- Email: [email protected]
Polar positional information in Escherichia coli spherical cells
Biochemical and Biophysical Research Communications 353 (2007) 493–500 www.elsevier.com/locate/ybbrc
Polar positional information in Escherichia coli spherical cells Nathalie Pradel
a,1
, Claire-Lise Santini a,1, Alain Bernadac b, Yu-Ling Shih Marcia B. Goldberg d, Long-Fei Wu a,*
c,2
,
a
c
Laboratoire de Chimie Bacte´rienne, UPR9043, Institut de Biologie Structurale et Microbiologie, CNRS, 31 chemin Joseph Aiguier, F-13402 Marseille cedex 20, France b Service de Microscopie Electronique, Institut de Biologie Structurale et Microbiologie, CNRS, 31 chemin Joseph Aiguier, F-13402 Marseille cedex 20, France Department of Molecular, Microbial, and Structural Biology, University of Connecticut Health Center, Farmington, CT 06032, USA d Bacterial Pathogenesis Laboratories, Infectious Disease Division, Massachusetts General Hospital, Cambridge, MA 02139, USA Received 5 December 2006 Available online 18 December 2006
Abstract Shigella surface protein IcsA and its cytoplasmic derivatives are localized to the old pole of rod-shaped cells when expressed in Escherichia coli. In spherical mreB cells, IcsA is targeted to ectopic sites and close to one extremity of actin-like MamK filament. To gain insight into the properties of the sites containing polar material, we studied the IcsA localization in spherical cells. GFP was exported into the periplasm via the Tat pathway and used as a periplasmic space marker. GFP displayed zonal fluorescence in both mreB and rodA-pbpA spherical E. coli cells, indicating an uneven periplasmic space. Deconvolution images revealed that the cytoplasmic IcsA fused to mCherry was localized outside or at the edge of the GFP zones. These observations strongly suggest that polar material is restricted to the positions where the periplasm possesses particular structural or biochemical properties. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Periplasm; Polar localization information; Spherical cells; MreB; GFP; mCherry
Polarity is an intrinsic property of cells. Many different bacterial species utilize polar localization for diverse cellular processes including chromosome segregation, protein secretion, macromolecule delivery, chemotaxis, adherence, motility, cell division, cell shape and virulence [1]. The Shigella outer membrane protein IcsA (VirG) is one of the most extensively studied polar proteins. It localizes to the pole of Shigella, where it mediates assembly of the actin tail and movement of this intracellular pathogen within the cytoplasm of a host cell [2]. IcsA is targeted and exported at the bacterial pole [3]. Each of two segments, residues 1–104 (IcsA1–104) or 507–620 (IcsA507–620), is sufficient for *
Corresponding author. Fax: +33 491718914. E-mail address: [email protected] (L.-F. Wu). 1 These authors contributed equally to this work. 2 Present address: Institute of Biological Chemistry, Academia Sinica, 128 Sec. 2, Academia Road, Nankang, Taipei 115, Taiwan. 0006-291X/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.12.054
targeting IcsA to the pole on the cytoplasmic side of the inner membrane of rod-shaped cells [3,4]. The bacterial cytoskeletal protein MreB forms helical structures along the long axis of the cell, just beneath the cytoplasmic membrane (for review see [5]). Interestingly, the mreB mutation leads to ectopic localization of IcsA and its truncated IcsA1–104 and IcsA507–620 derivatives in the spherical cells [6,7]. Janakiraman and Goldberg have shown that polar positional information required for IcsA localization is independent of both known components of the cell division machinery and nucleoid occlusion [8]. Recently, two laboratories have independently identified a landmark protein, TipN, which acts as a spatial and temporal cue for setting up the correct polarity in the bacterium Caulobacter crescentus [9,10]. In contrast to IcsA, TipN does not localize to the division site in cells depleted of the division initiation protein FtsZ [9,10] or in cells in which the division site peptidoglycan synthesis protein FtsI has been inactivated [9].
494
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
These observations would suggest that different mechanisms might be used for polar localization of IcsA and TipN. However, the polar positional information that mediates proper localization of the polar proteins and the mechanisms by which polar proteins recognize this positional information are poorly understood [1]. MamK actin proteins are found only in magnetotactic bacteria and involved in positioning bacterial organelles, magnetosomes [11,12]. When expressed in Escherichia coli, MamK appears as a linear filament which is structurally and functionally distinct from the MreB filament [13]. Interestingly, one extremity of the MamK filament is located close to the co-expressed IcsA in spherical mreB mutant cells [13], suggesting that the filament may not be randomly positioned within a geometrically symmetric spherical cell. In E. coli, like the mreB mutation, rodA-pbpA mutations affect periplasmic sidewall peptidoglycan synthesis and lead to spherical cells [5]. The cellular localization of penicillinbinding protein 2 (encoded by pbpA) depends upon the localization of both MreB and MreC in C. crescentus [14–16]. To gain more information regarding the polar material localization information in spherical cells, we compared the localization of IcsA in mreB and rodA-pbpA spherical cells. Green fluorescence protein (GFP) was exported into the periplasm via the twin-arginine translocation pathway and used as a periplasmic space marker under physiological conditions. Periplasmic GFP exhibited zonal localization in both mutants. Interestingly, IcsA507–620mCherry was located outside or at the edge of the GFP zones, but not overlaid on the GFP zones. These results would suggest that polar positional information recognized by IcsA resides at positions where the periplasm possesses particular structural or biochemical properties.
Materials and methods Bacterial strains, plasmids, and media. Escherichia coli strains used in this study are: MC1000 (araD139 (ara, leu)7697 lacX74 galU galK rpsL) and its derivative YLS3 (mreB) [6], GC3904 (argS D(pbpA-rodA)::Kan, zbf::Tn10 [17] and TG1 (D(lac-pro) supE thi hsdD5/F 0 traD36 proA+B+ lacIq lacZDM15). Plasmids pBAD-IcsA1–104-GFP, pBAD-IcsA507–620-GFP (=pMPR402), pBAD-IcsAD507–730-GFP and Ptac-IcsA507–620-mCherry (pAWY3) express hybrid proteins consisting of various segments of IcsA fused in frame to green fluorescent protein or mCherry under the control of the arabinose promoter of pBAD24 or the tac promoter [7]. To improve signal/noise ratio of periplasmic GFP fluorescence, the gfpmut2 in our previously described pRR-GFP plasmid [17] was changed with the gene encoding pH sensor derivative of GFP, the ratiometric pHluorin [18]. Compared to the wild type GFP, pHluorin displays a reversible excitation ratio change between pH 7.5 and 5.5; the major excitation peak is at 475 nm under acidic conditions [18]. Since the periplasm is more acidic than the cytoplasm due to the proton motive force, ratiometric pHluorin increases the signal/background ratio comparing to other green fluorescent protein derivatives when the same portion of the marker proteins are exported into the periplasm. The DNA fragment encoding for ratiometric pHluorin was amplified by PCR using PHR5NHE (5 0 -cgg tgc tag caa agg aga aga act ttt cac-3 0 ) and PHR3H3 (5 0 -gaa tta agc tta ttt gta tag ttc atc cat gcc-3 0 ) as primers, and the reaction was performed by using the Expand High Fidelity PCR System according to the manufacturer’s instructions (Roche). The amplified fragment was
purified, double digested by NheI and HindIII and cloned into the corresponding sites of the plasmid pRR-GFP, resulting to plasmid p8799. The bacteria were routinely grown in Luria–Bertani (LB) medium or on LB plates [19]. As required, ampicillin (Amp) (100 lg/ml), chloramphenicol (30 lg/ml), glucose (0.2% w/v), arabinose (0.2% w/v) or IPTG (0.5 mM) were added. Microscopy and fluorescence spectrometry. Overnight cultures were diluted 1:100 for wild type strain and 1:50 for the mutants in LB+Amp+glucose medium and incubated at 37 °C with shaking for 3 h. Cells were centrifuged, washed once with LB+Amp medium, and resuspended in LB+Amp with 0.2% arabinose or/and IPTG (0.5 mM), and grown at 20 °C for 80 min with shaking. The cells were examined directly, or after the addition of 0.15 M NaCl or other treatment as described in the text. Protein synthesis was blocked by the addition of rifampicin (0.15 lg/ml) to the culture. Samples were taken 1 h later and examined under fluorescence microscope. Cells were fixed in 0.25% agarose on slides. Images and Z-stack of 25–33 images were captured with step distance ranging from 0.15 or 0.25 lm with Zeiss Axiovert 200M connected with a Hamamatsu ORCA ER camera. Image restoration was obtained by deconvolution using Huygens Essential software (SVI). Three-dimensional visualization was performed with Imaris software package (Bitplane). Immuno-gold staining of ultra-thin frozen sections were performed by using 7 nm gold-conjugated protein A as described by Anba et al. [20]. Antisera used are polyclonal rabbit anti GFP [17] or RFP (Chemicon International, Temecula, California, USA) at dilution of 1/500 and 1/100, respectively.
