by N-terminal sequencing analysis (Figure I E). Cleavage was not detected with ATP and additional EDTA, with AMP-PNP instead of ATP, or with Lopl 4A (Figure 9C). The inventors observed that Lop l Ai-59 is a monomeric protein in solution (data not shown). As Lop l Ai. ₅₉ is generated upon reaction of Lopl with ATP, the enzymatic parameters (Km, Vmax) of the full-length protein could not be calculated. Alternatively, the inventors could determine Lopl A], ₅₉ Km=0.25±0.1 mM and Vmax= 7.98 ± 0.87 mM Pi.min ^"1 (Figure 9D). Interestingly, the generation of Lopl Ai _-59 was associated with the assembly of protein polymers observed by negative stain electron-microscopy (Figure I F). In addition, Lopl A _{1 -59} alone was able to associate as polymers in the presence of ATP, indicating that the N-terminal fragment is not required for polymerization (Figure IF). In addition, incubating Lopl in the presence of ATP the inventors could isolate polymers by ultracentrifugation and demonstrate that they are composed of Lopl and LoplAi g, as Lopl only remained in the soluble fraction and Lop 1Δ 1 .59 accumulates in the pellet fraction (Figure 1G).

E. coli Lopl interacts with FtsZ in vitro.

To further define role of Lopl the inventors searched for potential interactions between Lopl and components of the divisome. Using the BATCH two- hybrid system (Karimova et al., 1998; 2005; Adams and Errington, 2009), they found interactions between Lopl and FtsQ, FtsL, FtsI and FtsN respectively (Student's T test p<0.01). No interaction was observed with FtsA, ZipA and FtsK. A similar result was obtained using Lopl from S. flexneri (M90T) as a bait (Figure 2A). Interactions with FtsZ could not be tested using the two-hybrid system, as previously described (Mukherjee and Lutkenhaus, 1994; Karimova et al., 2005).

The interactions were confirmed by pulldown. FtsQ-GFP (pDSW240), FtsL (pDSW326), FtsI-GFP (pDSW234) and FtsN-GFP (pDSW238) from E. coli lysates with and interact with Lopl bound to beads. Although no interaction was observed with ZipA (Figure 2B). An interaction between Lopl and FtsZ-GFP (pDSW230) was detected (Figure 2C). This interaction is dependent of the N-terminal region of Lopl but not its ATP-binding site (Figure 2C).

Next, the cellular localization of divisome components was analysed in the absence of Lopl ( \2::Alopl). Loss of Lopl resulted in multiple FtsZ-rings along filamentous bacteria (Figure 2D), while the localization of FtsQ, FtsL and FtsN was affected by the absence of Lopl (Figure 10), suggesting that Lopl recruitment was downstream of Z-ring maturation step. This phenotype was more marked during the growth in rich media (LB) and at higher temperatures (Figure 2E) or during stationary phase (Figure 1 1 A), consistent with the temperature-dependent phenotype seen in the K\2::Alopl mutant (Figure 1C and ID). Eventually, the filamentation of the Κ12::Δ/ορ7 strain expressing FtsZ-GFP (pDSW231) was observed using live fluorescent microscopy (Figure 1 IB).

FtsZ-GFP is not fully functional but does not impair the cell division process when expressed at a basal level, tolerated by K12 wild-type strain (Oliva et al., 2004; Thanedar and Margolin, 2004). However, the observation of dysfunctional Z-rings associated with the formation of filamentous bacteria in Kl2: Alopl expressing FtsZ-GFP suggests a role of Lopl in the control of Z-ring dynamics.

Next, as aberrant cell division phenotype was observed in the Κ12: :Δ/ορ7 mutant, the localisation of Lopl and FtsZ in the K12 wild-type and K12:A - lopl mutant strains was analysed by Western blot using rabbit polyclonal antibodies. The inventors demonstrated that Lop l was a cytosolic protein as FtsZ was found in cytosolic and membrane fractions (Figure 2F) as described previously in dividing bacteria (Shlomovitz and Gov, 2009) and modelled in liposomes (Osawa et al., 2009). The localisation of Lopl was further confirmed by EM analysis using a polyclonal anti-Lop 1 associated with an immunogold staining which allowed the detection of Lopl predominantly cytoplasmic, in the close vicinity of the bacterial inner membrane (Figure 2G), which is consistent with its ability to interact with FtsZ.

Lopl recruitment co-localises with constricting Z-rings in vivo. The inactivation of lopl leads to an accumulation of immature Z-rings.

In order to further examine the relationship between Lopl and FtsZ during cell division, the inventors performed timelapse observations with E. coli Κ12: :Δ/ορ>7 expressing Lopl -mCherry under the lopl promoter control (pSUC plasmid, Figures 12) and FtsZ-GFP (pDSW230) (Weiss et al., 1999) without IPTG, allowing a basal level of FtsZ-GFP expression. By live microscopy, the inventors demonstrated the recruitment of Lopl -mCherry at the Z-ring at late stage of cell division (Figure 3 A). This recruitment of Lopl -mCherry was correlated with a decrease of the Z-ring diameter (Figures 3B and 3C), while the Lopl -mCherry signal was co-localizing with the FtsZ-GFP signal at each time point (Figure 3B, n=5 represents five individual observations extracted from three independent experiments).

Considering the maximum of Z-ring constriction as a final state, these quantifications were performed at -50min, -15min, -l Omin, -5min. As compared to the -50min time point (before Maximum constriction), the level of the Lopl - mCherry signal and the diameter of the Z-ring were inversely correlated at the -15min, -l Omin, -5min time points and at the Maximal constriction (Max. const.) (Figure 3C, Student's T test p<0.001 ). In the absence of Lop l-mCherry, no restriction of the Z- ring was observed (Figures 3D and 1 IB). Lopl lacking residues 1 -59 was not associated with the Z-ring (Figure 3E), which was consistent with the inability of LoplAi.59 to interact with FtsZ in vitro (Figure 2C).

Overexpression of Lopl induces bacterial filamentation and Z- ring disruption in vivo. Lopl is processed in vivo into LoplA _{1 -5} 9.

Different versions of Lopl were constitutively expressed to further study the influence of this protein on Z-ring stability and constriction.

Overexpression of the Lopl protein induced the formation of twisted filamentous bacteria (Figure 4A and 4B), while overexpressing Lopl _K 84A and Lopl Ai. 59 had no consequence on the bacterial shape (Figure 4B). Overexpression of the Lopl 59aa N-terminal fragment only (Lop l ,_s<r6xHis) was toxic for the bacteria (data not shown). Reducing the IPTG dose to 0.1 mM induced the formation of filamentous bacteria, suggesting that this fragment interfered with FtsZ (Figure 4B).

In order to evaluate the level of Lopl turnover, overexpression of Lopl was performed at a lower level and stopped by eliminating IPTG form the media (t=0 min). At t=0 the inventors could recapitulate the results described in Figure 4B. The inventors could then observe at t=30 min and t=60 min that the filamentous phenotype associated to Lopl overexpression was reversed. In controls (Lopl 4A and LoplAi_59) the bacterial shape remained normal as compared to the wild-type at all time points (Figure 4C). As observed previously in vitro, the ATP-dependent cleavage of Lopl (Figure IE), was recapitulated by Western blot using an anti-His antibody. Indeed, LoplA]. ₅₉ was detected in the pellet fraction upon Lopl overexpression at t=60 min (Figure 4D), which was not observed upon Lopl i<84A overexpression (Figure 4D). These observations were confirmed using an anti-Lopl antibody (Figures 4E). Interestingly, overexpression of Lopl-6xHis contributed to accumulation of an abnormally high level of FtsZ in the pellet fraction, while overexpressing LoplAi_5 ₉ caused a reduced amount of FtsZ polymers in the pellet fraction (Figure 4E).

As Lopl overexpression seemed to perturb the Z-ring constriction, the inventors aimed at visualising FtsZ polymerization in this condition. The FtsZ- GFP protein fusion was well tolerated by the E. coli wild-type strain, even though the fluorescent signal was low (Figure 5 A), while it led to the formation of filamentous bacteria in the Κ\ 2 Δΐορ1 mutant (Figure 5A), as previously observed (Figure 2D). Subsequently, the overexpression of Lopl-mCherry perturbed Z-ring constriction in a dose-dependent manner (Figure 5B), while the overexpression of Lop l K. ₈₄ A-mCherry or LoplAi-sg-mCherry did not lead to either bacterial shape modification (Figure 5B). For unclear reasons, in the presence of pJC104, the Lopl- mCherry signal was undetectable (data not shown) as compared to the wild-type strain in the absence of pJC104 (Figures 5B and 13).