Results and discussion Localization of polar IcsA507–620-GFP fusion in spherical rodA-pbpA cells In E. coli, mutation of the mreB gene results in spherical cells [21,22]. Interestingly, residues 507–620 of IcsA, which are sufficient to localize a GFP fusion to the pole of rodshaped cells, localizes GFP to ectopic sites in spherical mreB cells [6,7]. In addition, the Vibrio cholerae type II secretion protein EpsM [7] and the E. coli chemotaxis protein Tar [6] are also localized at ectopic sites in the mreB mutant. Recent studies showed that MreB and its homologue Mbl govern spatial localization of proteins involved in peptidoglycan biosynthesis and influence the cellular morphology [14,15,23]. Penicillin-binding protein 2, encoded by pbpA, and RodA play important roles in peptidoglycan synthesis [24], and mutation of either the rodA or the pbpA gene results in round cells [25,26]. MreB localization is disrupted in RodA-depleted cells [27]. We sought to assess whether the distribution of the polar protein IcsA in rodA-pbpA spherical cells would be similar to its distribution in mreB cells. As has been reported for the mreB cells [7], IcsA507–620GFP appeared as fluorescent foci in rodA-pbpA cells. The number of foci per cell ranged from 1 to 4, increasing proportionally to the cell volume (Fig. 1A). The similar localization of the fluorescent foci to multiple sites in mreB and rodA-pbpA cells is consistent with the observation that mreB depletion leads to mislocalization of Pbp2, hence spherical cellular morphology [16]. Whereas cells of stationary phase cultures could have more foci, most cells of exponential phase cultures had only one spot per cell, probably because they have recently divided and have not yet grown. To avoid formation of inclusion bodies due to over-
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
A1
A2
B1
B2
495
Fig. 1. Localization of IcsA-GFP fusions in the spherical rodA-pbpA cells. Nomarski images (A1,B1) or fluorescence images (A2,B2) of strains rodApbpA/IcsA507–620-GFP (A1,A2) or rodA-pbpA/IcsAD507–730-GFP (B1,B2). Scale bar: 3 lm.
production of the fusion protein, only exponential state cultures with a controlled time of IcsA fusion expression were used in this study. In addition, the expression of the fusion was performed at 20 °C with low concentration of the inducer (see Materials and methods). In contrast to IcsA507–620-GFP, the IcsAD507–730-GFP fusion lacks the polar targeting sequence and gives a diffuse signal in both rod-shaped cells and spherical mreB cells [4,7]. As in the spherical mreB cells, the IcsAD507–730-GFP fusion displayed diffuse fluorescence in rodA-pbpA cells (Fig. 1B). As the expression conditions of the two fusion proteins were identical, the localization of IcsA507–620-GFP at ectopic sites in rodA-pbpA spherical cells depends on the polar targeting sequence and is unlikely to be due to formation of inclusion bodies. Recently, we studied production of Magnetospirillum sp. MamK actin protein in E. coli. The mamK gene alone is sufficient to direct the assembly of the MamK filament which is structurally and functionally distinct from the spiral MreB filament [13]. Interestingly, one extremity of the MamK actin-filament is co-localized with the IcsA507–620-mCherry foci in spherical mreB mutant [13]. In the spherical cells, the localization of the MamK-GFP filament extremities is not restricted by geometric parameters. The co-localization of MamK-GFP filament end with IcsA is thus meaningful, but the biological implication is elusive. Periplasmic space of spherical cells Previously, we have shown that folded GFP can be exported into the E. coli periplasm via the Tat system
[17]. The green fluorescence appears as a halo in most cells and occasionally shows a polar localization in wild type strains. In this study, we examined the distribution of periplasmic GFP in spherical cells. In most rodA-pbpA cells, the periplasmic GFP exhibited a halo appearance (Fig. 2A1 and A2). Formation of vesicles filled with GFP was occasionally observed in cells at exponential growth phase. The depletion of the cytoskeletal gene mreB also leads to spherical cells [22]. The mreB cells were bigger than those of the rodA-pbpA mutants. Otherwise they showed a similar morphology, fluorescence distribution, and vesicle formation property as those observed in the rodA-pbpA cells (Fig. 2A3 and A4). Interestingly, despite the overall halo appearance, the GFP fluorescent circles were not homogenous; bright areas were observed (Fig. 2). Deconvolution images confirmed that the GFP was discontinuously localized as zones in the periplasm of both rodA-pbpA and mreB spherical cells (Fig. 2B–G). In dividing cells, the septal area exhibited stronger green fluorescence than other places (see above and Fig. 2D and G). Therefore, the distribution of GFP indicated a zonal organization of the periplasmic space. Localization of the polar IcsA507–620-mCherry fusion with respect to the periplasmic GFP distribution Increasing data suggest that outer membrane proteins including LamB [28] and TonB-dependent receptors [15] localize in spiral, punctate or banding patterns similar to those adopted by MreC and Pbp2 in the periplasm. In addition, synthesis of the sidewall of Bascillus subtilis occurs in a helical pattern [23]. It has been proposed that
496
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
A1
A2
A4
A3
B1
B2
C1
C2
D1
D2
E1
E2
F1
F2
G1
G2
Fig. 2. Distribution of periplasmic GFP in spherical cells. (A1,A3) Nomarski images and (A2,A4) fluorescence images of rodA-pbpA/p8799 (A1,A2) or mreB/p8799 (A3,A4) strains. Arrows indicate vesicles with GFP inside; scale bar: 2 lm. For (B–G), the first image of each group is the clipping image and the second is the 3D deconvolution image of 33 Z-stacks of the same cells with step distance of 0.15 lm. (B–D) rodA-pbpA/p8799, (E–G) mreB/p8799. The red, green and blue lines indicate the X, Y, and Z directions, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
the subcellular localization pattern of these extracytoplasmic proteins is fastened in place by interaction with a helical structure present in the peptidoglycan layer [15]. Our observation of zonal GFP distribution of GFP in the periplasms of spherical cells is consistent with a highly organized periplasmic space. To gain insight into the relationship between the localization of polar information
and the periplasmic space organization, we analyzed the distribution of periplasmic GFP and the localization of the polar IcsA507–620-mCherry in the mreB and rodA-pbpA mutants. Since the GFP-filter excluded fluorescence from mCherry (data not shown), there was no bleed through under the conditions used here. By deconvolution images, periplasmic GFP was evident as green fluorescent zones.
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
A1
A2
B1
B2
C1
C2
D1
D2
E1
E2
F1
F2
G
H
I
J
K
L
M
N
O
497
Fig. 3. Distribution of IcsA507–620-mCherry and periplasmic GFP. (A–F) Images of rodA-pbpA/p8799+pAWY3 (A to C), and mreB/p8799+pAWY3 (D–F). The first image of each set shows clipping plan whereas the second one shows three-dimensional deconvolution reconstitution images from 33 zstacks. (G–I) Immuno-gold staining of ultra-thin frozen sections of transmission electron micrographs of MC1000/pMPR402 (G,H) and MC1000/ pAWY3 (I), respectively. (J–O) Optical sections of MC1000/p8799+pAWY3 cells. The cells shown in (M–O) were subjected to a mild hypertonic treatment.
498
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
Intriguingly, the red fluorescence of IcsA507–620-mCherry was not overlaid with the green GFP fluorescence. The same results were obtained for the two strains (Fig. 3A– F). Instead, the red fluorescence was located at the edge of the green fluorescence zones. When a single cell had two red fluorescent foci, they were separated by green zones (Fig. 3B and E). In the septating cells, the red fluorescence was not located at, but close to the common green fluorescence zone (Fig. 3C). As a control, the IcsAD507–730-GFP fusion lacking the polar targeting sequence exhibited a diffuse fluorescent signal in the spherical rodA-pbpA (Fig. 1B2) or mreB cells (data not shown). Therefore, the polar localization of IcsA507–620mCherry is unlikely to result from aggregation, self-association or the tendencies to occupy certain regions of the cell which gives artefactual results. This finding suggests that the polar material may be positioned as soon as the daughter cells are septated. The dark areas at the periphery of the spherical cells might reflect two possibilities. First, these areas might be inaccessible to the periplasmic space marker GFP. Second, these areas might possess particular biochemical properties that quench the GFP fluorescence. It has been reported by Bayer [29] that the inner membrane is closely associated with the murein-outer membrane layer at numerous sites within the cell envelope, forming zones of adhesion in the periplasm that can be seen in thin-section electron micrographs. The adhesion zones are assumed to be implicated in several cellular processes, including the export of proteins and lipopolysaccharides from the inner membrane to the outer membrane [30]. Therefore, the adhesion zones might be a physical barrier excluding the localization of the periplasmic GFP. The Bayer bridges have only been observed with rodshaped cells (for a review, see [31]), so we analyzed the localization of IcsA507–620-mCherry and IcsA507–620-GFP in rod-shaped cells by immuno-gold staining of ultra-thin frozen sections using transmission electron microscopy. The gold particles were concentrated at one pole of cells in an electron dense area (Fig. 3G–I), but no adhesion sites were evident under the conditions used. It appears to be very difficult to establish a direct relation between IcsA targeting sites and the Bayer bridges. To gain more information about a possible relationship between IcsA targeting sites and the periplasmic space, we compared the localization of IcsA507–620-mCherry with that of the periplasmic GFP in the rod-shaped cells. Among 170 cells inspected, 161 (94.7%) cells had the IcsA507–620-mCherry located at one pole of the cells (Fig. 3J–L) and in the 9 (5.3%) remaining cells, IcsA507–620-mCherry was at septa, which is consistent with the previous report of polar localization of IcsA [3,4]. Interestingly, about 41% of the IcsA507–620-mCherry polar foci were not centered (Fig. 3L). Since the periplasmic located GFP appeared as a thin halo, it is difficult to determine the periplasmic property at the position where IcsA507–620-mCherry is located. Application of a mild hypertonic treatment provokes plasmolysis, which could
be visualized by increasing of GFP signal at the poles [17]. In about 90% of the cells, the IcsA507–620-mCherry and the GFP were located at the same pole of the cell, whereas in about 10% of the cells, they were located at opposite poles of the cell (Fig. 3M). When the red IcsA507–620-mCherry and the periplasmic green GFP were located at the same pole, the two colors were adjacent, but barely overlaid (Fig. 3N and O). These observations are fully consistent with the results obtained with spherical cells, and confirm that IcsA507–620-mCherry was localized at the edge of the periplasmic GFP zones. The segregation of IcsA-mCherry and periplasmic GFP could be due to competition-exclusion for export between two protein species, or to a negative effect of areas where periplasm is expanded on the association of IcsA to the membranes. At present, the biochemical properties at these sites and the mechanism of specific targeting of polar proteins to the sites remain an open question. The dynamic structure of a periplasm creates a special problem for the transport of diverse molecules into and out of a bacterial cell. To avoid the diluting effect of the periplasm, all molecules destined to move into an external medium or into the cytoplasm must traverse the two membranes in a coordinated manner. Yet the dynamic structure of the periplasm should not be compromised. An ultimate solution to the coordination of processes spatially separated into two different membranes is the existence of intermembrane multiprotein complexes that span both membranes and create a bridge across the periplasm. Numerous transporters of Gram-negative bacteria involved in the extracellular secretion of protein and the efflux of toxic molecules have been shown to operate by forming intermembrane complexes (reviewed in [32]). The type 1 secretion system is composed of an ABC-transporter, providing energy through ATP hydrolysis (and perhaps the initial channel across the inner membrane), linked to a multimeric Membrane Fusion Protein (MFP) spanning the initial part of the periplasm and forming a continuous channel to the surface with an outer membrane trimeric protein. In E. coli the HlyA toxin interacts with both the MFP (HlyD) and the ABC protein HlyB, triggering, via a conformational change in HlyD, recruitment of the third component, TolC, into the transenvelope complex [33]. A similar AcrA–AcrB–TolC complex of E. coli is used to expel substrates directly into the medium [34]. These complexes temporally connect the inner membrane with cell wall, excluding the periplasmic space from these sites. Since they are dedicated for the export of specific substrates, these complexes are unlikely to be used for the membrane targeting of IcsA. Is there a similar complex specific for polar protein targeting and assembly? Although many proteins are specifically targeted to the poles, they do not contain a conserved common targeting sequence. It has been reported that covalent modification of Tar may stabilize its polar localization [35]. Therefore, conformational signal or sequence modification might contribute to the specific polar targeting.