Fluorescence-based assay of FtsZ proteolysis

In order to confirm the FtsZ polymer proteolysis by LoplAi.59, the inventors designed a fluorescence-based assay as polymerization of a FtsZ/FtsZ-GFP mixture leads to an increase of the solution fluorescence (Trusca and Bramhill, 2002). In the presence of GTP, purified FtsZ and FtsZ-GFP form polymers, as described previously (Yu and Margolin, 1997b); however the level of the detected fluorescence remains low and the polymers length was reduced (Figures 17A and 17B uppermost row). In the presence of LoplAi^, the inventors observed a rapid and significant increase of the fluorescence level (Figure 17A), which correlates with the formation of an homogeneous FtsZ/FtsZ-GFP polymer network (Figure 17B middle row, t=0). Interestingly, these polymers were rapidly degraded (Figure 17B middle row, t^lOmin), which correlated with a significant decrease of the fluorescence level (Figure 5C, Student's T test p<0.01). As a control, the simultaneous addition of benzamidin with LoplAi^ did not impair the fluorescence increase (Figure 17B lowermost row), although preventing the FtsZ/FtsZ-GFP polymers degradation (Figure lowermost row), in association with a stable level of the fluorescent signal (Figure 17A). Interestingly, the addition of full length Lopl led to a reduced fluorescence increase (Figure 17A), led to the formation of large and bundled polymers of FtsZ/FtsZ-GFP, which remained stable over time (Figure 17A).

LoplAi.sg proteolyses FtsZ polymers in vitro.

To determine whether Lopl acts directly on FtsZ polymers, the inventors performed FtsZ polymerization assays in the presence of purified proteins (Lopl , Lopl _84A and LoplA] -5 ); the soluble and polymerized forms of FtsZ were separated by ultracentrifugation. FtsZ formed polymers in a GTP-dependent manner in 3 min at 30°C (Figure 14). The FtsZ polymers incubation with Lopl lead to their slight degradation at t=3min in an ATP-dependent manner (Figure 6A), with no obvious increase in soluble FtsZ level suggesting a proteolysis of FtsZ by Lopl . In contrast, no degradation of FtsZ polymers was observed with Lopl _K84A (Figure 6 A). Strikingly, incubation of FtsZ polymers with LoplAj.sg in the absence of ATP caused its significant degradation (t=lmin, no FtsZ signal in the soluble fraction), in association with a complete degradation of LoplAi_59 (Figure 6A). In turn, the inventors hypothesize that the late initiation of FtsZ polymers proteolysis by Lopl in the presence of ATP might be due to the formation of small amount of LoplA] _59-6xHis, as described above.

This result was confirmed by TEM analysis of FtsZ polymers (Yu and Margolin, 1997a; Adams and Errington, 2009) (Mukherjee and Lutkenhaus, 1994; Mateos-Gil et al., 2012), as the simultaneous addition of Lopl with ATP or Lopl Ai. ₅₉ in the absence of ATP induced a disruption or complete degradation of FtsZ polymers respectively (Figure 6B). Conversely, no effect was observed in the presence of Lopl with ATP and additional EDTA or LoplAi. ₅ 9 with PMSF (Figure 6B). This result was consistent with the FtsZ polymers degradation activity of Lopl-A]^ and confirmed the ATP-dependent proteolysis of FtsZ by Lopl and indirectly the ATP-dependent autoproteolysis of Lopl (Figure 2F). This experiment showing an inhibition of the proteolytic activity of LoplAi.59 by PMSF suggested that this protein has a serine protease activity.