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
In rod-shaped B. subtilis cells, new murein precursors are incorporated into the sidewall in spiral pattern governed by the MreB homologue, Mbl [23]. Increasing evidence converges to the concept that proteins involved in establishment and maintenance of rod-shaped morphology (e.g. MreB, MreC and Pbp2) adopt helical or banded patterns along the cell length in an inter-dependent manner [15,16,23]. The coordinated localization of these partners could determine the spatial material transport path from the cytoplasm to the cell wall. In rodA spherical cells, incorporation of new precursors is apparently a zonal process, localized at positions at which division had previously taken place [36]. The peptidoglycan precursors should be exported at specific sites under these zones. It would be interesting to determine whether there is a correlation between the polar protein targeting sites and the peptidoglycan synthesis zone. The next challenge would be the identification of the materials located at these sites and elucidation of the polar protein targeting mechanism. Acknowledgments We thank D. Vinella for the GC3904 strain, and J.E. Rothman for the plasmid pHluorin. This work was supported in part by LSHB-CT-2004-005257 and HFSPRGP0035/2004-C. References [1] A. Janakiraman, M.B. Goldberg, Recent advances on the development of bacterial poles, Trends Microbiol. 12 (2004) 518–525. [2] M.B. Goldberg, O. Barzu, C. Parsot, P.J. Sansonetti, Unipolar localization and ATPase activity of IcsA, a Shigella flexneri protein involved in intracellular movement, J. Bacteriol. 175 (1993) 2189– 2196. [3] L.D. Brandon, N. Goehring, A. Janakiraman, A.W. Yan, T. Wu, J. Beckwith, M.B. Goldberg, IcsA, a polarly localized autotransporter with an atypical signal peptide, uses the Sec apparatus for secretion, although the Sec apparatus is circumferentially distributed, Mol. Microbiol. 50 (2003) 45–60. [4] M. Charles, M. Perez, J.H. Kobil, M.B. Goldberg, Polar targeting of Shigella virulence factor IcsA in Enterobacteriacae and Vibrio, Proc. Natl. Acad. Sci. USA 98 (2001) 9871–9876. [5] M.T. Cabeen, C. Jacobs-Wagner, Bacterial cell shape, Nat. Rev. Microbiol. 3 (2005) 601–610. [6] Y.L. Shih, I. Kawagishi, L. Rothfield, The MreB and Min cytoskeletal-like systems play independent roles in prokaryotic polar differentiation, Mol. Microbiol. 58 (2005) 917–928. [7] T. Nilsen, A.W. Yan, G. Gale, M.B. Goldberg, Presence of multiple sites containing polar material in spherical Escherichia coli cells that lack MreB, J. Bacteriol. 187 (2005) 6187–6196. [8] A. Janakiraman, M.B. Goldberg, Evidence for polar positional information independent of cell division and nucleoid occlusion, Proc. Natl. Acad. Sci. USA 101 (2004) 835–840. [9] E. Huitema, S. Pritchard, D. Matteson, S.K. Radhakrishnan, P.H. Viollier, Bacterial birth scar proteins mark future flagellum assembly site, Cell 124 (2006) 1025–1037. [10] H. Lam, W.B. Schofield, C. Jacobs-Wagner, A landmark protein essential for establishing and perpetuating the polarity of a bacterial cell, Cell 124 (2006) 1011–1023.
499
[11] A. Scheffel, M. Gruska, D. Faivre, A. Linaroudis, J.M. Plitzko, D. Schuler, An acidic protein aligns magnetosomes along a filamentous structure in magnetotactic bacteria, Nature 440 (2006) 110– 114. [12] A. Komeili, Z. Li, D.K. Newman, G.J. Jensen, Magnetosomes are cell membrane invaginations organized by the actin-like protein MamK, Science 311 (2006) 242–245. [13] N. Pradel, C.-L. Santini, A. Bernadac, Y. Fukumori, L.-F. Wu, Biogenesis of actin-like bacterial cytoskeletal filaments destined for positioning prokaryotic magnetic organelles, Proc. Natl. Acad. Sci. USA 103 (2006) 17485–17489. [14] R.M. Figge, A.V. Divakaruni, J.W. Gober, MreB, the cell shapedetermining bacterial actin homologue, co-ordinates cell wall morphogenesis in Caulobacter crescentus, Mol. Microbiol. 51 (2004) 1321–1332. [15] A.V. Divakaruni, R.R. Loo, Y. Xie, J.A. Loo, J.W. Gober, The cellshape protein MreC interacts with extracytoplasmic proteins including cell wall assembly complexes in Caulobacter crescentus, Proc. Natl. Acad. Sci. USA 102 (2005) 18602–18607. [16] N.A. Dye, Z. Pincus, J.A. Theriot, L. Shapiro, Z. Gitai, Two independent spiral structures control cell shape in Caulobacter, Proc. Natl. Acad. Sci. USA 102 (2005) 18608–18613. [17] C.-L. Santini, A. Bernadac, M. Zhang, A. Chanal, B. Ize, C. Blanco, L.-F. Wu, Translocation of jellyfish green fluorescent protein via the Tat system of Escherichia coli and change of its periplasmic localization in response to osmotic up-shock, J. Biol. Chem. 276 (2001) 8159–8164. [18] G. Miesenbock, D.A. De Angelis, J.E. Rothman, Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins, Nature 394 (1998) 192–195. [19] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1972. [20] J. Anba, A. Bernadac, J.-M. Pages, C. Lazdunski, The periseptal annulus in Escherichia coli, Biol. Cell 50 (1984) 273–278. [21] M. Wachi, M. Doi, Y. Okada, M. Matsuhashi, New mre genes mreC and mreD, responsible for formation of the rod shape of Escherichia coli cells, J. Bacteriol. 171 (1989) 6511–6516. [22] M. Wachi, M. Doi, S. Tamaki, W. Park, S. Nakajima-Iijima, M. Matsuhashi, Mutant isolation and molecular cloning of mre genes, which determine cell shape, sensitivity to mecillinam, and amount of penicillin-binding proteins in Escherichia coli, J. Bacteriol. 169 (1987) 4935–4940. [23] R.A. Daniel, J. Errington, Control of cell morphogenesis in bacteria: two distinct ways to make a rod-shaped cell, Cell 113 (2003) 767– 776. [24] F. Ishino, W. Park, S. Tomioka, S. Tamaki, I. Takase, K. Kunugita, H. Matsuzawa, S. Asoh, T. Ohta, B.G. Spratt, et al., Peptidoglycan synthetic activities in membranes of Escherichia coli caused by overproduction of penicillin-binding protein 2 and rodA protein, J. Biol. Chem. 261 (1986) 7024–7031. [25] H. Matsuzawa, K. Hayakawa, T. Sato, K. Imahori, Characterization and genetic analysis of a mutant of Escherichia coli K-12 with rounded morphology, J. Bacteriol. 115 (1973) 436– 442. [26] B.G. Spratt, Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12, Proc. Natl. Acad. Sci. USA 72 (1975) 2999–3003. [27] T. Kruse, J. Bork-Jensen, K. Gerdes, The morphogenetic MreBCD proteins of Escherichia coli form an essential membrane-bound complex, Mol. Microbiol. 55 (2005) 78–89. [28] K.A. Gibbs, D.D. Isaac, J. Xu, R.W. Hendrix, T.J. Silhavy, J.A. Theriot, Complex spatial distribution dynamics of an abundant Escherichia coli outer membrane protein LamB, Mol. Microbiol. 53 (2004) 1771–1783. [29] M.E. Bayer, Areas of adhesion between wall and membrane of Escherichia coli, J. Gen. Microbiol. 53 (1968) 395–404. [30] M. Bayer, The fusion sites between outer membrane and cytoplasmic membrane of bacteria: their role in membrane assembly and virus
500
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
infection, in: M. Inouye (Ed.), Bacterial Outer Membrane, M. Wiley, New York, 1979, pp. 167–202. [31] C.L. Woldringh, Significance of plasmolysis spaces as markers for periseptal annuli and adhesion sites, Mol. Microbiol. 14 (1994) 597–607. [32] C. Andersen, C. Hughes, V. Koronakis, Protein export and drug efflux through bacterial channel-tunnels, Curr. Opin. Cell Biol. 13 (2001) 412–416. [33] I.B. Holland, L. Schmitt, J. Young, Type 1 protein secretion in bacteria, the ABC-transporter dependent pathway, Mol. Membr. Biol. 22 (2005) 29–39.
[34] S. Murakami, A. Yamaguchi, Multidrug-exporting secondary transporters, Curr. Opin. Struct. Biol. 13 (2003) 443–452. [35] D. Shiomi, S. Banno, M. Homma, I. Kawagishi, Stabilization of polar localization of a chemoreceptor via its covalent modifications and its communication with a different chemoreceptor, J. Bacteriol. 187 (2005) 7647–7654. [36] M.A. de Pedro, W.D. Donachie, J.V. Holtje, H. Schwarz, Constitutive septal murein synthesis in Escherichia coli with impaired activity of the morphogenetic proteins RodA and penicillin-binding protein 2, J. Bacteriol. 183 (2001) 4115–4126.
Polar positional information in Escherichia coli spherical cells Nathalie Pradel
a,1
, Claire-Lise Santini a,1, Alain Bernadac b, Yu-Ling Shih Marcia B. Goldberg d, Long-Fei Wu a,*
c,2
,
a
c
Laboratoire de Chimie Bacte´rienne, UPR9043, Institut de Biologie Structurale et Microbiologie, CNRS, 31 chemin Joseph Aiguier, F-13402 Marseille cedex 20, France b Service de Microscopie Electronique, Institut de Biologie Structurale et Microbiologie, CNRS, 31 chemin Joseph Aiguier, F-13402 Marseille cedex 20, France Department of Molecular, Microbial, and Structural Biology, University of Connecticut Health Center, Farmington, CT 06032, USA d Bacterial Pathogenesis Laboratories, Infectious Disease Division, Massachusetts General Hospital, Cambridge, MA 02139, USA Received 5 December 2006 Available online 18 December 2006
Abstract Shigella surface protein IcsA and its cytoplasmic derivatives are localized to the old pole of rod-shaped cells when expressed in Escherichia coli. In spherical mreB cells, IcsA is targeted to ectopic sites and close to one extremity of actin-like MamK filament. To gain insight into the properties of the sites containing polar material, we studied the IcsA localization in spherical cells. GFP was exported into the periplasm via the Tat pathway and used as a periplasmic space marker. GFP displayed zonal fluorescence in both mreB and rodA-pbpA spherical E. coli cells, indicating an uneven periplasmic space. Deconvolution images revealed that the cytoplasmic IcsA fused to mCherry was localized outside or at the edge of the GFP zones. These observations strongly suggest that polar material is restricted to the positions where the periplasm possesses particular structural or biochemical properties. Ó 2006 Elsevier Inc. All rights reserved. Keywords: Periplasm; Polar localization information; Spherical cells; MreB; GFP; mCherry
Polarity is an intrinsic property of cells. Many different bacterial species utilize polar localization for diverse cellular processes including chromosome segregation, protein secretion, macromolecule delivery, chemotaxis, adherence, motility, cell division, cell shape and virulence [1]. The Shigella outer membrane protein IcsA (VirG) is one of the most extensively studied polar proteins. It localizes to the pole of Shigella, where it mediates assembly of the actin tail and movement of this intracellular pathogen within the cytoplasm of a host cell [2]. IcsA is targeted and exported at the bacterial pole [3]. Each of two segments, residues 1–104 (IcsA1–104) or 507–620 (IcsA507–620), is sufficient for *
Corresponding author. Fax: +33 491718914. E-mail address: [email protected] (L.-F. Wu). 1 These authors contributed equally to this work. 2 Present address: Institute of Biological Chemistry, Academia Sinica, 128 Sec. 2, Academia Road, Nankang, Taipei 115, Taiwan. 0006-291X/$ - see front matter Ó 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2006.12.054
targeting IcsA to the pole on the cytoplasmic side of the inner membrane of rod-shaped cells [3,4]. The bacterial cytoskeletal protein MreB forms helical structures along the long axis of the cell, just beneath the cytoplasmic membrane (for review see [5]). Interestingly, the mreB mutation leads to ectopic localization of IcsA and its truncated IcsA1–104 and IcsA507–620 derivatives in the spherical cells [6,7]. Janakiraman and Goldberg have shown that polar positional information required for IcsA localization is independent of both known components of the cell division machinery and nucleoid occlusion [8]. Recently, two laboratories have independently identified a landmark protein, TipN, which acts as a spatial and temporal cue for setting up the correct polarity in the bacterium Caulobacter crescentus [9,10]. In contrast to IcsA, TipN does not localize to the division site in cells depleted of the division initiation protein FtsZ [9,10] or in cells in which the division site peptidoglycan synthesis protein FtsI has been inactivated [9].
494
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
These observations would suggest that different mechanisms might be used for polar localization of IcsA and TipN. However, the polar positional information that mediates proper localization of the polar proteins and the mechanisms by which polar proteins recognize this positional information are poorly understood [1]. MamK actin proteins are found only in magnetotactic bacteria and involved in positioning bacterial organelles, magnetosomes [11,12]. When expressed in Escherichia coli, MamK appears as a linear filament which is structurally and functionally distinct from the MreB filament [13]. Interestingly, one extremity of the MamK filament is located close to the co-expressed IcsA in spherical mreB mutant cells [13], suggesting that the filament may not be randomly positioned within a geometrically symmetric spherical cell. In E. coli, like the mreB mutation, rodA-pbpA mutations affect periplasmic sidewall peptidoglycan synthesis and lead to spherical cells [5]. The cellular localization of penicillinbinding protein 2 (encoded by pbpA) depends upon the localization of both MreB and MreC in C. crescentus [14–16]. To gain more information regarding the polar material localization information in spherical cells, we compared the localization of IcsA in mreB and rodA-pbpA spherical cells. Green fluorescence protein (GFP) was exported into the periplasm via the twin-arginine translocation pathway and used as a periplasmic space marker under physiological conditions. Periplasmic GFP exhibited zonal localization in both mutants. Interestingly, IcsA507–620mCherry was located outside or at the edge of the GFP zones, but not overlaid on the GFP zones. These results would suggest that polar positional information recognized by IcsA resides at positions where the periplasm possesses particular structural or biochemical properties.
Materials and methods Bacterial strains, plasmids, and media. Escherichia coli strains used in this study are: MC1000 (araD139 (ara, leu)7697 lacX74 galU galK rpsL) and its derivative YLS3 (mreB) [6], GC3904 (argS D(pbpA-rodA)::Kan, zbf::Tn10 [17] and TG1 (D(lac-pro) supE thi hsdD5/F 0 traD36 proA+B+ lacIq lacZDM15). Plasmids pBAD-IcsA1–104-GFP, pBAD-IcsA507–620-GFP (=pMPR402), pBAD-IcsAD507–730-GFP and Ptac-IcsA507–620-mCherry (pAWY3) express hybrid proteins consisting of various segments of IcsA fused in frame to green fluorescent protein or mCherry under the control of the arabinose promoter of pBAD24 or the tac promoter [7]. To improve signal/noise ratio of periplasmic GFP fluorescence, the gfpmut2 in our previously described pRR-GFP plasmid [17] was changed with the gene encoding pH sensor derivative of GFP, the ratiometric pHluorin [18]. Compared to the wild type GFP, pHluorin displays a reversible excitation ratio change between pH 7.5 and 5.5; the major excitation peak is at 475 nm under acidic conditions [18]. Since the periplasm is more acidic than the cytoplasm due to the proton motive force, ratiometric pHluorin increases the signal/background ratio comparing to other green fluorescent protein derivatives when the same portion of the marker proteins are exported into the periplasm. The DNA fragment encoding for ratiometric pHluorin was amplified by PCR using PHR5NHE (5 0 -cgg tgc tag caa agg aga aga act ttt cac-3 0 ) and PHR3H3 (5 0 -gaa tta agc tta ttt gta tag ttc atc cat gcc-3 0 ) as primers, and the reaction was performed by using the Expand High Fidelity PCR System according to the manufacturer’s instructions (Roche). The amplified fragment was
purified, double digested by NheI and HindIII and cloned into the corresponding sites of the plasmid pRR-GFP, resulting to plasmid p8799. The bacteria were routinely grown in Luria–Bertani (LB) medium or on LB plates [19]. As required, ampicillin (Amp) (100 lg/ml), chloramphenicol (30 lg/ml), glucose (0.2% w/v), arabinose (0.2% w/v) or IPTG (0.5 mM) were added. Microscopy and fluorescence spectrometry. Overnight cultures were diluted 1:100 for wild type strain and 1:50 for the mutants in LB+Amp+glucose medium and incubated at 37 °C with shaking for 3 h. Cells were centrifuged, washed once with LB+Amp medium, and resuspended in LB+Amp with 0.2% arabinose or/and IPTG (0.5 mM), and grown at 20 °C for 80 min with shaking. The cells were examined directly, or after the addition of 0.15 M NaCl or other treatment as described in the text. Protein synthesis was blocked by the addition of rifampicin (0.15 lg/ml) to the culture. Samples were taken 1 h later and examined under fluorescence microscope. Cells were fixed in 0.25% agarose on slides. Images and Z-stack of 25–33 images were captured with step distance ranging from 0.15 or 0.25 lm with Zeiss Axiovert 200M connected with a Hamamatsu ORCA ER camera. Image restoration was obtained by deconvolution using Huygens Essential software (SVI). Three-dimensional visualization was performed with Imaris software package (Bitplane). Immuno-gold staining of ultra-thin frozen sections were performed by using 7 nm gold-conjugated protein A as described by Anba et al. [20]. Antisera used are polyclonal rabbit anti GFP [17] or RFP (Chemicon International, Temecula, California, USA) at dilution of 1/500 and 1/100, respectively.