In a strain expressing Lopl-mCherry and FtsZ-GFP, the inventors observed that a small population of bacteria (1.5%(±2), n=234) expressed a strong Lopl -mCherry signal (Figure 6C), as most of the bacterial population was associated with a Z-ring (89%(±7), n=234) (Figure 6C). In the whole population, the Lopl- mCherry signal and Z-ring detection were found to be exclusive (Student's T test pO.001 , n=232). In contrast, Lopl _K84A -mCherry or LoplAi^-mCherry positive cells showed no such effect on Z-rings (Figure 15). Focusing on Lopl-mCherry positive cells for up to 120 min of growth, no Z-ring could be detected and over the time, as they did not divide (Figure 6D). These observations support the previous results showing a Lo l -dependent degradation of Z-ring in E. coli. The inventors demonstrated that LoplAi.sg proteolytic activity was independent from the ClpX/ClpP complex which was recently described as being involved in FtsZ proteolysis in vitro (Camberg et al., 2009) (Oliva et al., 2004; Sugimoto et al., 2010) (Romberg and Mitchison, 2004; Camberg et al., 201 1 ). Briefly, through a gel-filtration analysis, the inventors demonstrated that Lopl in the presence of AMP-PNP was not forming complex with ClpP in vitro (Figure 16A and 16B). As a confirmation of this result, Lopl is a monomeric protein in vitro, in contrast ClpX and ClpP organised as a proteolytic hexameric complex in vitro (Mukherjee and Lutkenhaus, 1998; Maillard et al., 201 1).

As a final experiment, in order to confirm that LoplAi. ₅₉ was a serine protease, the inventors repeated the FtsZ polymers proteolysis assay in the presence of Lop l Λ 1. and various protease inhibitors. The inventors found that serine protease inhibitors (PMSF and leupeptin) inhibited the proteolysis of FtsZ polymers, whereas the the presence of EDTA or pepstatin had no effect (Figure 6E). Indeed the inventors found that the 73 aa C-terminal fragment of LoplAi_ ₅ 9 was required for the proteolytic degradation of FtsZ polymers (Figure 6F).

DISCUSSION

In this work, the inventors have characterized Loopinl (Lopl ), an ATPase conserved among Gram-negative bacteria. Lopl was named in relation to its function as the inventors present evidence that this protein has a "looping effect" on the Z-ring, allowing its constriction and the cell division process to end (Figure 3A).

Lopl was not an obvious candidate to play a role in the cell division control. First, lopl is neither an essential gene in E. coli nor in Shigella, so the initial screens aiming at identifying essential cell division genes missed lopl . Second, when fluorescent fusions (mCherry or GFP) of Lopl are expressed in E. coli, no recruitment at the Z-ring is observed (Figures 6C and 6D), as classically reported for most of the divisome complex components (Strieker et al., 2002; de Boer, 2010). The inventors could observe a dynamic recruitment of Lopl -mCherry at the Z-ring using recent imaging technologies such as live microscopy (Figure 3 A and Movie S I). To further Lopl function, the inventors demonstrated that it autoproteolyses in an ATP- dependent manner, generating a N-terminal truncation (1 -59) leading to the production of Lopl A _{] -59} (or tLopl , Figure 7) (Figure IE). LOPI A _S Q is a serine protease (Figure 6E) and its C-terminal part (between aa 303-375) is required for its activity (Figure 6F).

LoplAi-59 is an active serine protease generated from Lopl through an ATP-dependent autoproteolytic process

The inventors identified FtsZ as a target of Lopl and shown that the

N-terminal fragment of Lopl is not essential for its ATP-dependent polymerization (Figure I F), but is required for the interaction between Lopl and FtsZ, which has been observed in vitro using an His-pulldown approach (Figure 2B). This result was confirmed using a live microscopy approach, allowing the visualisation of Lopl recruitment during the constriction of the Z-ring (Figure 3 A). The inventors showed that Lopl recruitment occurs downstream of the Z-ring maturation, as the localization of FtsQ, FtsL and FtsN are not impaired in the absence of Lopl (Figure 1 1A). Indeed, the inventors have demonstrated that the accumulation of Lopl at the Z-ring correlated to its constriction (Figure 4B, 4C and 4D). As the results show in vitro that the LoplAi.sg truncated form proteolyses FtsZ polymers (Figure 6B), the inventors speculate that this activated form of Lopl might be generated upon interaction with FtsZ during the cell division (Figure 7). This hypothesis is supported by the observation that generation of LoplAi _-5 9 is required for the formation of Lopl /Lop 1Δ _{1 -5} 9 mixture composed filaments (Figure IF), which are observed during the cell-division process (Figure 3A).