Results and discussion Localization of polar IcsA507–620-GFP fusion in spherical rodA-pbpA cells In E. coli, mutation of the mreB gene results in spherical cells [21,22]. Interestingly, residues 507–620 of IcsA, which are sufficient to localize a GFP fusion to the pole of rodshaped cells, localizes GFP to ectopic sites in spherical mreB cells [6,7]. In addition, the Vibrio cholerae type II secretion protein EpsM [7] and the E. coli chemotaxis protein Tar [6] are also localized at ectopic sites in the mreB mutant. Recent studies showed that MreB and its homologue Mbl govern spatial localization of proteins involved in peptidoglycan biosynthesis and influence the cellular morphology [14,15,23]. Penicillin-binding protein 2, encoded by pbpA, and RodA play important roles in peptidoglycan synthesis [24], and mutation of either the rodA or the pbpA gene results in round cells [25,26]. MreB localization is disrupted in RodA-depleted cells [27]. We sought to assess whether the distribution of the polar protein IcsA in rodA-pbpA spherical cells would be similar to its distribution in mreB cells. As has been reported for the mreB cells [7], IcsA507–620GFP appeared as fluorescent foci in rodA-pbpA cells. The number of foci per cell ranged from 1 to 4, increasing proportionally to the cell volume (Fig. 1A). The similar localization of the fluorescent foci to multiple sites in mreB and rodA-pbpA cells is consistent with the observation that mreB depletion leads to mislocalization of Pbp2, hence spherical cellular morphology [16]. Whereas cells of stationary phase cultures could have more foci, most cells of exponential phase cultures had only one spot per cell, probably because they have recently divided and have not yet grown. To avoid formation of inclusion bodies due to over-
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
A1
A2
B1
B2
495
Fig. 1. Localization of IcsA-GFP fusions in the spherical rodA-pbpA cells. Nomarski images (A1,B1) or fluorescence images (A2,B2) of strains rodApbpA/IcsA507–620-GFP (A1,A2) or rodA-pbpA/IcsAD507–730-GFP (B1,B2). Scale bar: 3 lm.
production of the fusion protein, only exponential state cultures with a controlled time of IcsA fusion expression were used in this study. In addition, the expression of the fusion was performed at 20 °C with low concentration of the inducer (see Materials and methods). In contrast to IcsA507–620-GFP, the IcsAD507–730-GFP fusion lacks the polar targeting sequence and gives a diffuse signal in both rod-shaped cells and spherical mreB cells [4,7]. As in the spherical mreB cells, the IcsAD507–730-GFP fusion displayed diffuse fluorescence in rodA-pbpA cells (Fig. 1B). As the expression conditions of the two fusion proteins were identical, the localization of IcsA507–620-GFP at ectopic sites in rodA-pbpA spherical cells depends on the polar targeting sequence and is unlikely to be due to formation of inclusion bodies. Recently, we studied production of Magnetospirillum sp. MamK actin protein in E. coli. The mamK gene alone is sufficient to direct the assembly of the MamK filament which is structurally and functionally distinct from the spiral MreB filament [13]. Interestingly, one extremity of the MamK actin-filament is co-localized with the IcsA507–620-mCherry foci in spherical mreB mutant [13]. In the spherical cells, the localization of the MamK-GFP filament extremities is not restricted by geometric parameters. The co-localization of MamK-GFP filament end with IcsA is thus meaningful, but the biological implication is elusive. Periplasmic space of spherical cells Previously, we have shown that folded GFP can be exported into the E. coli periplasm via the Tat system
[17]. The green fluorescence appears as a halo in most cells and occasionally shows a polar localization in wild type strains. In this study, we examined the distribution of periplasmic GFP in spherical cells. In most rodA-pbpA cells, the periplasmic GFP exhibited a halo appearance (Fig. 2A1 and A2). Formation of vesicles filled with GFP was occasionally observed in cells at exponential growth phase. The depletion of the cytoskeletal gene mreB also leads to spherical cells [22]. The mreB cells were bigger than those of the rodA-pbpA mutants. Otherwise they showed a similar morphology, fluorescence distribution, and vesicle formation property as those observed in the rodA-pbpA cells (Fig. 2A3 and A4). Interestingly, despite the overall halo appearance, the GFP fluorescent circles were not homogenous; bright areas were observed (Fig. 2). Deconvolution images confirmed that the GFP was discontinuously localized as zones in the periplasm of both rodA-pbpA and mreB spherical cells (Fig. 2B–G). In dividing cells, the septal area exhibited stronger green fluorescence than other places (see above and Fig. 2D and G). Therefore, the distribution of GFP indicated a zonal organization of the periplasmic space. Localization of the polar IcsA507–620-mCherry fusion with respect to the periplasmic GFP distribution Increasing data suggest that outer membrane proteins including LamB [28] and TonB-dependent receptors [15] localize in spiral, punctate or banding patterns similar to those adopted by MreC and Pbp2 in the periplasm. In addition, synthesis of the sidewall of Bascillus subtilis occurs in a helical pattern [23]. It has been proposed that
496
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
A1
A2
A4
A3
B1
B2
C1
C2
D1
D2
E1
E2
F1
F2
G1
G2
Fig. 2. Distribution of periplasmic GFP in spherical cells. (A1,A3) Nomarski images and (A2,A4) fluorescence images of rodA-pbpA/p8799 (A1,A2) or mreB/p8799 (A3,A4) strains. Arrows indicate vesicles with GFP inside; scale bar: 2 lm. For (B–G), the first image of each group is the clipping image and the second is the 3D deconvolution image of 33 Z-stacks of the same cells with step distance of 0.15 lm. (B–D) rodA-pbpA/p8799, (E–G) mreB/p8799. The red, green and blue lines indicate the X, Y, and Z directions, respectively. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this paper.)
the subcellular localization pattern of these extracytoplasmic proteins is fastened in place by interaction with a helical structure present in the peptidoglycan layer [15]. Our observation of zonal GFP distribution of GFP in the periplasms of spherical cells is consistent with a highly organized periplasmic space. To gain insight into the relationship between the localization of polar information
and the periplasmic space organization, we analyzed the distribution of periplasmic GFP and the localization of the polar IcsA507–620-mCherry in the mreB and rodA-pbpA mutants. Since the GFP-filter excluded fluorescence from mCherry (data not shown), there was no bleed through under the conditions used here. By deconvolution images, periplasmic GFP was evident as green fluorescent zones.