The ATP-dependent autoproteolytic activation of Lopl is comparable to the subtilisin (serine protease) autoproteolytic maturation consisting in a 77 aa prodomain processing (Bryan et al, 1995). In analogy with the subtilisin catalytic triade (Asp-32, His-64, and Ser-221), the Lopl active site will have to be characterized in details through a mutagenesis approach and structural analysis.

Until this work, the regulation mechanism by which a Z-ring will initiate its constriction has been a controversial question. A first model suggested that FtsZ polymers could mediate their own constriction through GTP hydrolysis leading to their depolymerization (Scheffers and Driessen, 2001). Another model suggested that MAP-like proteins to be identified could be recruited at the Z-ring to modulate FtsZ polymers stability, bundling and disassembly (Adams and Errington, 2009). Only recently, the hypothesis of a proteolysis-dependent control of the Z-ring constriction emerged.

Z-ring proteolysis is associated with its constriction

It has been reported previously in Caulobacter crescentus that the rate of FtsZ degradation increases after the initiation of the cell division, leading to a decrease of the Z-ring diameter (Kelly et al., 1998) however the underlying mechanism was not described. This proteolytic model of FtsZ constriction was supported by recent observations describing the proteolytic degradation of FtsZ polymers by the ClpX (ATPase)/ClpP (protease) hexameric complex (Hensel et al., 1995; Camberg et al., 2009) (West et al., 2005; Sugimoto et al., 2010) (Camberg et al., 201 1), although no direct recruitment at the Z-ring of this complex was observed during a cell division process. Interestingly, the Lopl N-terminal fragment has a role in the FtsZ interaction (Figures 2B and 3E) which can be compared to the N-terminal (65aa) domains of ClpX, which is also involved in the recognition of its substrate. The overexpression of the ClpX N-terminal fragment (but also the full length protein) cause filamentation and perturbates the Z-ring constriction (Karimova et al., 1998; 2005; Glynn et al., 2009) (Karimova et al., 2005; Sugimoto et al., 2010), as observed for Lopl (Figure 4B). Indeed, in the absence of the N-terminal 59aa, LoplAi^ did not interact with FtsZ (Figure 2B). Its overexpression no longer perturbates any longer the FtsZ polymerization (Figures 4B).

As lopl is not an essential gene in E. coli or in Shigella, it is still unclear to which the extent of Lopl redundancy as the ClpX/ClpP complex and putatively other proteases provide FtsZ depredatory functions. This hypothesis would be supported by the vital role played by the Z-ring constriction control in the cell division process and survival of bacteria. Indeed, to date five AAA+-containing proteolytic systems have been identified in bacteria in general and more particularly in E. coli that are ClpX/P, ClpA/P, HslU/V, FtsH and Lon (reviewed in (Thanedar and Margolin, 2004; Hanson and Whiteheart, 2005)), which could putatively be involved in Z-ring proteolysis, as demonstrated for ClpX/ClpP in vitro. However, similarly to our observations concerning Lopl , their speculative dynamic recruitment at the Z-ring during the late stage of bacteria division will have to be demonstrated. This hypothesis is particularly true considering Gram-positive bacteria in which cell division process is controlled by FtsZ (as reviewed in (Errington et al., 2003; Shlomovitz and Gov, 2009) and (Goehring and Beckwith, 2005; Osawa et al., 2009)) but do not possess the lopl gene. Further study will deepen our knowledge on Z-ring constriction modulation in bacteria.

Lopl abundance regulation is critical for Z-ring constriction

Lopl abundance regulation seems to be important for Z-ring constriction as the inventors show that the absence (Figure 1 IB) or the overexpression (Figure 4B) of lopl alters the cell division process through Z-ring constriction. In addition, our live microscopy allowed the detection of Lopl cyclic accumulation at the Z-ring during each division (Figure, 3A). Based on these observations, the inventors propose that Lopl expression and degradation have to be tightly and cyclically controlled by the cell to allow functional Z-ring constriction and daughter cells separation processes. Our results show that Lopl autoproteolyses in an ATP- dependent manner (Figure IE) and that the FtsZ proteolysis by Lopl A ₁ -59 is associated with a degradation of the latter (Figure 6A). Our data suggest that the autoproteolysis of LoplAi-59 upon activation might be considered as a self-regulation of its proteolytic action. The question of a cyclic expression regulation of lopl is still unsolved. Addressing this question will provide information on the cell cycle control as it could be shown in the case of KiaC in the circadian clock control of cell division in cyanobacteria (Weiss et al., 1999; Dong et al, 2010).

The inventors propose that Lopl plays a key role in the cell division process control through the proteolysis of FtsZ polymers. As the inactivation of lopl led to the loss of Shigella virulence in vivo, this protein should have to be considered as a novel putative antibiotic target for Gram-negative bacterial infection. Since the initial identification of the filamentous temperature sensitive fts) genes and as highlighted by Beckwith and colleagues a decade ago (Buddelmeijer and Beckwith, 2002), the characterization of the whole set of proteins involved in the Z-ring- dependent bacterial division is still on going. It will be essential for a better comprehension of this key vital biological process.

REFERENCES

Adams, D.W., and Errington, J. (2009). Bacterial cell division: assembly, maintenance and disassembly of the Z ring. Nat. Rev. Microbiol. 7, 642- 653.

Benaroudj, N. et al., 201 1. Assembly and proteolytic processing of mycobacterial ClpPl and ClpP2. BMC biochemistry, 12, p.61.

Bi, E.F., and Lutkenhaus, J. (1991). FtsZ ring structure associated with division in Escherichia coli. Nature 354, 161-164.

Bryan, P., Wang, L., Hoskins, J., Ruvinov, S., Strausberg, S., Alexander, P., Almog, O., Gilliland, G., and Gallagher, T. (1995). Catalysis of a protein folding reaction: mechanistic implications of the 2.0 A structure of the subtilisin-prodomain complex. Biochemistry 34, 10310-10318.

Buddelmeijer, N., and Beckwith, J. (2002). Assembly of cell division proteins at the E. coli cell center. Curr. Opin. Microbiol. 5, 553-557.

Camberg, J.L., Hoskins, J.R., and Wickner, S. (2009). ClpXP protease degrades the cytoskeletal protein, FtsZ, and modulates FtsZ polymer dynamics. Proc. Natl. Acad. Sci. U.S.a. 106, 10614-10619.

Camberg, J.L., Hoskins, J.R., and Wickner, S. (201 1). The interplay of ClpXP with the cell division machinery in Escherichia coli. J. Bacteriol. 193, 191 1 - 1918.

Chen, J.C., Weiss, D.S., Ghigo, J.M., and Beckwith, J. (1999). Septal localization of FtsQ, an essential cell division protein in Escherichia coli. J. Bacteriol. 181, 521-530.

de Boer, P. A.. I. (2010). Advances in understanding E. coli cell fission. Curr. Opin. Microbiol. 13, 730-73 '.

Datsenko, .A., and Wanner, B.L. (2000). One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc. Natl. Acad. Sci. U.S.a. 97, 6640-6645.

Dong, G., Yang, Q., Wang, Q., Kim, Y.-I., Wood, T.L., Osteryoung, K.W., van Oudenaarden, A., and Golden, S.S. (2010). Elevated ATPase activity of KaiC applies a circadian checkpoint on cell division in Synechococcus elongatus. Cell 140, 529-539.

Errington, J., Daniel, R.A., and Scheffers, D.-J. (2003). Cytokinesis in bacteria. Microbiol. Mol. Biol. Rev. 67, 52-65-tableofcontents.

Galli, E., and Gerdes, K. (2010). Spatial resolution of two bacterial cell division proteins: ZapA recruits ZapB to the inner face of the Z-ring. Mol.

Microbiol. 76, 1514-1526.

Ghigo, J.M., Weiss, D.S., Chen, J.C., Yarrow, J.C., and Beckwith, J. (1999). Localization of FtsL to the Escherichia coli septal ring. Mol. Microbiol. 31, 725-737.

Glynn, S.E., Martin, A., Nager, A.R., Baker, T.A., and Sauer, R.T. (2009). Structures of asymmetric ClpX hexamers reveal nucleoti de-dependent motions in a AAA+ protein-unfolding machine. Cell 139, 744-756.