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
A1
A2
B1
B2
C1
C2
D1
D2
E1
E2
F1
F2
G
H
I
J
K
L
M
N
O
497
Fig. 3. Distribution of IcsA507–620-mCherry and periplasmic GFP. (A–F) Images of rodA-pbpA/p8799+pAWY3 (A to C), and mreB/p8799+pAWY3 (D–F). The first image of each set shows clipping plan whereas the second one shows three-dimensional deconvolution reconstitution images from 33 zstacks. (G–I) Immuno-gold staining of ultra-thin frozen sections of transmission electron micrographs of MC1000/pMPR402 (G,H) and MC1000/ pAWY3 (I), respectively. (J–O) Optical sections of MC1000/p8799+pAWY3 cells. The cells shown in (M–O) were subjected to a mild hypertonic treatment.
498
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
Intriguingly, the red fluorescence of IcsA507–620-mCherry was not overlaid with the green GFP fluorescence. The same results were obtained for the two strains (Fig. 3A– F). Instead, the red fluorescence was located at the edge of the green fluorescence zones. When a single cell had two red fluorescent foci, they were separated by green zones (Fig. 3B and E). In the septating cells, the red fluorescence was not located at, but close to the common green fluorescence zone (Fig. 3C). As a control, the IcsAD507–730-GFP fusion lacking the polar targeting sequence exhibited a diffuse fluorescent signal in the spherical rodA-pbpA (Fig. 1B2) or mreB cells (data not shown). Therefore, the polar localization of IcsA507–620mCherry is unlikely to result from aggregation, self-association or the tendencies to occupy certain regions of the cell which gives artefactual results. This finding suggests that the polar material may be positioned as soon as the daughter cells are septated. The dark areas at the periphery of the spherical cells might reflect two possibilities. First, these areas might be inaccessible to the periplasmic space marker GFP. Second, these areas might possess particular biochemical properties that quench the GFP fluorescence. It has been reported by Bayer [29] that the inner membrane is closely associated with the murein-outer membrane layer at numerous sites within the cell envelope, forming zones of adhesion in the periplasm that can be seen in thin-section electron micrographs. The adhesion zones are assumed to be implicated in several cellular processes, including the export of proteins and lipopolysaccharides from the inner membrane to the outer membrane [30]. Therefore, the adhesion zones might be a physical barrier excluding the localization of the periplasmic GFP. The Bayer bridges have only been observed with rodshaped cells (for a review, see [31]), so we analyzed the localization of IcsA507–620-mCherry and IcsA507–620-GFP in rod-shaped cells by immuno-gold staining of ultra-thin frozen sections using transmission electron microscopy. The gold particles were concentrated at one pole of cells in an electron dense area (Fig. 3G–I), but no adhesion sites were evident under the conditions used. It appears to be very difficult to establish a direct relation between IcsA targeting sites and the Bayer bridges. To gain more information about a possible relationship between IcsA targeting sites and the periplasmic space, we compared the localization of IcsA507–620-mCherry with that of the periplasmic GFP in the rod-shaped cells. Among 170 cells inspected, 161 (94.7%) cells had the IcsA507–620-mCherry located at one pole of the cells (Fig. 3J–L) and in the 9 (5.3%) remaining cells, IcsA507–620-mCherry was at septa, which is consistent with the previous report of polar localization of IcsA [3,4]. Interestingly, about 41% of the IcsA507–620-mCherry polar foci were not centered (Fig. 3L). Since the periplasmic located GFP appeared as a thin halo, it is difficult to determine the periplasmic property at the position where IcsA507–620-mCherry is located. Application of a mild hypertonic treatment provokes plasmolysis, which could
be visualized by increasing of GFP signal at the poles [17]. In about 90% of the cells, the IcsA507–620-mCherry and the GFP were located at the same pole of the cell, whereas in about 10% of the cells, they were located at opposite poles of the cell (Fig. 3M). When the red IcsA507–620-mCherry and the periplasmic green GFP were located at the same pole, the two colors were adjacent, but barely overlaid (Fig. 3N and O). These observations are fully consistent with the results obtained with spherical cells, and confirm that IcsA507–620-mCherry was localized at the edge of the periplasmic GFP zones. The segregation of IcsA-mCherry and periplasmic GFP could be due to competition-exclusion for export between two protein species, or to a negative effect of areas where periplasm is expanded on the association of IcsA to the membranes. At present, the biochemical properties at these sites and the mechanism of specific targeting of polar proteins to the sites remain an open question. The dynamic structure of a periplasm creates a special problem for the transport of diverse molecules into and out of a bacterial cell. To avoid the diluting effect of the periplasm, all molecules destined to move into an external medium or into the cytoplasm must traverse the two membranes in a coordinated manner. Yet the dynamic structure of the periplasm should not be compromised. An ultimate solution to the coordination of processes spatially separated into two different membranes is the existence of intermembrane multiprotein complexes that span both membranes and create a bridge across the periplasm. Numerous transporters of Gram-negative bacteria involved in the extracellular secretion of protein and the efflux of toxic molecules have been shown to operate by forming intermembrane complexes (reviewed in [32]). The type 1 secretion system is composed of an ABC-transporter, providing energy through ATP hydrolysis (and perhaps the initial channel across the inner membrane), linked to a multimeric Membrane Fusion Protein (MFP) spanning the initial part of the periplasm and forming a continuous channel to the surface with an outer membrane trimeric protein. In E. coli the HlyA toxin interacts with both the MFP (HlyD) and the ABC protein HlyB, triggering, via a conformational change in HlyD, recruitment of the third component, TolC, into the transenvelope complex [33]. A similar AcrA–AcrB–TolC complex of E. coli is used to expel substrates directly into the medium [34]. These complexes temporally connect the inner membrane with cell wall, excluding the periplasmic space from these sites. Since they are dedicated for the export of specific substrates, these complexes are unlikely to be used for the membrane targeting of IcsA. Is there a similar complex specific for polar protein targeting and assembly? Although many proteins are specifically targeted to the poles, they do not contain a conserved common targeting sequence. It has been reported that covalent modification of Tar may stabilize its polar localization [35]. Therefore, conformational signal or sequence modification might contribute to the specific polar targeting.
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
In rod-shaped B. subtilis cells, new murein precursors are incorporated into the sidewall in spiral pattern governed by the MreB homologue, Mbl [23]. Increasing evidence converges to the concept that proteins involved in establishment and maintenance of rod-shaped morphology (e.g. MreB, MreC and Pbp2) adopt helical or banded patterns along the cell length in an inter-dependent manner [15,16,23]. The coordinated localization of these partners could determine the spatial material transport path from the cytoplasm to the cell wall. In rodA spherical cells, incorporation of new precursors is apparently a zonal process, localized at positions at which division had previously taken place [36]. The peptidoglycan precursors should be exported at specific sites under these zones. It would be interesting to determine whether there is a correlation between the polar protein targeting sites and the peptidoglycan synthesis zone. The next challenge would be the identification of the materials located at these sites and elucidation of the polar protein targeting mechanism. Acknowledgments We thank D. Vinella for the GC3904 strain, and J.E. Rothman for the plasmid pHluorin. This work was supported in part by LSHB-CT-2004-005257 and HFSPRGP0035/2004-C. References [1] A. Janakiraman, M.B. Goldberg, Recent advances on the development of bacterial poles, Trends Microbiol. 12 (2004) 518–525. [2] M.B. Goldberg, O. Barzu, C. Parsot, P.J. Sansonetti, Unipolar localization and ATPase activity of IcsA, a Shigella flexneri protein involved in intracellular movement, J. Bacteriol. 175 (1993) 2189– 2196. [3] L.D. Brandon, N. Goehring, A. Janakiraman, A.W. Yan, T. Wu, J. Beckwith, M.B. Goldberg, IcsA, a polarly localized autotransporter with an atypical signal peptide, uses the Sec apparatus for secretion, although the Sec apparatus is circumferentially distributed, Mol. Microbiol. 50 (2003) 45–60. [4] M. Charles, M. Perez, J.H. Kobil, M.B. Goldberg, Polar targeting of Shigella virulence factor IcsA in Enterobacteriacae and Vibrio, Proc. Natl. Acad. Sci. USA 98 (2001) 9871–9876. [5] M.T. Cabeen, C. Jacobs-Wagner, Bacterial cell shape, Nat. Rev. Microbiol. 3 (2005) 601–610. [6] Y.L. Shih, I. Kawagishi, L. Rothfield, The MreB and Min cytoskeletal-like systems play independent roles in prokaryotic polar differentiation, Mol. Microbiol. 58 (2005) 917–928. [7] T. Nilsen, A.W. Yan, G. Gale, M.B. Goldberg, Presence of multiple sites containing polar material in spherical Escherichia coli cells that lack MreB, J. Bacteriol. 187 (2005) 6187–6196. [8] A. Janakiraman, M.B. Goldberg, Evidence for polar positional information independent of cell division and nucleoid occlusion, Proc. Natl. Acad. Sci. USA 101 (2004) 835–840. [9] E. Huitema, S. Pritchard, D. Matteson, S.K. Radhakrishnan, P.H. Viollier, Bacterial birth scar proteins mark future flagellum assembly site, Cell 124 (2006) 1025–1037. [10] H. Lam, W.B. Schofield, C. Jacobs-Wagner, A landmark protein essential for establishing and perpetuating the polarity of a bacterial cell, Cell 124 (2006) 1011–1023.