Goehring, N.W., and Beckwith, J. (2005). Diverse paths to midcell: assembly of the bacterial cell division machinery. Curr. Biol. 15, R514-R526.

Hanson, P.I., and Whiteheart, S.W. (2005). AAA+ proteins: have engine, will work. Nat. Rev. Mol. Cell Biol. 6, 519-529.

Hensel, M., Shea, J.E., Gleeson, C, Jones, M.D., Dalton, E., and Holden, D.W. (1995). Simultaneous identification of bacterial virulence genes by negative selection. Science 269, 400-403.

Hu, Z... Mukherjee, A., Pichoff, S., and Lutkenhaus, J. (1999). The MinC component of the division site selection system in Escherichia coli interacts with FtsZ to prevent polymerization. Proc. Natl. Acad. Sci. U.S. a. 96, 14819-14824.

Karimova, G., Dautin, N., and Ladant, D. (2005). Interaction network among Escherichia coli membrane proteins involved in cell division as revealed by bacterial two-hybrid analysis. J. Bacteriol. 187, 2233-2243.

Karimova, G., Pidoux, J., Ullmann, A., and Ladant, D. (1998). A bacterial two-hybrid system based on a reconstituted signal transduction pathway. Proc. Natl. Acad. Sci. U.S.a. 95, 5752-5756.

Kelly, A.J., Sackett, M.J., Din, N., Quardokus, E., and Brun, Y.V.

(1998). Cell cycle-dependent transcriptional and proteolytic regulation of FtsZ in Caulobacter. Genes Dev. 12, 880-893.

Mai Hard. R.A., Chistol, G., Sen, M., Righini, M., Tan, J., Kaiser, C.M., Hodges, C., Martin, A., and Bustamante, C. (201 1). ClpX(P) generates mechanical force to unfold and translocate its protein substrates. Cell 145, 459-469.

Marteyn, B., West, N.P., Browning, D.F., Cole, J.A., Shaw, J.G., Palm, F., Mounier, J., Prevost, M.C., Sansonetti, P., and Tang, CM. (2010). Modulation of Shigella virulence in response to available oxygen in vivo. Nature 465, 355-358.

Martinez, E., Bartolome, B., and la Cruz, de, F. (1988). pACYC 184- derived cloning vectors containing the multiple cloning site and lacZ alpha reporter gene of pUC8/9 and pUC 18/19 plasmids. Gene 68, 159-162.

Mateos-Gil, P., Paez, A., l lorger, I., Rivas, G., Vicente, M., Tarazona, P., and Velez, M. (2012). Depolymerization dynamics of individual filaments of bacterial cytoskeletal protein FtsZ. Proc. Natl. Acad. Sci. U.S. a. 109, 8133-8138.

Mukherjee, A., and Lutkenhaus, J. (1994). Guanine rmcleotide- dependent assembly of FtsZ into filaments. J. Bacteriol. 176, 2754-2758.

Mukherjee, A., and Lutkenhaus, J. (1998). Dynamic assembly of FtsZ regulated by GTP hydrolysis. Embo J. 17, 462^69.

Mukherjee, A., Cao, C, and Lutkenhaus, J. (1998). Inhibition of FtsZ polymerization by SulA, an inhibitor of septation in Escherichia coli. Proc, Natl. Acad. Sci. U.S.a. 95, 2885-2890.

Mukherjee, A., Saez, C, and Lutkenhaus, J. (2001). Assembly of an FtsZ mutant deficient in GTPase activity has implications for FtsZ assembly and the role of the Z ring in cell division. J. Bacteriol. 183, 7190-7197.

Oliva, M.A., Cordell, S.C., and Lowe, J. (2004). Structural insights into FtsZ protofilament formation. Nat. Struct. Mol. Biol. 11, 1243-1250.

Osawa, M., Anderson, D.E., and Erickson, I LP. (2009). Curved FtsZ protofilaments generate bending forces on liposome membranes. Embo J. 28, 3476- 3484.

Pazos, M., Natale, P., and Vicente, M. (2012). A specific role for the

ZipA protein in cell division: stabilization of the FtsZ protein. J. Biol. Chem.

Raskin, D.M., and de Boer, P.A. (1997). The MinE ring: an FtsZ- independent cell structure required for selection of the correct division site in E. coli. Cell 91, 685-694.

Romberg, L., and Mitchison, T.J. (2004). Rate-limiting guanosine

5 '-triphosphate hydrolysis during nucleotide turnover by FtsZ, a prokaryotic tubulin homologue involved in bacterial cell division. Biochemistry 43, 282-288. Rothfield, L., Taghbalout, A., and Shih, Y.-L. (2005). Spatial control of bacterial division-site placement. Nat. Rev. Microbiol. 3, 959-968.

Sansonetti, P.J., d'Hauteville. H.. Formal, S.B., and Toucas, M. (1982). Plasmid-mediated invasiveness of "Shigella-like" Escherichia coli. Ann.

Microbiol. (Paris) 133. 351 -355.

Scheffers, D., and Driessen, A.J. (2001). The polymerization mechanism of the bacterial cell division protein FtsZ. FEBS Lett. 506, 6-10.

Shlomovitz, R., and Gov, N.S. (2009). Membrane-mediated interactions drive the condensation and coalescence of FtsZ rings. Phys Biol 6, 046017.

Sliusarenko, O., Heinritz, J., Emonet, T., and Jacobs-Wagner, C. (201 1). High-throughput, subpixel precision analysis of bacterial morphogenesis and intracellular spatio-temporal dynamics. Mol. Microbiol. 80, 612-627.

Slot, J.W., Geuze, H.J., Gigengack, S., Lienhard, G.E., and James, D.E. (1991). Immuno-localization of the insulin regulatable glucose transporter in brown adipose tissue of the rat. J Cell Biol 113, 123-135.

Strieker, J., Maddox, P., Salmon, E.D., and Erickson, H.P. (2002). Rapid assembly dynamics of the Escherichia coli FtsZ-ring demonstrated by fluorescence recovery after photobleaching. Proc. Natl. Acad. Sci. U.S. a. 99, 3171— 3175.

Sugimoto, S., Yamanaka, K., Nishikori, S., Miyagi, A., Ando, T., and Ogura, T. (2010). AAA+ chaperone ClpX regulates dynamics of prokaryotic cytoskeletal protein FtsZ. J. Biol. Chem. 285, 6648-6657.

Thanedar, S., and Margolin, W. (2004). FtsZ exhibits rapid movement and oscillation waves in helix-like patterns in Escherichia coli. Curr. Biol. 14, 1 167-1 173.

Trusca, D., Scott, S., Thompson, C, and Bramhill, D. (1998). Bacterial SOS checkpoint protein SulA inhibits polymerization of purified FtsZ cell division protein. J. Bacteriol. 180, 3946-3953.

Twining, S.S. Fluorescein Isothiocyanate Labeled Casein Assay for

Proteolytic Enzymes Anal. Biochem. 143, 30-34(1984).

Valdivia, R.H., and Falkow, S. (1996). Bacterial genetics by flow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by differential fluorescence induction. Mol. Microbiol. 22, 367-378.

Weiss, D.S., Chen, J.C., Ghigo, J.M., Boyd, D., and Beckwith, J. (1999). Localization of FtsI (PBP3) to the septal ring requires its membrane anchor, the Z ring, FtsA, FtsQ, and FtsL. J. Bacteriol. 757, 508-520.

West, N.P., Sansonetti, P., Mounier, J., Exley, R.M., Parsot, C, Guadagnini, S., Prevost, M.C., Prochnicka-Chalufour, A., Delepierre, M, Tanguy, M., et al. (2005). Optimization of virulence functions through glucosylation of Shigella LPS. Science 307, 1313-1317.

Yu, X.C., and Margolin, W. (1997a). Ca2+-mediated GTP- dependent dynamic assembly of bacterial cell division protein FtsZ into asters and polymer networks in vitro. Embo J. 16, 5455-5463.

Yu, X.C., and Margolin, W. (1997b). Ca2+-mediated GTP- dependent dynamic assembly of bacterial cell division protein FtsZ into asters and polymer networks in vitro. Embo J. 16, 5455-5463.