499
[11] A. Scheffel, M. Gruska, D. Faivre, A. Linaroudis, J.M. Plitzko, D. Schuler, An acidic protein aligns magnetosomes along a filamentous structure in magnetotactic bacteria, Nature 440 (2006) 110– 114. [12] A. Komeili, Z. Li, D.K. Newman, G.J. Jensen, Magnetosomes are cell membrane invaginations organized by the actin-like protein MamK, Science 311 (2006) 242–245. [13] N. Pradel, C.-L. Santini, A. Bernadac, Y. Fukumori, L.-F. Wu, Biogenesis of actin-like bacterial cytoskeletal filaments destined for positioning prokaryotic magnetic organelles, Proc. Natl. Acad. Sci. USA 103 (2006) 17485–17489. [14] R.M. Figge, A.V. Divakaruni, J.W. Gober, MreB, the cell shapedetermining bacterial actin homologue, co-ordinates cell wall morphogenesis in Caulobacter crescentus, Mol. Microbiol. 51 (2004) 1321–1332. [15] A.V. Divakaruni, R.R. Loo, Y. Xie, J.A. Loo, J.W. Gober, The cellshape protein MreC interacts with extracytoplasmic proteins including cell wall assembly complexes in Caulobacter crescentus, Proc. Natl. Acad. Sci. USA 102 (2005) 18602–18607. [16] N.A. Dye, Z. Pincus, J.A. Theriot, L. Shapiro, Z. Gitai, Two independent spiral structures control cell shape in Caulobacter, Proc. Natl. Acad. Sci. USA 102 (2005) 18608–18613. [17] C.-L. Santini, A. Bernadac, M. Zhang, A. Chanal, B. Ize, C. Blanco, L.-F. Wu, Translocation of jellyfish green fluorescent protein via the Tat system of Escherichia coli and change of its periplasmic localization in response to osmotic up-shock, J. Biol. Chem. 276 (2001) 8159–8164. [18] G. Miesenbock, D.A. De Angelis, J.E. Rothman, Visualizing secretion and synaptic transmission with pH-sensitive green fluorescent proteins, Nature 394 (1998) 192–195. [19] J.H. Miller, Experiments in Molecular Genetics, Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, 1972. [20] J. Anba, A. Bernadac, J.-M. Pages, C. Lazdunski, The periseptal annulus in Escherichia coli, Biol. Cell 50 (1984) 273–278. [21] M. Wachi, M. Doi, Y. Okada, M. Matsuhashi, New mre genes mreC and mreD, responsible for formation of the rod shape of Escherichia coli cells, J. Bacteriol. 171 (1989) 6511–6516. [22] M. Wachi, M. Doi, S. Tamaki, W. Park, S. Nakajima-Iijima, M. Matsuhashi, Mutant isolation and molecular cloning of mre genes, which determine cell shape, sensitivity to mecillinam, and amount of penicillin-binding proteins in Escherichia coli, J. Bacteriol. 169 (1987) 4935–4940. [23] R.A. Daniel, J. Errington, Control of cell morphogenesis in bacteria: two distinct ways to make a rod-shaped cell, Cell 113 (2003) 767– 776. [24] F. Ishino, W. Park, S. Tomioka, S. Tamaki, I. Takase, K. Kunugita, H. Matsuzawa, S. Asoh, T. Ohta, B.G. Spratt, et al., Peptidoglycan synthetic activities in membranes of Escherichia coli caused by overproduction of penicillin-binding protein 2 and rodA protein, J. Biol. Chem. 261 (1986) 7024–7031. [25] H. Matsuzawa, K. Hayakawa, T. Sato, K. Imahori, Characterization and genetic analysis of a mutant of Escherichia coli K-12 with rounded morphology, J. Bacteriol. 115 (1973) 436– 442. [26] B.G. Spratt, Distinct penicillin binding proteins involved in the division, elongation, and shape of Escherichia coli K12, Proc. Natl. Acad. Sci. USA 72 (1975) 2999–3003. [27] T. Kruse, J. Bork-Jensen, K. Gerdes, The morphogenetic MreBCD proteins of Escherichia coli form an essential membrane-bound complex, Mol. Microbiol. 55 (2005) 78–89. [28] K.A. Gibbs, D.D. Isaac, J. Xu, R.W. Hendrix, T.J. Silhavy, J.A. Theriot, Complex spatial distribution dynamics of an abundant Escherichia coli outer membrane protein LamB, Mol. Microbiol. 53 (2004) 1771–1783. [29] M.E. Bayer, Areas of adhesion between wall and membrane of Escherichia coli, J. Gen. Microbiol. 53 (1968) 395–404. [30] M. Bayer, The fusion sites between outer membrane and cytoplasmic membrane of bacteria: their role in membrane assembly and virus
500
N. Pradel et al. / Biochemical and Biophysical Research Communications 353 (2007) 493–500
infection, in: M. Inouye (Ed.), Bacterial Outer Membrane, M. Wiley, New York, 1979, pp. 167–202. [31] C.L. Woldringh, Significance of plasmolysis spaces as markers for periseptal annuli and adhesion sites, Mol. Microbiol. 14 (1994) 597–607. [32] C. Andersen, C. Hughes, V. Koronakis, Protein export and drug efflux through bacterial channel-tunnels, Curr. Opin. Cell Biol. 13 (2001) 412–416. [33] I.B. Holland, L. Schmitt, J. Young, Type 1 protein secretion in bacteria, the ABC-transporter dependent pathway, Mol. Membr. Biol. 22 (2005) 29–39.
[34] S. Murakami, A. Yamaguchi, Multidrug-exporting secondary transporters, Curr. Opin. Struct. Biol. 13 (2003) 443–452. [35] D. Shiomi, S. Banno, M. Homma, I. Kawagishi, Stabilization of polar localization of a chemoreceptor via its covalent modifications and its communication with a different chemoreceptor, J. Bacteriol. 187 (2005) 7647–7654. [36] M.A. de Pedro, W.D. Donachie, J.V. Holtje, H. Schwarz, Constitutive septal murein synthesis in Escherichia coli with impaired activity of the morphogenetic proteins RodA and penicillin-binding protein 2, J. Bacteriol. 183 (2001) 4115–4126.